Tobacco Documents Online

TIMN 428549

Smoking Behavior and Policy Conference Series The Pharmacologic Treatment of Tobacco Dependence: Proceedings of the World Congress Nvvember 4-5, 1985 Ir>>titutc tor thr Stud-v uf Smokim; Bchavior aud Po(ic~, Harv,.mi L »ircrsity (ohn F. Kcnilcdy School ot Govc•rnntcilt %() John F. K(2nncd- - Str«t Cambridgr. Massachusctts 021 -18 TIMN 428550

J. K. Ockene, Editor © Copyright 1986 President and Fellows of Harvard College All rights reserved Printed in the United States of America Additional copies of these Proceedings are available at a cost of Ten Dollars ($10.00 U.S.) per single copy, or Seven Dollars and Fifty Cents ($7.50 U.S.) per copy for 5 or more copies mailed together to the same address within the United States. Books are shipped via U.S. Postal Service at book rate. Irrstitute for the Study of Smoking Behavior and Policy John F. Kennedy School of Government Harvard University 79 John F. Kennedy Street Cambridge, MA 02138 For citations: Ockene, JK (ed.). The Pharmacologic Trentment of Tobacco Dependence: Prvreed iirgs of the Wrld ConAress, Novembcr 4-5, 1985. Cambridge, Massachusetts: Institute for the Study of Smoking Behavior and Policy, 1986. Design and Printing: Puritan Press, Inc.., Nashua, New Hampshire Foreword The conference from which these proceedings are drawn is part of an on- going series begun by the Institute for the Study of Smoking Behavior and Policy in April, 1985. Through this conference series, we hope to focus the attention of researchers and policymakers on those issues shaping our collective approach to the problems of cigarette smoking in Ameri- can society. Early in the life of the Institute, we determined that there was an emerg- ing consensus that tobacco smoking is an addiction and as such, the policy implications were worthy of careful attention. In order to explore the is- sues further, we joined with the National lnstitute on Drug Abuse in July, 1985 to sponsor a working meeting that reviewed the evidence on tobacco addiction and key issues such as the implications for treatment. It was apparent from that meeting that despite some remaining dis- agreements about precise terminology of addiction and dependence, the field of smoking research and treatment had moved far beyond purely theoretical considerations of pharmacologic treatments to grappling with the implications of the first commercial product, nicotine gum, and the possibility of others to follow. It was also apparent that all those knowledgeable and concerned about issues of the use, abuse and efficacy of pharmacologic approaches could benefit from a structured interchange and exploration of research find- ings and issues. This led the Institute to welcome the opportunity offered by Merrell Dow Pharmaceuticals to cosponsor a World Congress on the Pharmacologic Treatment of Tobacco Dependence. This volume represents the substance of that conference. This collection of presentations by scientists from many countries and many disciplines reflects the remarkable growth that has taken place in our knowledge of smoking behavior in the short span of six years. In 1979, the Surgeon General's Report endeavored to present the most com- prehensive review of the literature on smoking behavior ever produced. Of the over 1400 pages in that report, fewer than 125 were devoted to discussing the psychosocial and psychopharmacologic aspects of smoking behavior. Less than one page was needed to review the knowledge on specific pharmacologic treatments. In his opening remarks, Dr. Jerome Jaffe sums up one of the primary reasons for this progress: "Collectively the scientific enterl3rise-the phar- maceutical industry and the academic community-has `shipped'. It has delivered a product and demonstrated that in the proper context the product can have a significant impact on smoking behavior."

In this case, the product is nicotine gum. But as the presentations that follow show, nicotine gum is likely to prove to be the forerunner of other products. The progress that has been made in understanding smoking behavior and the fact that over 50 million Americans and over a billion people worldwide continue to smoke, ensure a high probability that other pharmacologic approaches will be developed, tested and marketed. The "collective" venture that brought nicotine gum is significant for many reasons, not the least of which is its enormous contribution to our understanding of the behavioral and pharmacological components of tobacco dependence. It has also given physicians a means to become more actively and effectively involved in smoking cessation - an involvement that may lead to increased cessation rates and increasing interest in new treatments. In general, the acceptance of this new approach to treatment by those who have labored long to help smokers quit has been marked by thought- fiil, gradual acceptance based on demonstrated efficacy. As a field, our experience with this first product should guide our response when other treatment approaches inevitably arrive on the scene. It is the Institute's hope that these Proceedings will not only educate and inform about the state of today's knowledge, but that they will also help guide our actions, attitudes and policies toward future developments in the pharmacologic treatment of tobacco dependence. Thomas C. Schelling, Director John M. Pinney, Executive Director Institute for the Study of Smoking Behavior and Policy John F. Kennedy School of Government Harvard University Preface In response to mounting evidence of the hazards of cigarette smoking, a growing demand for effective strategies to decrease the prevalence of smoking has culrninated in Surgeon General C. Everett Koop's challenge for a smoke-free society by the year 2000. Over the last decade, research has contributed greatly to our understanding of the factors related to the etiology, development and cessation of cigarette smoking. While this progress has been substantial, many individuals still have a difficult time breaking the smoking habit. We now realize that for many smokers, nicotine is an addictive sub- stance, and smokers therefore benefrt from the integration of pharma- cology and behavioral treatment. The development of nicotine-containing chewing gum has increased our ability to work more effectively with the more nicotine-dependent snioker, and we are well on our way in the de- velopment of other methods for nicotine replacement. These proceed- ings of the World Congress on Pharmacologic "Treatment ofTobacco De- penderrce attest to this progress, and demonstrate the need for continued research in this area. The proceedings also emphasize that we must use what we learn to help develop health policies which can have a benefr- cial effect on the health of our nation. The fine papers examine in great depth the problems of nicotine dc- pendence and pharmacologic treatment. Many scientists have contributed to this report, and I appreciated the opportunity to interact with them and learn from their expertise. For this I am deeply grateful. Judith K. Ockene, Ph.D. Director, Division of Preventive and Behavioral Medicine Department of Medicine University of Massachusetts Medical School Worcester, Massachusetts iii

Acknowledgements This volume contains papers and discussions presented at the World Con- gress on the Pharmacologic Treatment of Tobacco Dependence, held in New York City, November 4-5, 1985. The World Congress brought together researchers and scientists from all over the world to review the latest findings relating to pharmacologic measures for smoking cessa- tion. The Institute for the Study of Smoking Behavior and Policy joint- ly sponsored the event with Merrell Dow Pharmaceuticals Inc., and Aktiebolaget Leo, respectively the distributor and manufacturer of Nicorette chewing gum. Ted Klein and Company of New York coordi- nated the World Congress. These proceedings, produced by the Institute for the Study of Smok- ing Behavior and Policy, would have been impossible without the sup- port and assistance of Merrell Dow Pharmaceuticals. Likewise, the work of the Institute, which housed this publication project, is made possible through grants from the Carnegie Corporation, the W. K. Kellogg Foun- dation, the Cabot Family Charitable Trust, the Conrad Hilton and Alfred P. Sloan Foundations and the National Cancer Institute. Judith Ockene, Ph.D., the editor of this volume, provided our contri- butors with scientific guidance and crafted a volume of unique useful- ness and consistency. Her comments and suggestions to authors were universally thoughtful and appreciated. Sue Hyde, the project coordi- nator, shepherded the proceedings through the editorial and publishing process and provided continuity to authors, editor, and publishing com- pany representatives. Richard Minton painstakingly prepared this manuscript for typography and printing by Puritan Press, Inc. And, of course, our contributors have given the proceedings and the World Con- gress its heart and soul and without them, neither would have occurred. Participants Neal L. Benowitz, M.D. University of California, San Francisco, School of Medicine San Francisco, California, U.S.A. J. Allan Best, Ph.D. University of Waterloo Waterloo, Ontario, Canada Renee Bittoun, Ph.D. St. Vincent's Hospital Sydney, Australia Karl-Olov Fagerstrom, Ph.D. Aktiebolaget Leo Helsingborg, Sweden Alexander H. Glassman, M.D. Colurnbia University College of Physicians and Surgeons New York, New York, U.S.A. Ellen R. Gritz, Ph.D. University of California, Los Angeles, School of Medicine Los Angeles, California, U.S.A. Peter Hajek, Ph.D. Maudslcy Hospital London, England Jack E. Henningfield, Ph.D. National Institute on Drug Abuse Baltimore, Maryland, U.S.A. John R. Hughes, M.D. University of Vermont College of Medicine Burlington, Vermont, U.S.A. Jerome Jaffe, M.D. National Institute on Drug Abuse 13altimore, Maryland, U.S.A. Murray E. Jarvik, M.D. University of California, Los Angeles, School of Medicine Los Angeles, California, U.S.A. Martin Jarvis, M.A, B.Sc., M.Phil. Maudsley Hospital London, England Marcel Kornitzer, M.D. Universite Libre de Bruxelles Brussels, Belgium Edward Lichtenstein, Ph.D. University of Oregon Eugene, Oregon, U.S.A. Teresa Salvador-Llivina, Ph.D. Hospital Clinic i Provincial de Barcelona Barcelona, Spain Judith K. Ockene, Ph.D. University of Massachusetts Medical School Worcester, Massachusetts, U.S.A. Gerry Oster, Ph.D. Policy Analysis Inc. Brookline, Massachusetts, U.S.A. Ovide Pomerleau, Ph.D. University of Michigan School of Medicine Ann Arbor, Michigan, U.S.A. Jed E. Rose, Ph.D. Veterans Administration Medical Center, West Los Angeles Los Angeles, California, U.S.A. M.A.H. Russell, M.R.C.P. Maudsley Hospital London, England v iv

David P. L. Sachs, M.D. Stanford Univcrsity School of Medicine Stanford, California, U.S.A. Nina G. Schneider, Ph.D. University of California, Los Angeles, School of Medicine Los Angeles, California, U.S.A. Saul Shiffman, Ph.D. University of Pittsburgh Pittsburgh, Pcnnsylvania, U.S.A. Stephen Sutton, Ph.D. Maudslcy Hospital London, England Philip Toennesen, M.D. AMTS Hospital of Copenhagen Lyngby, Dennnark Douglas M. C. Wilson, M.D. McMastcr University Hamilton, Ontario, Canada Contents Foreword ............................................... i ... Preface ................................................. tir Acknowledgernents ....................................... iv Participants ............................................. v I: Keynote Remarks by Jerome Jaffe .......................... 1 II: Smoking Behavior and Tobacco Dependence .............. 11 Ovcrvicw by Ellen R. Gritz ........................... . .... 12 HoW Tobacco Produces DruA Dependence by Jack E. Henningfield ...... 19 The "Why" of 7oGacco Dependertce: Undcrlyin~ Reinforcin,q Mechanisms in Nirotine Se1f-Administration by Ovide Pomerleau ............. 32 Pcychosocial Factors in Srnoking and Quitting: Health Be(ie/s, Self-eFficacy, and Stress by Saul Shiffnian ............................... 48 Discussion .............................................63 III: Development of Nicorette: Its Uses and Limitations ....... 81 Overview by John R. Hughes . . . : . . . . . . . . . . . . ...... . ....... 82 Conceptual Framework f r Nicotine Substitution by M.A.1-L Russell ..... 90 Clinical PharmacoloXy cf Nicotine Gum by Neal L. Benowitz ........ 108 Nicotine Polacrilex: Clinical Promises Delivered and Yet to Come by David P. L. Sachs ................................... 120 Problerns of Nicotine Cum by John R. Hughes ....... . ........... 141 Discussion ............................................148 IV: Alternate Forms of Pharmacologic Treatment ............. 153 Overview by Murray E. Jarvik ............................. 154 7i-ansdermal Nicotine as a StrateQy /or Nicotine Replacentent byJed E.Rose ........................................158 Nasal Nicotine Solntion: Its Potential in Srnokin~ Cessation rmd as a Resear(h T<)ol by Martin Jarvis ............................ 167 Clonidine and Ci~arctte Srnokinq Withdrawal by Alexander H. Glassntan, Fay Stetner, and Pamela Raizman ................. 174 Discussion ............................................. 181 V: An International Update on Nicorette .................... 1 87 Pancl Moderated by Karl-Olov Fagcrstriim (Sweden) Panel Participants: Teresa Salvador-Llivina (Spain) Rcnec Bittoun (Australia) Marcel Kornitzcr (Bclgium) Philip Tocnncscn (Denrnark) \1 i vii

VI: Integrating Pharmacologic and Behavioral Approaches .... 195 Ovcrview by Saul Shiffman ............................... 196 Clinic Based Cessation Strategies by Edward Lichtenstein ........... 205 Can Training Family Physicians Improve Compliance with Nicotine Gum Use? by Douglas M.C. Wilson, J. Allan Best, Elizabeth Lindsay-Mclntyre, J. Raymond Gilbert, D. Wayne Taylor, and Joel Singer ......................... 218 Nicotine Chewing Gum in Group Treatment of Smokers by Peter Hajek ....................................... 226 Use o(2 mg and 4 mg Gum in an Individual Treatment Trial by Nina G.Schneidcr ..................................233 Nicotine Guni in the Workplace: Preliminary Report of Two Randomized Trials by Stephen Sutton and Robert Hallett ................. 249 Discussion ............................................254 VII: Economics of Treatment .............................. 263 Overview by Judith K. Ockenc ............................ 264 Cost-Benefit Analysis of Tobacco Dependency Treatment by David P.L.Sachs ................................... 270 The Cost-effectiveness of Nicotine Chewing Gum as an Adjunct to Physician's Advice Against Cigarette Smoking in a Primary Care Setting by Gerry Oster, Danicl M. Huse, Thotttas E. Delea, and Graham A. Colditz ................................ 281 lliscussion ............................................293 VIII: The Future of Pharmacological Treatment for Smoking by M.A.H. Russell ............................ 295 I: Keynote Remarks viii

Keynote Remarks Jerome Jaffe, M.D. National Institute on Drug Abuse Baltimore, Maryland Introduction I'm ccrtainly pleased and honored to be asked to open this first scien- tific session of what appears to me to be the most concentrated gather- ing of researchers on smoking behavior in the free world. I can remem- ber a time when all of those working on the biology and the pharmacology of smoking would have filled only a very, very small corner of this room. That so many have devoted a significant part of their profes- sional careers to pursue the very important questions being considered at this meeting is a credit to the scientists themselves, for this is not an area that fits neatly into any of the established professional disciplines. And it is a credit, also, to the organizations and administrators whose de- cisions provided the economic support for this research. Twenty years ago, at the time of the first Surgeon General's Report, it had been hoped that education and prevention efforts would have a more profound effect on smoking rates. But young people continue to take up smoking, although in smaller numbers than did their parents; 20 percent of high school seniors are daily smokers by the time of gradua- tion. If this generation is like the one that preceded it, at some time in the near future they will try to stop smoking, and many of them will find it difficult to do so. Let us hope that the tools available to help them will be better than those we have had to help their parents, and even better than those we have today. Truly, in the words that end every scientific report, more research is needed. Unanswered research questions In opening this first session, I'd like to mention some of the research questions that seem to me to remain unanswered. The first question is the context of treatment. What is the proper context for using pharmaco- logical treatments, such as nicotine chewing gum? There seem to be two very different perspectives. The work of Russell and Jarvis and their co- workers (13) certainly suggests that minimal intervention, a small effort by the many, many primary care physicians, can produce more cases of cessation than intensive intervention by those who specialize in smok- ing cessation. Yet, there are two drawbacks to this position. One has to do with maintaining the credibility of the pharmacological agents, and the other has to do with the attitudes of third party payers. It does seem that the least effective way to use nicotine replacements is to have them used by those with very little training (13,14), and the most effective way is to combine the pharmacological treatments with other behavioral supports (3,14). To the degree that one uses general practitioners in order to produce the highest number of cases of cessation, one runs the risk of destroying the credibility of this form of treatment, since most smokers who are merely offered a prescription for nicotine gum do not succeed in stop- ping. A few days ago I was talking to a salesperson. She began to talk about her problems with smoking, and she asked if I knew anybody who could help her stop smoking. I hinted that perhaps I knew a few people. I asked her had she ever tried to stop? What did she try? Did she ever try nicotine chewing gum? She said, "Yeah, I tried that. Didn't like the taste." And I asked, "Well, what were you told about it?" As far as I could tell, she was told nothing. At least her memory of her experience was that she had been given no special instruction or information. She was simply given a prescription for gum and told to chew it. It did nothing for her. She was somebody who might have responded had she been given the gum by someone with more training and experience. She was obvi- ously a very heavy smoker, but now she was no longer interested in trying it. We see here the loss of a potential success. The other potential loss inherent in using the minimal intervention ap- proach (which, again, I don't question might actually produce more cases of cessation over some short period of time), is that eventually third party payers will look at the effectiveness of what happens, see the small im- pact, and elect to not support the prescription of pharmacological adju- vants to smoking cessation. In general, the funders of treatment can often be persuaded to support something which is unequivocally or even usually effective. The other issue, however, is one that touches on ethics. If, in fact, the likelihood of success of using a product or a drug is low in the hands of general practitioners, should the patient be so informed? Should the patient be told, "In the hands of a specialist this particular approach would be quite likely to help you. In my hands, since I have very little experience with it, it probably won't help much" What are the ethics of prescribing a drug to somebody with low levels of dependence if the data suggest that it is the most helpful to those with high levels? 2 3

Obviously, some of these questions can be answered by further research, but others involve complex questions of politics, ethics, and eco- nomics. The fact that they're complex, however, does not mean that we can turn away from trying to answer them. The next question is one which is much more amenable to scientific investigation: why is it that nicotine itself, in the form of gum, is rela- tively ineffective in controlling craving for nicotine by inhalation in the form of tobacco (6,17)? Is it simply a matter of dosage or peak plasma levels that are achieved? Would nicotine given by other routes be more effective? A number of researchers who are here at this meeting have con- sidered this question, and it may be that some answers to this question will emerge during this meeting. In the meantime it would be useful to pay attention to just what a prescriber should or ought to tell a patient. As John Hughes said, if nicotine gum doesn't really relieve craving, should we not tell the patient so and suggest that some willpower is still needed? But what does a patient think when told that the part of cessa- tion which for so many is the most difficult to endure-the continued craving for cigarettes-is not affected by the medicine we're about to prescribe? Relapse Next, I'd like to touch on the problem of relapse. While we seem to have made some progress in helping people give up smoking initially, we have not managed the problem of relapse as effectively. Smokers who stop with the help of products like Nicorette are still quite likely to take up smoking again over the subsequent years. Based on work done with other drugs, we can reasonably assume that after repeated ingestion of a drug, the organism is no longer the same. One possible explanation for the change-that is, the vulnerability to relapse-is that learning and condi- tioning of drug effects last far longer than the acute withdrawal syndrome. Can pharmacological agents be developed that would facilitate a process of deconditioning or unlearning? Abe Wikler, who contributed so much to the understanding ofopioid dependence, postulated that learning played a major role in the dependence process, and that if the dependent individu- al, the addict, could be made to self-administer the drug, but get no rein- forcement, eventually the conditioned responses that underlie drug crav- ing would be extinguished. Wikler suggested that if narcotic antagonists could be used to block the effects of self-administered opioids, condi- tioned drug effects would be extinguished. To the degree that relapse was due to such factors, relapse would be prevented and the dependence process could be undone (18). Such a drug has been developed for opioid-dependence. The drug is naltrexone. There are short-acting antagonists as well. An analogous drug exists for nicotine: mecamylamine. It can block the subjective effects of nicotine (9), although it has a number of significant effects on its own. These two drugs have something else in common. As agents for the treat- ment of drug dependence, they work best in theory. Opioid addicts have not flocked to ingest naltrexone (1). Although many clinicians are con- vinced that it has significant benefits in preventing relapse, the primary problem seems to be in persuading addicts that they should take a drug that will help them avoid relapse (1,11,16). ln some instances, this may be due to subtle dysphoric effects of high doses of naltrexone (5). It may also be that opioid addicts are ambivalent about giving up the opioids. Some ambulatory smokers have been given mecanrylamine as a ther- apeutic agent. Anecdotal reports by Dr. Forrest Tennant and co-workers indicate that some smokers find the drug helpful, and that in doses toler- ated by ambulatory patients it does indeed block the rewarding effects of nicotine and does seem to help them stop (15). If a nicotine-blocking agent with fewer side effects could be developed, would it help with relapse? It would be important to know. However, given the history of failure to find a profitable market for naltrexone, who will make the in- vestment to get such a nicotine-blocking drug approved? Criteria for successful treatment Let me turn now to a very different and even more difficult issue- the criteria for successful treatment. If a smoker takes a medicine designed to help him or her stop smoking, and then never smokes again, there's no problem in judging the success of treatment. But what if the smoker merely cuts down from thirty cigarettes to 10 and continues to take whatever pharmacological substance we've prescribed to help, or con- tinues to participate in some other behavioral support? Is that a partial success? Have we interfered with the natural history of smoking? How much benefit has the smoker really obtained? Might that smoker in another program have gone on to stop entirely? And what are the ethics of continuing to prescribe nicotine gum for those smokers who seem to get stuck at "partial abstinence;' i.e., smoking fewer cigarettes? If we look to the experiences of the medical profession with other ad- dictive disorders for guidance on this issue, we will find little help. Even among those clinicians who endorse the use of oral methadone as a use- ful way to control the far more harmful patterns of intravenous heroin use, there is disagreement on what to do about the patient who takes the methadone, butt continues to use I.VV opioids. Some clinicians try to 4 5

increase the dose of methadone. Others, in exasperation, simply discharge the patient, and hope that someday he or she will be more prepared to make a change. It's obvious that we need further studies on the natural history of those who seem to get only halfway with currently available treatment tools. I should probably make it clear that I am not questioning the propri- ety of continued use of therapeutic pharmacological agents such as Nicorette or methadone if these agents are fully effective in controlling the original, far more deleterious drug use patterns. If a patient is able to refrain from tobacco smoking only when provided with nicotine in an oral form, I would support such long-term use (with periodic efforts at gradual withdrawal). New drugs, new combinations Lastly, I'd like to touch upon opportunities for new drugs and new com- binations. Work is progressing in many laboratories that is aimed at un- derstanding the neural substrates and biochemical mechanisms under- lying the reinforcing actions of nicotine and the nicotine withdrawal syndrome. Sorne of the speakers at this conference will undoubtedly talk about nicotine receptors and the sites ofnicotine'.s reinforcing effects. These studies should eventually lead to new ideas for agents that can alleviate craving for nicotine or alleviate one or more of the aversive components of the nicotine withdrawal syndrome. There is already one report from Glassnian and co-workers (2) that clonidine, a drug that appears to sup- press sonic of the components of the opioid withdrawal syndrome, has some effects on craving for tobacco. Should such findings be replicated, one obvious question will be whether drugs like clonidine could be com- bined with nicotine gum or other forms of nicotine to produce enhanced effects. Conclusion I'd like to end by reminding you of what I believe is the most signifi- cant achicvcnrcnt of the research conducted over the past 15 years. In suc- Z cessive reports of the U.S. Surgeon General, cigarette smoking has been identified as the largest snlgle preventable cause of morbidity and mor- tality in the United States. To the best of my knowledge, the same dubi- ous distinction belongs to cigarette smoking in most of the developed countries of the western world. Over the years, the evidence that links smoking to numerous illnesses has merely been strengthened. But in the way that the scientific community views smoking behavior itself, the progress is tantamount to the breaking of new conceptual ground. Despite the fact that some research remains to bee done, there is now more than suffrcient evidence for two propositions. First, for the overwhelming rnajority of regular smokers, smoking is primarily a way of obtaining nicotine; and second, the methods and concepts that have helped to elu- cidate the factors involved in other forms of drug dependence can be use- fully applied to cigarette smoking (4,7,10,12,14). The question is no longer whether tobacco use is a form of drug de- pendence, but how, in an era when not all research possibilities can be pursued, we can best prioritize the important questions about tobacco dependence and its treatment which remain to be answered. Stcven Jobs, one of the founders of Apple Computer Corporation, the little company that shook the computer world with its innovations, was praised for his actions in supporting creativity and in freeing innovators from bureaucratic constraints. But with all of his reverence for creativi- ty, one ofJobs' favorite slogans was: "Geniuses ship." These two words contain the essence of the successful creative enterprise in the entrepre- neurial world, scientific or otherwise. It is necessary to be talented. In- deed, brilliance is taken for granted. But when work requires continued resources and support, true genius consists of not merely theorizing about better understanding or better ways to do something, but, at somc point, of packaging and shipping a useful product. Collectively the scientific enterprise -the pharmaceutical industry and the academic community - has "shipped." It has delivered a product and demonstrated that in the proper context the product can have a significant impact on smoking be- havior. It now remains for us to use the profits to build on the momen- tum generated by this advance and to develop still better ways to help those people who want to give up smoking. Thank you. References 1. Ginzburg HM. Pp. 83-101 in B Stimmel (Ed.) Advances in Alcohol & Substance Abuse, Volume 5. New York: The Haworth Press, 1986. 2. Glassman AH, Jackson WK, Walsh BT, Roose SP. Cigarette crav- ing, smoking withdrawal, and clonidinc. Sciencc, 1984; 226:864-866. 3. Hall SM, Killen Jll. Psychological and pharmacological approaches to smoking relapse prevention. Pp. 131-143 in J Grabowski, SM Hall (Eds.) Pharmacological Adjuncts in SmokinA Cessation, NIDA Research Monograph 53, DHHS Pub. No. (ADM)85-1333, 1985. 4. Henningfield JE. Pharrnacologic basis and treatment of cigarette srnoking. J. Clin. Psychiat. 1984; 45:24-34. 7 6

5. Hollister LE, Johnson K, Boukhabza D, Gillespie HK. Aversive ef- fects of naltrexone in subjects not dependent on opiates. Drug Alco- hol Depend. 1981; 8:37-41. 6. Hughes JR, Hatsukami DK, Pickens RW, Krahn D, Malin S, Luknic A. Effect of nicotine on the tobacco withdrawal syndrome. Psy- chopharm. 1984; 83:82-87. 7. Jaffc JH, Kanzler M. Smoking as an addictive disorder. Pp. 4-23 in NA Krasnegor (Ed.) Cigarette Smokinq as a Dependence Process, Washing- ton, DC: U.S. Government Printing Office, NIDA Research Mono- graph, 1979. 8. Jarvis MJ, Raw M, Russell MAH, Feyerahend C. Randomized con- trolled trial of nicotine chewing gum. Brit. Med. J. 1982; 285:537-540. 9. Nemeth-Coslett R, Henningfield JE, O'Keeffe MK, Griffiths RR. Effects of inecamylamine on human cigarette smoking and subjec- tive ratings. Psychopharm. 1986; 88:420-425. 10. Polin, W. The role of the addictive process as a Key Step in causa- tion of all tobacco-related diseases. J. Am. Med. Assoc. 1984; 252:2874. 11. Resnick RB, Schuyten-Resnick E, Washton AM. Assessment of nar- cotic antagonists in the treatment of opioid dependence. Ann. Rev. Pharmacol. Toxicol. 1980; 20:463-474. 12. Russell MAH. Tobacco smoking and nicotine dependence. Pp. 1- 46 in RJ Gibbons, Y Israel, H Kalant, RE Popham, W Schmidt, RG Smart (Eds.) Research Advances in Alcohol and Drug Problems, New York: Wiley, 1976. 13. Russell MAH, Merrirnan R, Stapleton J, Taylor W. Effect of nico- tine chewing gmn as an adjunct to general practitioners' advice against smoking. Brit. Med. J. 1983; 287:1782-1785. 14. Schneider NG, Jarvik ME. Nicotine gum vs. placebo gum: Com- parisons of withdrawal symptoms and success rates. Pp. 83-101 in J Grabowski, SM Hall (Eds.) Pharmacological Adjuncts in Smoking Cessation, Washington, DC: U.S. Government Printing Office, NIDA Research Monograph 53, DHHS Pub. No. (ADM)85-1333, 1979. 15. Tennant FS, Tarver AL. Withdrawal from nicotine dependence us- ing niecamylamine: Comparison of threc-week and six-wcek dosage schedules. Pp. 291-297 in LS Harris (Ed.) Problems of Drug Dependence 1984, Washington, DC: U.S. Government Printing Office, NIDA Research Monograph 55, DHHS Pub. No. (ADM)85-1393, 1985. 16. Washton AM, Gold MS, Pottash AC. Naltrexonc in addicted phy- sicians and business executives. Pp. 185-190 in LS Harris (Ed.) Problems of Dri±g Dependence 1984, Washington, DC: U.S. Govern- ment Printing Office, NIDA Research Monograph 55, DHHS Pub. No. (ADM)85-1393, 1985. 17. West RJ, Jarvis MJ, Russell MAH, Carruthers ME, Feyerabend C. Nicotine replacement on cigarette withdrawal syndrome. Brit. J. Ad- dict. 1984; 79:215-219. 18. Wikler A. Opioid Dependence: Mechanisms and Treatment. New York: Plenum Press, 1980. 9 8

II: Smoking Behavior and Tobacco Dependence 00 fA ~ ®

Overview: Smoking Behavior and Tobacco Dependence Ellen R. Gritz, Ph.D. Director, Division of Cancer Control Jonsson Comprehensive Cancer Center University of California, Los Angeles Introduction In 1985, biomedical and social scientists have at their command the most sophisticated research techniques ever developed to study the phar- macological and psychoactive properties of chemical-containing sub- stances. The papers presented in this section summarize widely accept- ed findings regarding the dependence-producing properties of nicotine and the psychosocial factors influencing regular cigarette smoking. However, defining cigarette smoking as a form of substance dependence, or addiction, does not occur solely in the context of science and medi- cine. It has profound consequences for society in sociocultural, economic, legal and political contexts, and thus implications for public policy. Remember, we are discussing the dependence-producing properties of a self-administered substance that is the leading cause of premature death and disease in the U.S. today. Before introducing the three papers which constitute this section, a brief historical survey of tobacco may broaden our understanding of this public health perspective. Tobacco use in society Tobacco, native to the western hemisphere, was well integrated into the cultures of the South, Central and North American peoples when European explorers first reached our shores in the 15th and 16th centu- ries. An early record of tobacco use is depicted on a Mayan stone monu- ment in the remote Yucatan site of Palenque (c. 600-900 A.D.): "God L smoking a large cigar" portrays a man-figure taking in smoke in a magico- religious representation (11). The gods and spirits, it is widely held, crave tobacco smoke so intensively that they are unable to resist it.... Just as the tobacco shaman of the Warao requires tobacco smoke with tremendous physiological and psychological urgency, and is literally sick without it, so the gods await their gift of tobacco smoke with the craving of the addict, and will enter into mutually beneficial relationships with man so long as he is able to provide the drug.... No wonder that in the indigenous world tobacco was considered too sacred for secular or purely hedonistic use. (16, pp. 455-457) While we know little about the early specific ritual use of the tobacco plant (primarily the species Nicotiana rustica, higher in nicotine content than Nicotiana tabacum, our common commercial species), we do know it served as a ritual narcostimulant, especially in South America (16). The prominent anthropologist, Johannes Wilbert, has described the potent , pharmacologlc effects oI LIIe most ancient ivr,ii of tobacco ingestion, drinking a liquid preparation of tobacco boiled or steeped in water or macerated in spittle; alternatively, it could be licked from the fingers or from a stick dipped in the liquid or taken through the nose (16). The toxic effects of nicotine (e.g., pallor and tremor) were apparent after as few as two or three doses of the concentrated liquid and repeated ingestion of large doses was used by shamans to induce intensive dream visions, ac- companied by the more severe toxic effects of nausea, vomiting and even a comatose state. Tobacco was taken by itself or mixed with hallucino- gens such as Coca datura, Banisteriopsis caapi (yahuasca) or the cactus Trichocer- eus pachanoi. In contrast to its currently recognized role as a major cause of the leading killers, heart disease and cancer, tobacco served as a medicinal and hygenic agent in the folk medicine of many pre-Columbian cultures (11,16). Tobacco's reputation as a healing agent was carried to Europe. Nicot, the 16th century French ambassador to Lisbon after whom the tobacco ge- nus is named, defined Nicotiana in his French-Latin dictionary as follows (16): "This is an herb of marvelous virtue against wounds, ulcers, noli me tangere [lupus of the face], herpes, and all other things." And in mid-17th century Europe, the smoking and chewing of tobacco served as a prophylactic agent and therapeutic remedy for the plague (10,11). It is a matter of history how rapidly tobacco use spread around the world and became a part of the everyday culture of most societies, a.so- cially accepted recreational and habitual behavior. This occurred despite a variety of civil banning attempts, religious prohibitions and horrible punishments, including mutilation and the death penalty. Disapproval ot tobacco consumption was not based on known disease-producing ac- tions; in 1568, Benzoni called it" .. a pestiferous and wicked poison from the devil .."(11). Nor was it based on more than anecdotal reports of dependence. According to Las Casas in 1877, "I knew Spaniards who were accustomed to take tobacco, and, being reprimanded for it by telling 12 13

them it was vice, they replied that they were unable to cease using it. I do not know what pleasure or benefit they found in it" (11). Thus, the policy governing licit use of tobacco dcpended upon the attitudes and beliefs of governmental and/or religious leaders. Of course, economic fac- tors were important then, as now, for sales and revenue purposes (taxa- tion), only the stakes arc much higher in the 20th century. High tar and nicotine cigarettes, those with high dependence-producing potential, are aggressively marketed in the developing countries by American and Eu- ropean tobacco companies and the growing of tobacco as a major cash crop is encouraged (2,3). Tobacco dependence This discussion brings us to the 20th century, when an exploration of the pharmacologic properties of nicotine was undertaken and the role of nicotine as a reinforcing and dependence-producing pharmacologic agent first became recognized. The powerful range of effects of tobacco on the central nervous system, peripheral nervous system and directly upon major body organs is largely attributed to nicotine, which is respon- sible for about 95% of the total alkaloid content in tobacco and is the most powerful alkaloid (5,6). However, the scientific study and dcfini- tion of nicotine dependence occurred much later than the investigation of the pharmacodynamics and physiologic actions of the chernical. Indced, the landmark Report of the Surgeon General in 1964, while estab- lishing the causal relationship between smoking and lung cancer, was only able to term smoking an "habituation" (15). Beginning in the early 1970s, the role of nicotine as a primary rein- forcer in smoking behavior was seriously proposed, supporting scicn- tific evidence brought forth, and the dependence-producing nature of cigarette smoking discussed (7,9,12,13,14). A major issue in the consider- ation of dependence was the similarities and differences between cigarette smoking and other well known drugs of abuse, such as alcohol, heroin and barbiturates (8). That issue, at least on a medical decision-making level, was resolved in 1980 by the American Psychiatric Association in the Diagnostic and Statistical Manual (f Mental Disorders (DSM-III). That prestigious medical association established a Tobacco Dependence Disorder, which located cigarette smoking in the category of Substance Use Disorders and re- quired evidence of tolerance or withdrawal as part of the diagnostic criteria (1). Impairment in social or occupational functioning, while a potential consequence of the reaction of others to smoking, is not a direct effect of tobacco use and so was excluded from the definition, as was the concept of tobacco dependence as substance abuse. The essential features of the Tobacco Dependence Disorder are continuous use of tobacco for at least one month with either: (1) unsuccessful attempts to stop or significantly reduce the arnount of tobacco use on a permanent basis; (2) the development of Tobacco Withdrawal; or (3) the presence of a serious physical dis- order (e.g., respiratory or cardiovascular disease) that the individu- al knows is exacerbated by tobacco use (1, p. 178). [The Tobacco Withdrawal Syndrome was separately defined and included among the Organic Mental Disorders in the DSM-III.]. The American Psychiatric Association was careful to qualify its defini- tion of Tobacco Dependence so that dysphoria or illness in connection with the behavior was a necessary criterion: In practice, this diagnosis will be given only when the individu- al is seeking professional help to stop smoking, or, in thc judg- nunt of the diagnostician, the use of tobacco is seriously affect- ing the individual's physical health. It should also be noted that a heavy smoker who has never tried to stop smoking, who has never developed Tobacco Withdrawal, and who has no tobacco- related serious disorder, according to the criteria in this manual, does not have the disorder of Tobacco Dependence, even though physiologically the individual is almost certainly dependent on tobacco (1, pp. 176-177). Scientific evidence for tobacco dependence There are clearly some social and policy-directed interpretations im- plicit in this discussion of the syndrome. The APA was establishing a criterion that stipulated that if his/her smoking wasn't bothering the smok- cr, either in terms of health or behavior (regarding stopping), then the diagnosis of a dependency disorder was not warranted. The scientific determination of the criteria for tobacco dependence has continued to progress and, in particular, the Addiction Research Center of the National Institute on Drug Abuse has conducted a series of care- ful studies comparing nicotine and smoking to a variety of drugs of abuse. The conclusion which has emerged from this research is a strong state- ment of the dependence-producing properties of nicotine. Jack Henning- field, a noted scientist in this area, will present the findings of these studies. His multifaceted discussion of dependence will include: the orderliness ofsmoking behavior and its sensitivity to nicotine i>> nipulations; its dis- criminability, psychoactivity, reinforcing and euphoriant properties; and its physiologic dependence potential. 14 15

The second presenter in this section is Ovide Pomerleau, who has ex- plored in his research the ways in which nicotine and smoking may pro- vide direct physiological and psychological reinforcement, independent of the relief of withdrawal. In this biobehavioral conceptualization of smoking, he discusses both the positive and negative reinforcing conse- quences of smoking. The former include pleasure and its enhancement, the facilitation of task performance and the improvement of memory; the latter include the reduction of anxiety and tension, antinociception, the avoidance of weight gain and the relief of nicotine withdrawal. Dr. Pomerleau also outlines the putative neuroregulatory mechanisms for these reinforcing actions in an intriguing linking together of the multi- ple chemical systems governing our brains and behavior. Both Drs. Henningfield and Pomerleau have been careful to cmpha- size the complexity of smoking behavior and its multifactorial nature. Saul Shiffman develops this sector of the overall diagram, pointing out the interrelationship and interaction between pharmacologic and psy- chosocial factors in smoking. His paper concentrates on two issues: the role of health beliefs in continued smoking and motivation to stop; and the role of stress as it may facilitate the initiation of smoking and impede cessation. Dr. Shiffman will illustrate his discussion with findings from his own research on the situational and affective analysis of relapse crises and on successful and unsuccessful coping mechanisms. He points to the increased effectiveness of combining the pharmacologic agent and a be- havioral program in studies evaluating nicotine chewing gurrr as an ad- junct in smoking cessation treatment. Conclusion In concluding this overview, let me offer a few remarks on the policy implications of defining tobacco smoking in terms of nicotine depen- dence. First of all, underscoring the medical aspect of nicotine depen- dence by placing a treatment tool (a prescription product) solely under the control of physicians/dentists can be both a blessing and a curse. For the first time, practitioners have an agent in their pharmacopcia to treat smoking whic lriias survived the stringent tests of placebo-controlled de- signs. Nicotine gum and, possibly, other new pharmacologic substitutes currently being developed, provide a potentially effective smoking- intervention tool. Formerly, many practitioners felt they lacked the cotn- petence to treat or counsel patients in smoking cessation. On the other hand, the dangers of providing a "magic bullet;' at least in the patient's eyes, loom large; hard work and behavioral/psychosocial learning and practice are still an integral part of every successful smoking cessation effort. If too much faith or emphasis is placed upon this tool, then both physicians and patients can become discouraged when the effort is difficult or initially unsuccessful. Both may fail to recognize the complexity of the disorder and the need for a "coping" model rather than a "mastery" model in treatment strategy. Second, there are policy implications for third party reimbursers of medical expenses. Providers of treatment for nicotine dependence, such as physicians and psychologists, should receive reimbursement at a rate comparable to treatment for other medical and mental health conditions. Certainly the dollar expenditure for treating smoking dependence is ex- trao_rdinarily cost-effective compared with that for treating lung cancer, not to mention the inestimable price of human suffering. Third, the widespread acceptance of tobacco dependence may have pro- found implications for the tobacco litigation being conducted in this coun- try (4). Dependence introduces a non-voluntary quality into smoking be- havior; this is especially poignant when one considers that most smoking initiation occurs in the teenage years when adolescents consider them- selves healthy and invulnerable to the diseases of later life, no matter what the cause. The tobacco industry is firmly opposed to acknowledging tobacco dependence and worked long and hard to have such a warning removed from the set of new Surgeon General's rotational warnings on cigarette packages. Finally, the inclusion of tobacco use among the dependence-producing behaviors carries powerful implications for substance abuse prevention programs with youth. It means we have the charge to educate and arm vulnerable adolescents against the dangers of tobacco as a companion and prototype of drugs of abuse, as a conveyor of adverse health effects both immediately and distantly in time, and finally, as perhaps the most in- sidious form of substance dependence they may ever face in their lives. This charge alone, if carried out successfully, can help us attain the Sur- geon General's goal of producing a smokcfree society in the year 2000. References 1. American Psychiatric Association. Diagnostic and Statistical Manual, Third Edition. Washington, DC: American Psychiatric Association, 1980. 2. Canadian Council on Smoking and Health. Proceedings on the 5th World Conference on Smoking and Health, Volumes 1 and 2. Win- nipeg, Canada, 1983. 3. Currie K, and Ray L. Going up in smoke: The case of British Ameri- can Tobacco in Kenya. Social Sci. & Med. 1984; 19(11):1131-1139. 16 17

4. Garner DA. Cigarette dependency and civil liability: A modest proposal. Southern Cal f Law Review 1980; 53:1423-1465. 5. Goodman AG, Goodman LS, Gilman A. (Eds.) The Pharmacological Basis of Therapeutics, Sixth Edition. New York: MacMillan, 1980. 6. Gorrod JW Jenner P. The metabolism of tobacco alkaloids. Pp. 35- 78 in WJ Hayes, Jr. (Ed.) Essays in Toxicology, Volume 6. New York: Academic Press, 1975. 7. Gritz ER. Smoking behavior and tobacco abuse. Pp. 91-158 in NK Mello (Ed.) Advances in Substance Abuse. Greenwich, CT JAI Press, 1980. 8. Jaffe JH, Kanzler M. Smoking as an addictive disorder. Pp. 4-23 in NA Krasnegor (Ed.) Cigarette Smoking as a Dependence Process. NIDA Research Monograph 23. Rockville, MD: Department of Health, Education, and Welfare. PHS.ADAMHA.NIDA, 1979. 9. Jarvik ME. The role of nicotine in the smoking habit. Pp. 155-190 in WA Hunt (Ed.) Learning Mechanisms in Smoking. Chicago: Aldine, 1970. 10. Laufer B. Introduction of Tobacco into Europe. Chicago: Field Museum of National History, Anthropology Leaflet No. 19, 1924. 11. Robicsek F. The Smoking Gods. Norman, OK: University of Okla- homa Press, 1978. 12. Russell MAH. Cigarette smoking: Natural history of a dependence disorder. Brit. J. Med. Psych. 1971; 44:1-16. 13. Russell MAH. Tobacco smoking and nicotine dependence. Pp. 1-47 in RJ Gibbons, Y Israel, H Kalant, RE Popham, W Schmidt, RG Smart (Eds.) Research Advances in Alcohol and Drug Problems. New York: Wiley, 1976. 14. U.S. Dcpartment of Health, Education, and Welfare. NA Krasnegor (Ed.) CiXarette Smoking as a Dependence Process. NIDA Research Mono- graph 23. Rockville, MD: Department of Health, Education, and Wel- fare. PHS.ADAMHA.NIDA, 1979. 15. U.S. Public Health Service. Smoking and Health. Report of the Advi- sory Committee to the Surgeon General of the Public Health Serv- ice. Washington, DC: U.S. Department of Health, Education, and Welfare, Public Health Service Publication No. 1103, 1964. 16. Wilbert J. Magico-religious use of tobacco among South American Indians. Pp. 439-461 in V Rubin (Ed.) Cannabis and Culture. The Hague: Mouton, 1975. How Tobacco Produces Drug Dependence Jack E. Henningfield, Ph.D. Chief, Biology of Dependence and Abuse Potential Assessment Laboratory Addiction Research Center National Institute on Drug Abuse and Assistant Professor of Behavioral Biology Department of Psychiatry and Behavioral Sciences The,Johns Hopkins [Iniversity School cf Medicine Baltimore, Maryland Introduction The main purpose of this paper is to summarize some of the data which reveal tobacco use to be a dependence process and nicotine as a model dependence-producing drug in order to show the mechanism of tobac- co dependence. On the one hand, the mechanism of tobacco dependence is quite simple: when tobacco products are used as advertised by manufac- turers (and most consumers actually use them this way), nicotine is de- livered to the central nervous system; repeated use then leads to nicotine dependence. On the other hand, any form of drug dependence is a conr plex interaction of pharmacologic and nonpharrnaeologic factors in which the relative contribution of each factor may vary. An appreciation of the complexities is useful in understanding the biological process of tobacco dependence and may be necessary in the development of effective treatments. Tobacco use and cocaine use. To better understand tobacco depen- dence, it may be useful to place it in the context of another dependence process: cocaine dependence. Such a perspective provides a variety of in- teresting issues to consider when discussing why people use tobacco. If ten scientists, clinicians, and cocaine users were asked to explain why de- pendence to cocaine is so powerful, they would probably provide ten different reasons. Among them, they might say that cocaine is a power- ful reinforcer for animals and humans; it is a powerful cuphoriant; it en- hances mood and affect; it enhances performance on a variety of tasks; it controls behavior via its effects on catecholamines; compulsive 18 19

~10 cocaine users have personality defects which make them vulnerable to dependence; and once tolerant to cocaine, abstinence is unpleasant for users and is to be avoided at all costs. Alternatively, some might say that the process is essentially nonpharmacologic because some users do not have difficulty controlling or ending their patterns of use. There are many other reasons that might be given for the use of co- caine. Most of these explanations are not mutually exclusive: each one may contribute to any individual case of cocaine dependence. A similar illustration could be made for heroin, alcohol and barbiturates. It is now well-established that tolerance may develop to the intoxicating and de- bilitating effects of most drugs of abuse; physical dependence and with- drawal are neither necessary nor sufficient to produce compulsive drug use; and the effects of drugs that lead to their compulsive use arc diverse and not necessarily apparent to an observer. A few other observations about patterns of cocaine use may help shed some light on the extent and perniciousness of tobacco dependence. In just a few decades, we have seen the social attitude regarding cocaine use shift from relatively limited to a significant public health concern. Why? As cocaine use becomes widespread and continues to increase, cocaine- related emergency room admissions and deaths have dramatically in- creased. Additionally, there now exists a growing and vocal population of patients seeking treatment for a disorder over which they have little control. What brought about this drastic change? The pharmacology of cocaine has not changed. Rather, a highly addictive drug with perceived low toxicity became rnore widely available and socially acceptable, and its relative price declined, a process not dissimilar to that of tobacco in the first few decades of the Twentieth Century. Overview and definitions. We will review the data that now con- clusively show that tobacco use leads to nicotine dependence. When pat- terns of nicotine use are surveyed in the context of other drugs of de- pendence, it is evident that nicotine is more similar to other drugs than it is different from them. Some of these similarities will be described in the next section of this paper. To determine that a chemical can control the behavior of a potential user, there are relatively standard methods of testing used in studies with both human and animal subjects: the results of recent studies will be described later in this paper. Finally, a few in]- plications of these observations for the pharmacologic treatment of tobac- co dependence will be offered. The approach taken here is an empirical or descriptive approach in which tobacco/nicotine are characterized with respect to other substances of abuse, not necessarily with respect to any particular definition of a 20 dependence-producing substance. This course is taken because, while defi- iutions will continue to evolve, the pharmacology of known substances will not. In other words, as will be described later, nicotine is known to be a dependence-producing substance, not simply because it qualifies on the basis of prevalent definitions of dependence (13), but rather because its effects on critical measures are sitnilar to those of known dependence- producing substances, such as heroin, cocaine, and alcohol. For the purposes of the discussion at hand, it is helpful to define a few terms whose usages vary widely. Drug dependence or drug abuse: substance- seeking behavior that is controlled, in part, by the central nervous sys- tern activity of a constituent drug; nonpharmacologic factors may be oper- ative; tolerance and withdrawal are neither necessary nor sufficient. The term drug addiction is sometimes used interchangeably with drug depen- dence, but will not be used in this paper because of its more widespread imprecise usage that includes non-drug-mediated habitual behavior. De- pendence potential: the direct effects of a drug that lead to tolerance and to physical dependence. Tolerance is assessed by the occurrence of diminished responsiveness to the drug and physical dependence is assessed by a syn- drome of withdrawal symptoms when drug administration is terminat- ed. Abuse liability: the effects of the drug which result in its control over behavior and hence in its control over self-administration of the drug, often in the face of mounting costs and adverse effects. Comparison of tobacco dependence to other forms of drug dependence Tobacco and its use are complex. Tobacco use is a common behavior which, at first blush, would seem very simple to study and to understand. On the contrary, as we now know, tobacco products themselves, like other drugs, are exceedingly complex mixtures of naturally occurring and syn- thesized substances which can obscure studies of the action of individu- al components. The drug-seeking behavior, which is the interface be- tween the substances and their effects, is no less complex and is the result of interaction among drug and non-drug factors. Tobacco use is an orderly and controlled behavior. Despite the complexities of tobacco usage, a systematic analysis reveals that the be- havior is orderly and lawfully determined by the same parameters that control other forms of drug self-administratiou. Though estimates vary, it has been demonstrated that one-third to two-thirds of all people who experimentally smoke as few as two cigarettes go on to become regular and dependent smokers (6). Moreover, a recent survey (conducted by 21

01N myself and colleagues at the Addiction Research Centcr-unpublished data) of patterns of development of dependence has shown that, with both cigarettes and smokeless forms of tobacco (as with other drugs of abuse) there is a marked escalation of dosage over time ("graduation") which may continue for several years before relative stability of intake occurs (Figure 1). Once patterns of smoking have stabilized, they become regular from day to day and even from cigarette to cigarette. For instance, when cigarette smokers are permitted to smoke in a laboratory situation that permits measurement with little external restraint or experimental interference, patterns of cigarette smoking resemble those of animals who self- administer stimulants such as amphetamine and cocaine. A microanaly- sis of patterns of puffing and inhaling similarly rcvcals an orderly and controlled behavior. Depriving cigarette smokers of cigarettes for inter- vals as brief as 60 minutes can result in increased cigarette smoking and increased preference for nicotine. Tobacco use is sensitive to nicotine dose manipulations. IDose can be clinically and experimentally varied in many ways. These methods in- clude changing the nicotine and/or tar level of the cigarette smoke, provid- ing nicotine by other routes of administration (e.g., 1 V, polacrilex or chewing gum), or blocking nicotine's actions at the ganglia with peripherally-acting (e.g., pentolinium) or centrally-acting (e.g., mecarnyla- mine) blockers (4). The general findings which have emerged from such studies may be summarized as follows: (1) nicotine dose is a functional determinant of tobacco intake; (2) when nicotine intake levels deviate from an upper and lower boundary, compensatory changes in tobacco self-administration occur (15); (3) pretreating cigarette smokers with nicotine by other routes of administration decreases tobacco self-adrninistration and/or reduces the level of preferred nicotine dose level; (4) pretreatment of cigarette smokers with mecamylamine, but not pentolinium, increases levels of tobacco self-administration and/or increases levels of preferred nicotine dose, showing that the effect is centrally mediated; (5) increasing the rate of nicotine excretion from the body by decreasing urinary pH levels increases the amount of tobacco smoked (see Bcnowitz in this volume). These findings are often lumped under the rubric "the nicotine dose titra- tion hypothesis." In spite of the data summarized above, however, the va- lidity of the titration hypothesis continues to be debated, and the out- come seems more dependent on the definitions used; therefore it may 22 DEPENDENCE PROCESS: DOSE GRADUATION 1 30 20 r_ Q 10 9 SMOKELESS `~ ---- - -----a --"ff- CIGARETTES 0 0 I r 0 1 2 3 4 5 6 7 8 YEARS OF TOBACCO USE 0 Figure 1. This figure shows a preliminary graph of results from a sur- vcy of smokeless tobacco users and cigarette smokers (unpublished data from the Addiction Research Center) regarding their daily tobacco in- take (shown on the y-axis) at various time points (shown on the x-axis), including and following their first day of tobacco use. Data from 32 sub- jects are presented for the smokeless tobacco function, and data from 13 representative subjects are presented for the cigarette function. The gradual escalation of dosage followed by relative stability is similar to that reported by persons dependent on alcohol, sedatives, stimulants and narcotics (opioids). 23

be most useful to simply discuss the phenomena in terms of dose-response relations as is done when other drugs of abuse are discussed. Nicotine has a significant physiologic dependence potential. An- other feature which nicotine shares with many other drugs of abuse is the development of tolerance and physiological dependence when the drug is repeatedly administered. While tolerance is not a factor that readily distinguishes drugs of abuse from non-abused drugs (e.g., chlorproma- zine), and physiologic dependence does not necessarily lead to abuse (e.g., most post-operative patients treated with narcotic analgesics do not sub- sequently abuse the drugs) these factors are nonetheless important to the understanding and treatment of drug dependence. Tolerance to drug effects is established either by the observation of a diminished response to repeated doses of a drug, or to the requirement for increasing doses to achieve the same effect. Tolerance to a variety of behavioral and physiologic effects of nicotine has been widely studied (2,18). More recently, it has been confirmed that tolerance also develops to those effects of nicotine considered critical to its abuse liability; namely, its psychoactive and euphoriant effects. For instance, in a study of in- travenous nicotine administration, when a subject was asked to rate the pleasurable effects from successive injections given at ten-minute inter- vals, scores decreased with serial injections (6). Tertnination of chronic nicotine administration is followed by an orderly syndrome of withdrawal phenomena which may include the following signs and symptoms: decreased heart rate, altered clectroencephalographic activity and evoked cortical potentials, increased desire to smoke, anxi- ety, irritability, difficulty concentrating, impaired cognitive performance abilities, headache, drowsiness, insomnia, gastrointestinal disturbances, decreased catecholarnine excretion, tremor and decreased basal metabolic rate. The syndrome may be reversed by nicotine replacement; the degree of reversal is dose-related such that low doses of nicotine may partially relieve withdrawal-related discomfort, and higher doses, or return to ad libitum tobacco use, may fully reverse the syndrome. The above materi- al is a composite of information available from several reviews (6,7), re- cently completed studies (11) and an unpublished series of studies from the Addiction Research Center. Taken together, these findings indicate that nicotine itself is critical to both the establishment and treatment of the syndrome. The studies confirm that a qualitatively similar syndrome of withdrawal occurs when repeated administration of cigarettes, smoke- less tobacco or nicotine gum is terminated, that the syndrome is charac- terized by behavioral and physiologic signs and symptoms, and that nico- tine gum (when administered correctly) can be used to treat withdrawal due to tobacco abstinence. Tobacco may produce useful effects. Tobacco, like many other sub- stances of abuse, produces effects often considered of utility or benefit to the user. Such effects are sometimes termed "therapeutic"; however, such a term may be misleading since (1) the most widely used forms of tobacco cause death and disease, and (2) it has not been clearly established to what degree the beneficial effects are direct actions of nicotine rather than a reversal of detrimental effects of withdrawal following long-term use of tobacco. In the nicotine dependent person nicotine can relieve anxi- ety and stress, can help control appetite and weight, alter mood and feeling state, and perhaps enhance performance and memory (5,7,15). In one of our recent studies of cognitive performance at the Addiction Research Center we found that performance on learned tasks and memory tests was significantly impaired following 8 to 12 hours of tobacco depriva- tion, and that such impairments could be reversed in dose-related fashion by administration of nicotine in the form of the polacrilex (unpublished data). Abuse liability of nicotine As the foregoing summary review has indicated, many effects of tobac- co and patterns of tobacco self-administration have much in common with prototypic substances of abuse and the patterns of behavior which they engender. At this point it is reasonable to question whether nico- tine itself can control the user's behavior in the same way that heroin and cocaine can control a user's behavior. In other words, is nicotine a drug with significant liability for abuse? Recent investigations using methods which were developed to quantitate the abuse liability of other substances such as cocaine and heroin have confirmed that nicotine does meet es- tablished criteria as a drug with significant liability for abuse. I will pro- vide a brief summary of these findings. Nicotine is discriminated by animals and is psychoactive in humans. The most prominent feature of abused drugs is that they produce effects in the central nervous system that change the way the person or animal feels or behaves; such studies conducted with animals are often termed tests for "discriminative effects" whereas such studies with humans are commonly termed tests for "psychoactivity" Several well-controlled animal discrimination studies have been conducted with nicotine. The findings may be sununarized as follows: (1) nicotine produces dose-related discritninable effects; (2) effects are blocked by centrally-acting, but not peripherally-acting, ganglionic blockers; (3) animals trained to identify 24 25

other drugs rarely "niistake" (generalize) nicotine for saline or a barbiturate, but may mistake nicotine for amphetamine (17). Analogous studies with human subjects produce similar findings. These arc as follows: (1) nicotine produced dose-related psychoactivity (dis- criminable effects); (2) effects were attenuated by centrally-acting, but not peripherally-acting, ganglionic blockers; (3) above threshold dose lev- els, persons with histories of polydrug abuse often identified nicotine in- jections as cocaine or amphetamine (10). Nicotine is a euphoriant for humans. Another effect of most drugs of abuse is that the substance serves as a euphoriant. For drug abuse lia- bility studies, euphoria is operationally defined and quantitated using structured questionnaires. The most widely used and validated question- naires are various forms of the Drug Liking scale of the Single Dose Ques- tionnaire, and the Morphine Benzedrine Group (Nll3G) scale of the Ad- diction Research Center Inventory (14). In one such drug abuse liability study, volunteer subjects were given multiple doses of nicotine, both in- travenously and in the form of tobacco smoke. The results were clear and consistent: nicotine adtninistration produced dose-related increases in eu- phoriant scale scores. In fact, as shown in Figure 2, nicotine produced a steep dose effect curve that was similar to curves produced by cocaine, amphetarnine and morphine, but not to curves produced by the nona- bused analgesic, zomeperac. Interestingly, as shown in the figure, despite the common assumption that sugar is an addictive substance, intravenous injections of glucose did not produce the effects characteristic of mor- phine, cocaine, or nicotine. Nicotine is a reinforcer for animals. By definition, drugs that are voluntarily self-administered are positive reinforcers or rewards; a corol- lary of this observation is that the ability of a drug to control the behavior of a user is related to its efficacy as a reinforcer. Reinforcing efficacy of drugs is assessed using "self-administration" studies, in which the organism is permitted to take the drug to allow better control of non-drug fac- tors; the drug is often available via the intravenous route. Drug self- administration studies using animals are accurate tests of the ability of a compound to control behavior, and it has now been demonstrated that nicotine functions, to some degree, as a positive reinforcer for five spe- cies of animals (8). Whereas results in some studies were equivocal, re- cent studies, most notably by Goldberg and his colleagues (8) have con- firrned that nicotine can function as a highly efficacious reinforcer in animals. 2 1 w 0 lz 0 U 0 w J 2 U U) 1 1- Z Q 0 _ Ir O tZ D w 2 1 0 ~--0 ~ P 1 8 MORPHINE L (sc) ZOMEPERAC L (PO) 1 1 16 32 L 30 r° I '-!~' P 200 400 800 2 NICOTINE ~ I- (IV) 1 0 - O ~ I I I P .75 1.5 3.0 d-AMPHETAMINE 2 1 I I I P .75 15 30 GLUCOSE L (IV) . 2 1 0 LO I I I P 10g 20g 40g DRUG DOSE (MG) Figure 2. This figure presents data from a series of abuse liability studies conducted primarily at the Addiction Research Center. The cocaine data are standardized and reported from a study conducted by Fischman (3). The glucose data arc recently collected and previously unpublished from the Addiction Research Center. The findings that Liking Scale scores are directly related to dose and exceed placebo values are important in iden- tifying dependence-producing drugs. Intravcnous nicotine produced the same elevated dose-response function as the representative narcotic (mor- phinc) arid a prototypic stimulant (d-amphetaminc). These data are also consistent with the negligible abuse liability of zomeperac. 26 27

Nicotine is a reinforcer for humans. A series of nicotine self- administration studies using human cigarette smokers as subjects has re- cently been completed at the Addiction Research Center. Subjects were tested during daily three-hour test sessions in which they were not per- mitted to smoke, but they were permitted to take injections of nicotine. Each subject was equipped with a catheter to a forearm vein and an au- tomatic injection apparatus which was controlled by pressing a small lever. The main finding was that nicotine was voluntarily self-administered and served as a reinforcer (9). In addition, some subjects graduated their to- tal dose intake across sessions, in much the same manner as tobacco users had reported in the earlier described survey study (Figure 1). When nico- tine was replaced with saline (placebo), subjects first increased, then decreased their injection rates. One subject reported that it felt analogous to "switching from a real cigarette to Carltons [a highly ventilated, ultra- low nicotine brand of cigarette] and that taking injections just wasn't worth the cffort." Pharmacologic treatment implications In this general overview, it has been demonstrated that certain critical features are shared by cigarette smoking and classic forms of drug de- pendence, that nicotine is psychoactive and can function as a euphori- ant, and that nicotine serves as a reinforcer for humans and animals. Tobac- co use is appropriately categorized as a form of drug dependence in which nicotine is critical. Whereas sorne treatments of other forms of drug de- pendencc have been developed in which minimal attention is paid to the pharmacologic issues, the most prominent advances in drug abuse treat- ment have involved pharmacologic manipulations designed to counter the effects of the drug itself and/or withdrawal. For instance, detoxifica- tion strategies with alcohol-dcpcndent, sedative-dependent, and opioid- dependent persons have been developed to ease the discomfort of with- drawal; substitution of more controllable and less toxic forms of the behavior-controlling drug are used in both maintenance and longer term detoxification strategies; and drugs that block the dependence-producing effects of the abused drug may be given to patients in an effort to pre- vent relapse. With respect to the treatment of tobacco dependence, cloni- dine, nicotine gum, and mecamylamine, respectively, have been used, to varying degrees, in analogous fashion (7, other papers in this volume). Use of Nicotine Gum for Treatment ofTobacco Dependence Other papers will more thoroughly address the application of phar- macologic treatment and the implications that tobacco dependent per- sons are dependent on nicotine. I would just like to make a few points that directly follow from the work of my colleagues and myself regard- ing the use of nicotine gum for the treatment of tobacco dependence. One issue is the problem of dose: the preparation available in the United States (i.e., 2 mg) is one-half the dose that has been found to be equivalent, along several parameters, to an average cigarette; also in part because of its slow speed of action (pharmacokinetic and pharmacodynamic issues), nico- tine gum is a weak reinforcer that does not readily engender compliance with a therapeutic regimen. These facts, possibly combined with a fear of toxicity and the financial burden of taking adequate dose levels, has apparently led to undermedicating: undermedication in drug dependence treatment can lead to partial abstinence, followed by relapse. Another issue is the appropriate way to use the gum itself: smokeless tobacco users learn to avoid undesired effects and to maximize desired ones by appropriate- ly allowing the nicotine to be released from its vehicle (e.g., tobacco wad) into the saliva, and then for sufficient exposure of saliva in the mouth to permit absorption. The practice of many gum users to swallow as they chew may be counterproductive. In addition, typical patterns of chew- ing the gum may mitigate e4licacy; specifically, instructions to chew slowly may help to avoid acutely taking too much, but may also minimize the chances of the smoker attaining a satisfactory substitute. Finally, instruc- tions to chew a piece of gum "when you feel the urge to stnoke" may be "too little too late" for nicotine dependent persons (by analogy, oral metha- done treatment for intravenous heroin dependence would probably be of little efficacy if patients were given 5 mg doses of methadone to take when withdrawal symptoms began). These problems are not inherent to the current product formulation (Nicorette (R) ) itself: alternate instructions and further research into the appropriate use of the gum would seem likely ways to immediately im- prove its efficacy. As Dr. Hughes and his colleagues have shown, instruc- tions are no less important with regard to control of nicotine gum-taking behavior than any other form of medication self-adrninistration (11). Summary In summary, let me remind you ofwhere I began-discussing a form of drug dependence in which the effects on performance, mood and feel- ing state, and brain function are critical to the ability of the drug to con- trol the behavior of the user (i.e., to its abuse liability). As I noted, fac- tors such as price, availability, and relative social acceptability are also prominent in the actual spread and incidence of the dependence process, 29 28

aud consideration of botlr pharmacologic and nonphannacologic issues are important to the efficacious treatment. Originally, I was referring to cocaine, but as you now have heard, the same factors apply to tobacco. Tobacco and its patterns of use share many features with other dependence-producing drugs: development of dependence follows a process of dose graduation; tolerance and physical dependence may oc- cur; useful effects may be produced in users; the critical drug is discrimi- nated by animals and humans, and is a etrphoriant for humans; and nico- tine serves as a reinforcer for both animals and humans. The strength of' these findings is considerable: studies of tobacco use in humans show that the laboratory data are relevant to clinical issues; studies with animals and with other forms of rucotine show that key effects of tobacco are bio- logic in nature and are not just due to social and cultural factors unique to humans. A final word about the implications of the conclusion that tobacco use (both smoked and smokeless) may lead to nicotine dependence (1). Some might conclude that such an equation is cause for patients to give up hope of quitting. I would argue that, to the contrary, much has been learned about the treatment of drug dependence in recent years, and rnuch of this may be readily applied to the treatment of tobacco dependence; further- more, it is surely true that the known enemy is more easily overcome. This idea is not new and was much more eloquently expressed by Dr. John Dorsey in 1936 (19). If we accept the fact that the use of tobacco in its preparations is a form of drug addiction, even though a pleasant one, not af- fecting criminal statistics, we can more readily help our patient when lie finds that his problem has gotten out of hand. References 1. Connolly GN, Wirm DM, Hecht SS, Henningfield JE, Walker B, Hoffman Ill. The reemergence of smokeless tobacco. N Eng. J. Med. 1986; 314:1020-1027. 2. Domino EF. Neuropsychopharmacology of nicotine and tobacco smoking. Pp. 5-32 in WL Dunn (Ed.) Smoking Behavior: Motives and Ine-entives. New York: Holt, 1973. 3. Fischman MW, Schuster CR, Resnekov L, Shick JFE, Krasnegor NA, Fennell W, Freedman DX. Cardiovascular and subjective effects of intravenous cocaine administration in humans. Arch. Gen. Psychiat. 1976; 33:983-989. 4. Gritz ER. Smoking behavior and tobacco abuse. Pp. 91-158 in NK Mel- lo (Ed.) Advances in Substance Abuse. Greenwich, CT. JA( Press, 1980. 5. Grunberg NE, Bauwn A. Biological commonalities of stress and sub- stance abuse. In S. Shiffman and T Wills (Eds.) Coping Stress and Drugs. New York: Academic Press, in press 1986. 6. Henningfreld, JE. Behavioral Pharmacology of Cigarette Smoking. Pp. 131-210 in T Thompson, PB Dews, JE Barrett (Eds.) Advances in Behavioral Pharmacology, vol. 4, 1984a. 7. Henningfreld JE. Pharmacologic basis and treatment of cigarette smoking. J. Clin. Psychiat. 1984b; 45:24-34. 8. Henningfield JE, Goldberg SR. Nicotine as a reinforcer in human sub- jects and laboratory animals. Pharm., Biochem. and Behav. 1983; 19:989-992. 9. Henningfiel_d J_F,, Miyasato K, Jasinski DR. Cigarette smokers self- administer intravenous nicotine. Pharmacol., Biochem. and Behav. 1983; 19:887-890. 10. Henningfrcld JE, Miyasato K, Jasinski DR. Abuse liability and phar- macodynamic characteristics of intravenous and inhaled nicotine. J. Pharm. and Exper. Therapeu. 1985; 234:1-12. 11. Hughes JR, Hatsukami DK. Signs and symptoms of tobacco with- drawal. Arch. Gen. Psychiat. 1986; 43:289-294. 12. Hughes JR, Pickens RW, Spring W, Keenan RM. Instructions con- trol whether nicotine will serve as a reinforcer. J. Phatm. and Expee Therapeu. in press, 1986. 13. Jaffe JH. Drug addiction and drug abuse. Pp. 532-581 in AG Gil- man, LS Goodman, TW Rall, F. Murad (Eds.) Goodman and Gilman's Pharmacological Basis cf Therapeutics. New York: Macmillan, 1985. 14. Jasinski DR, Johnson RE, Henningfrcld JE. Abuse liability assessment in human subjects. Trends in Pharm. Sci. 1984; 5:196-200. 15. Kozlowski LT, Herman CP. The interaction of psycho-social and bi- ological determinants of tobacco use: more on the boundary model. J. Applied Soc. Psych. 1984; 14:244-256. 16. Pomerleau OF, Pomerleau CS. Ncuroregulators and the reinforce- ment of smoking: Towards a biobehavioral explanation. Neurosci, and Biobehav. Rev. 1984; 8. 17. Rosecrans JA, Meltzer LT. Central sites and mechanisms of action of nicotine. Neurosci. and Biobehav. Rev. 1981; 5:497-501 18. Russell MAH. Tobacco smoking and nicotine dependence. Pp. 1-46 in RJ Gibbons, Y Israel, H Kalant, RE Popham, W Schmidt, RG Smart (Eds.) Research Advances in Alcohol and Drug Problems. New York: Wiley, 1976. 19. Russell MAH, Jarvis MJ. T heoretical Background and Clinical Use of Nico- tine Chewing Gum. Pp. 110-130 in J Grabowski, SM Hall (Eds.) Na- tional Institute on Drug Abuse Research Monograph 53. Washing- ton, DC: US Government Printing Office, 1985. 30 31

The "Why" of Tobacco Dependence: Underlying Reinforcing Mechanisms in Nicotine Self-Administration Ovide F. Pomerleau, Ph.D. University of Michigan School of Medicine Ann Arbor, Michigan Tobacco is a dirty weed. I like it. It satisfies no normal need. I like it. It makes you thin, it makes you lean. It takes the hair right off your bean. It's the worst darn stuff I've ever seen. I like it. Graham Lee Heminger Penn State Froth, November, 1915. Introduction Pharmacological treatment using nicotine replacement is largely based on theories of smoking that stress physiological addiction to nicotine as a major component of the habit. While such formulations explain a num- ber of smoking-related phenomena, many cigarettes seem to be smoked in response to environmental stimuli rather than to nicotine-withdrawal symptoms. Recent research suggests that smoking can be used as a phar- macological "coping" response that promotes temporary improvements in performance or affect by altering the bioavailability of a variety of behaviorally-active neuroregulators. There is reason to believe that replacement of nicotine from smoking by using chewing gutn may deprive the smoker of certain reinforcing consequences, including facili- tation of coping and pleasurable effects. I will therefore identify sonie of the problematic aspects of the addiction model, provide supporting evidence for a biobchavioral formulation of smoking, and consider the implications for nicotine replacement strategies. I'reparation of this chapter was facilitated by aArantfrom United States National Cancer Institute (CA427_30) The nicotine addiction model Smoking has been characterized as negativcly-reinforced behavior-that is, as an escape/avoidance response to the aversive consequences of nico- tine withdrawal (19,41,46). What is postulated is a nicotine addiction cycle, consisting of a withdrawal syndrome that occurs periodically, presumably resulting from nicotine depletion. At some critical level, aversive symptorns-impaired concentration and psychomotor performance, along with increased anxiety, irritability, and craving for cigarettes- manifest themselves. By implication, pleasurable or other reinforcing aspects of the drug are discretionary or incidental, as can be seen in the widely-accepted definition of tobacco dependence in the Diagnostic and Statistical Manual [DSM-III] of the American Psychiatric Association (50). There is considerable empirical evidence for the nicotine addiction model (12,20), some of it from my own laboratory. For example, in a study contrasting heavy and light smokers (using plasma cotinine, a nicotine metabolite with a half-life of nearly 30 hours, as an indicator of usual nico- tine intake), we found a number of characteristic differences (34): a) heavy smokers took in more nicotine (using plasma nicotine as an indicator) from high or low nicotine research cigarettes (used to eliminate the con- founding effects of brand preference); b) heavy smokers showed greater heart rate and skin temperature tolerance for a given dose of nicotine taken in (i.e., they were less reactive physiologically than light smokers; c) heavy smokers demonstrated more severe nicotine withdrawal symptoms (par- ticularly craving) when they were deprived; and d) heavy smokers ex- hibited closer regulation of plasma nicotine (i.e., proportionally less vari- ation in plasma nicotine when smoking cigarettes of differing strengths). These findings suggest that amount of usual smoking and nicotine in- take are the result of differences in nicotine addiction, as predicted by theory. The addiction theory in its extreme form, articulated a few years ago by Stanley Schachter (46), holds that the sole function of smoking is to relieve the symptoms created by its absence. This theory implies that the probability of smoking is closely linked to the time since the last cigarette. Such a formulation has limitations as a comprehensive explanation of smoking. With the exception of the first cigarette of the day or of cigarettes smoked after an extended period of deprivation, much ad libitum smoking seems to be prompted by various environmental events (both internal and external stimuli) that are independent of the time since the last cigarette - for example, the termination of a meal or coffee-drinking (3,7), perfor- mance demands (56), and various dysphoric states-particularly anxie- ty (10). 33 32

There are no clear demonstrations of nicotine withdrawal symptoms prior to self-administration at inter-cigarette intervals that arc represen- tative of ordinary smoking (half an hour to an hour between cigarettes). Moreover, tobacco withdrawal symptoms seem to be idiosyncratic, vary- ing from smoker to smoker (15); symptoms arc relatively mild, even at their peak (49), compared with those produced by withdrawal from sub- stances such as barbiturates or ethanol. Even more disconcerting from the perspective of a simple addiction model is the recent observation that withdrawal symptoms in a novel environment are muted compared with those evidenced after the same length of abstinence from smoking in the usual environment (14). Furthermore, as several investigators have ob- served, smokers are able to undergo extended periods of deprivation under certain conditions without experiencing much discomfort (1,46). Final- ly, recent research has found that sensory blockade of the upper bronchi reduces smoking satisfaction and craving without affecting nicotine in- take or changing withdrawal status (40), lending support for a hedonic, pleasure-potentiating factor that involves inhalation of smoke through the upper airways and is independent of withdrawal relief. Nicotine-produced pain reduction and anxiety relief. Our recent research has attempted to differentiate those effects of smoking that con- stitute relief of nicotine withdrawal from those consequences that pro- vide independent reinforcement. We selected nicotine-produced pain reduction and anxiety relief as representative of the various behavioral effects of smoking, a choice supported by numerous investigations in- volving nicotine administration in both animals and humans (2,9). The choice of pain reduction and anxiety relief, known markers ofendogenous opioid activity, was also influenced by the demonstration of nicotine- stimulated beta-endorphin release in circulating plasma in my laborato- ry (33,48) and in that of others (28,61). In the research that I will describe, pain was induced using a cold pressor apparatus that involved the sub- ject's immersing his arm into freezing water; anxiety was induced by hav- ing subjects attempt to solve a puzzle (a six-letter anagram) that they had previously failed to complete. In the first experiments in the series (37), habitual smokers (smoking more than a pack a day for over five years) served as paid subjects. Each experiment involved a screening session followed by experimental con- ditions with and without nicotine administration. In the screening ses- sion for 1~xperinient I, a five-minute exposure to the freezing water was provided to allow subjects to define a criterion for cold-pain in subse- quent sessions. For Experiment II, the subjects were presented with an (unsolvable) anagram as a means of inducing anxiety in subsequent sessions; they were offered a$10 bonus if they solved the anagram in 60 seconds or a $5 bonus if they solved it in five minutes. Subjects were in- structed to smoke their usual-brand cigarette half an hour before each experimental session to ensure a standard state of minimal deprivation. In the session, cigarettes were smoked for five minutes and consisted of either usual-brand cigarettes with brand markings deleted (mean nico- tine content, 1.2 mg/cigarette) or zero-nicotine cigarettes. In Experiment I, after smoking the cigarette, the subject immersed his arm in freezing water for a maximum of five minutes; pain awareness and endurance thresholds were determined. In Experiment II, the subject was told pri- or to smoking a cigarette that in five minutes, they would have another opportunity to solve the anagram. The Spielberger State Anxiety Inven- tory was administered before and after completion of the cigarette. In Experiment I, the five subjects all showed increased pain awareness thresholds (i.e., reduced sensitivity to pain) after they smoked a usual- brand cigarette. (Means were 27.4 seconds for zero-nicotine versus 45.0 seconds for the usual cigarette; paired t-test, p < .05, one-tailed.) With respect to pain endurance, four of five subjects showed increased pain en- durance and a fifth subject kept his arm immersed for the five-minute limit in both sessions. (Means were 113.0 seconds for zero-nicotine ver- sus 143.4 for the usual cigarette.) In Experiment II, the five subjects showed a significantly greater decrease in anxiety on the Spielberger State Trait Inventory when they were allowed to smoke their usual nicotine- containing cigarette, as shown in Figure 1 (ANOVA anxiety [before and after smoking] by nicotine interaction, F[1,4] = 14.38, p < .02). These experiments left unanswered the question of whether pain and anxiety reduction could be attributed to smoking a usual brand cigarette or to nicotine itself. Accordingly, the procedure was repeated with some modifications, using an additional ten smokers (8). Instead of usual-brand cigarettes, subjects were exposed to the following conditions in succes- sive sessions: two high-nicotine research cigarettes five minutes apart (removing the contribution of brand preference); self-administered intra- nasal snuffing of fine tobacco over two five-minute periods (providing nicotine by a novel route of administration); two zero-nicotine cigarettes five minutes apart (controlling for smoke inhalation); and two sham- smoked cigarettes (smoking an unlit cigarette to control for psycholog- ical effects and motor behavior). As shown in Figure 2, the two nicotine conditions resulted in marked increases in plasma nicotine, whereas in the no-nicotine conditions, plasma nicotine decayed somewhat over the 55-minute session. Smoking high- nicotine cigarettes significantly increased the pain awareness threshold 34 35

Zero Nicotine CIGARETTE SMOKED Usual Nicotine Figure 1. Anxiety scores for five subjects before and after a usual-nicotine or a zero-nicotine cigarette. [From Pomerleau, Turk, and Fertig, 1984 (37).] compared with sham-smoking [F (1,8) = 66.28, p<.0001], as did ad- nlinistration of snuff [F (1,8) = 21.50, p < .002]; zero-nicotine cigarettes did not produce a significant increase in the pain awareness threshold. (Means were 24.2, 25.8, 14.2, and 11.7 seconds for the high-nicotine, snuff, zero-rncotine, and sham conditions respectively.) High-nicotine cigarettes also significantly increased the pain endurance threshold compared with sham-smoking [F (1,8) = 21.15, p < .002], as did the administration of snuff [F (1,8) = 17.09, p < .003]; but zero-nicotine cigarettes had no sig- nificant effect. (Means were 59.0, 56.9, 46.8, and 35.0 seconds for high- nicotine, snuff, zero-nicotine, and sham conditions respectively.) The possibility that rucotine administration caused relief ofwithdrawal that was unrecognized by the subject, however, could not be ruled out. To shed more light on this issue, we recruited 15 male ex-smokers (no cigarettes in the last year, as validated by the absence of detectable nico-e tine in plasma). In one session they snuffed a single dose of fine tobacco (about one-fourth the dose used for smokers) and in another they snuffed an inert substance of similar appearance. Following administration of tobacco snuff, pain endurance increased significantly. (Means were 68.3 seconds for snuff versus 54.2 seconds for placebo; ANOVA, F (1,13) = 7.65, p < .02.) Pain awareness likewise increased, though not significantly. (Means were 34.3 seconds for snuff versus 28.8 seconds for placebo.) The studies suggested that pain reduction was not the result of brand preference, of smoking ritual, or of inhalation of smoke as such, but was rather the result of nicotine intake. We concluded that pain (and anxiety) reduction following smoking was not simply a consequence of relief of withdrawal, since in our studies smokers were minimally deprived and exhibited no significant differences in withdrawal, comparing ad libitum smoking in the screening sessions with the interval just before the nico- tine/placebo administration. These conclusions were strengthened by the observation that ex-smokers-who are no longer pharmacologically ad- dicted to nicotine and do not report withdrawal symptoms -also exhibit- ed reduced pain in the nicotine condition. The issue is not fully settled, since in pilot studies, ex-stnokers took in somewhat larger doses of tobac- co snuff than never-smokers, suggesting that residual tolerance (and the- oretically, some minimal withdrawal effects) could still exist. Tests with human subjects who have never smoked may help to resolve the ques- tion by clitninating altogether the potential contribution of withdrawal effects. The above experiments and previous research with nicotine-naive animals, however, strongly support the existence of nicotine effects that are independent of relief of nicotine withdrawal. 36 37

4~1 SMOKE/ SNUFF r SMOKE/ COLD SNUFF PRESSOR A -. HIGH NICOTINE ~ ~ SNUFF A •••••••o ZERO NICOTINE I I I I I I I i I I 1 5 10 15 20 25 30 35 40 45 50 55 TIME IN MINUTES Figure 2. Plasma nicotine as a fiinction of time for the two nicotine and iio-nicotine conditions; means for ten subjects. [From Fertig, Pomerleau, and Sanders, 1986 (8).] A biobehavioral explanation of smoking A review of the pharmacology of inhaled tucotine makes clear that the drug exerts powerful effects. It acts initially upon cholinergic receptors, rnimicking the effects of acetylcholine at low doses but blocking trans- cnission after initial agonist activity at higher doses (55). It readily pene- trates the brain, where it has been shown to act upon central nicotinic cholinergic receptors; a biphasic response paralleling the peripheral pat- tern of activation superceded by blockade has been inferred (1,12). Nico- tine's central neuroregulatory effects have been reviewed at some length (35): nicotine increases rates of release and turnover of acetylcholine, cen- trally, and of the catecholamines, norepinephrine, epinephrine, and dopa- mine; it also alters and stimulates the release of a variety of neuromodula- tory peptides, including arginine vasopressin, growth hormone, prolactin, and cndogenous opioids. A number of the endogenous substances whose synthesis, release, and turnover in the central nervous system are affected by nicotine have been shown, independently of nicotine, to influence behavior and subjective state. Cholinergic mechanisms, for example, seem to play a role in learning and memory (5,57), alertness (23), and pain inhibition (11,22,31). Increased arousal is associated with nonadrenergic activity (38,39), and stimula- tion of central pathways is associated with improvements in selective at- tention, increased alertness, enhanced vigilance, and facilitation of rapid information processing (29,52). Several studies have identified dopaminer- gic neurons as a critical part of the brain mechanism by which reward occurs (58,59). Finally, endogenous opioids have been linked to pain reduction (11,53) and alleviation of anxiety (16,26); they may have a direct rewarding effect as well (54). Because nicotine alters the bioavailability of such substances, it is plau- sible that smokers learn to "use" the drug to regulate or finely tune the body's normal adaptive mechanisms. The congruence of the reported con- sequences of smoking and the psychological effects of the endogenous neuroregulators known to be stimulated by nicotine is striking. Table 1, based on a recently promulgated biobehavioral theory of smoking (35), suggests possible links between these reinforcing consequences of smok- ing and putative ncurorcgulatory mechanisms. In those situations in which the pre-smoking context is neutral or positive and does not involve depri- vation or aversive stimulation, the reinforcing consequences have been classified as positive reinforcers; alternatively, in those situations in which an aversive state or behavioral deficiency can be identified as part of the pre-smoking context, the reinforcing consequences have been classified as negative reinforcers. 39

Table 1 Reinforcement consequences of smoking and putative neuroregula- tory mechanisms. Positive reinforcement Pleasure/enhancement of pleasure Facilitation of task performance Improvement of memory Associated with increase in: dopamine norepinephrine beta-endorphin acetylcholine norepinephrine acetylcholine norepinephrine Negative reinforcement Reduction of anxiety and tension Antinociception Avoidance of weight gain Relief from nicotine withdrawal beta-endorphin acetylcholine beta-endorphin dopanune norepinephrine acetylcholine Source: Adapted from Pomerleau and Pomerleau, 1984 (35). Under this formulation, dysphoric states such as anxiety may prompt smoking because such distress has previously been alleviated by the anxio- lytic effects of nicotine-stimulated beta-endorphin release and/or choliner- gic activity. Similarly, work or performance demands may trigger smok- ing because sustained psychomotor performance and alertness have been enhanced in the past by increased cholinergic and/or noradrenergic ac- tivity. Evidence suggests that the prompting of smoking by these events is unrelated to the nicotine withdrawal cycle or the time since the last cigarette. The nurnber of stressors or challenges that might cue smok- ing independently of nicotine withdrawal is potentially very large, provid- ing an explanation for the thorough interweaving of the smoking habit into the fabric of daily life (35). The fact that nicotine does not produce dramatic intoxication or with- drawal may even add to its reinforcing value, in that the "benefits" of smoking may be obtained without disrupting ongoing activity. Moreover, from the perspective of learning theory (32), the immediacy of the con- sequences of smoking (one quarter of the nicotine inhaled in smoke crosses the blood/brain barrier in seven to ten seconds) and the sheer number of repetitions (a pack-a-day smoker obtains over 70,000 puffing reinforce- ments per year) help to explain the strength of the habit (20). ' Implications for nicotine replacement strategies The accumulated evidence that smoking is prompted by a large num- ber of interoceptive and exteroceptive stimuli, of which physiological withdrawal is only one, has a number of implications for nicotine replace- ment strategies in the treatment of smoking. Among them is that treat- ment will need to provide more than just relief from nicotine withdrawal; the smoker's tendency to respond to stress and challenge by smoking needs to be taken into account as well. Such considerations are supported by various observations: Hatsukami and her colleagues (14) found that nico- tine abstinence symptoms are more intense in the smoker's usual environ- ment, a context in which various adaptational demands have occurred. Hughes and his colleagues (18) observed that while nicotine chewing gum relieves some abstinence symptoms such as irritability, anxiety, and difficulty in concentrating, it does not restore function to the level of ad libitum smoking; in addition, other symptoms such as craving for cigarettes, eating, and insomnia are relatively unaffected by nicotine replacement. There has been speculation that the 4 mg nicotine gum might be a more satisfactory substitute, since it produces average plasma nico- tine levels comparable to those produced by smoking (25); however, a fairly high percentage of smokers are insensitive to the difference in nico- tine availability between the 2 mg and 4 mg nicotine gums (6). An additional difficulty is that smokers report minimal pleasure from chewing nicotine gum, possibly because the slower rate of absorption from the gum smooths out the sharp rise in nicotine reaching the brain (45). A number of researchers have come to believe that the fast rise-time ofnicotine from cigarette smoking, resulting in a trough-and-peak pat- tern of plasma nicotine, is critical for producing the reinforcements peculiar to smoking (1,2,42). Alternatively, Rose and his colleagues (40) contend that the nicotine "spike" is less important, at least as regards the produc- tion of short-term gratification from smoking, than irritation of the up- per airways provided by hot cigarette smoke. They speculate that smokers come to associate these immediate effects with nicotine's other reinforc- ing effects, and that these sensations are either intrinsically pleasurable or that they function as secondary positive reinforcers through condi- tioning. Further research is needed to resolve this important theoretical problem. Recognition of the concerns described above led to the suggestion some 41 40

years ago that replacement nicotine be administered in such a way as to mimic the sharp rise produced by inhalation of tobacco smoke. Though iucotine gum as currently formulated does not produce this "spike;" other modes have been suggested, including intranasal snuff (43) and a nasal nicotine solution (44). A serious limitation of such methods, however, is that while they may be more effective as substitutes for cigarette smok- ing, the enhancement of reinforcement value may be matched by an in- crease in dependency-producing potential. Another difficulty is that nico- tine is not without health risks of its own-particularly cardiovascular (4)-which means that the long-range goal of all interventions should still be discontinuation of tobacco products and nicotine. Accordingly, smokers need to be trained to function without resorting to the use of nicotine as a coping response. Support for the usefulness of behavioral retraining comes from recent outcome research showing that nicotine gum was not much better than placebo gum in the absence of behavioral in- tervention, whereas in the context of a habit-change program, nicotine gum was consistently more effective than placebo gurn (13,47). While tested only in a few case studies to date, behavioral substitutes for the calming or relaxing effects of smoking, such as deep muscle relax- ation (30) or aerobic exercise (27), show promise. In particular, physical exercise has powerful effects on neuroregulators, including the cateehola- mines and endogenous opioids (36), and a number of reports indicate favorable affective changes as a result of regular exercise (17,51). Adminis- tration of dietary precursors of neuroregulators (60) that constitute the reinforcement substrate for nicotine might also provide a method for stimulating selected effects (e.g., 24). On the whole, behavioral proce- dures and dietary supplementation are safe; they are also sustainable over extended periods of time. The potential of such techniques for allowing ex-smokers to copc with the demands of daily living without succumbing to the blandishments of nicotine has not been adequately tested. Conclusion An important implication of a biobehavioral conceptualization of smoking is the recognition of the multifactorial nature of the habit. From this perspective, it does not seem likely that any single category of in- tervention for smoking will be adequate in and of itself. An important contribution of the nicotine replacement concept may well be that it facili- tates and provides a rationale for the integration of pharmacological in- tervention with behavioral techniques. Moreover, it opens the door to the possibility of tailoring therapies to the needs of individual smokers; smokers selected for physiological addiction, for example, may be particularly responsive to replacement therapy, whereas smokers defi- cient in alternative coping strategies may profit from a greater emphasis on behavioral techniques (21). Looking to the future, a better understand- ing of underlying mechanisms, particularly the definition of the reinfarce- mcnt substrate for smoking, should promote the development of more sophisticated pharmacological and behavioral techniques than are cur- rently available. Taken together, these trends should foster the develop- ment of rational and more effective therapies for smoking. References 1. Ashton H, Stepney R. Smoking: Psychology and Pharmacoiogy. London: Tavistock Publications, 1982. 2. Battig K. The smoking habit and psychopharmacological effects of 3. nicotine. Activitas Nervosa Superior 1980; 22:274-288. Best JA, Hakstian AR. A situation specific model of smoking be- havior. Add. Beh. 1978; 3:79-82. 4. Clee MD, Clerk RA. Tbbacco smoking: The medical sequelae. Pp. 177-198 in DJK Balfour (Ed.) Nicotine and the tobacco smohinq habit. Ox- ford: Pergamon Press, 1984. 5. Davis KL, Yamamura HI. Cholinergic underactivity in human memory disorders. Life Sciences 1978; 23:1729-1734. 6. Ebert RV, McNabb ME, Snow SL. Effect of nicotine chewing gurn on plasma nicotine levels of cigarette sinokers. Clin. Pharm. and 7'her. 1984; 35:495-498. 7. Epstein L, Collins F. The measurement of situational influences of smoking. Add. Beh. 1977; 2:47-54. 8. Fertig J, Pomerleau OF, Sanders B. Nicotine-produced antinocicep- tion in minimally deprived smokers and ex-smokers. Add. Beh. 1986; 11:239-248. 9. Gilbert DG. Paradoxical tranquilizing and emotion-reducing effects of nicotine. Psychological Bulletin 1979; 86:643-661. 10. Golding J, Mangan GL. Arousing and de-arousing effects of cigarette smoking under conditions of stress and mild sensory isolation. Psy- chophysiology 1982; 19:449-456. 11. Green PG, Kitchen l. Antinociception, opioids, and the cholinergic system. Progress in Neurobio. 1986; 26:119-146. 12. Gritz ER. Smoking behavior and tobacco abuse. Pp. 127-158 in NK Mello (Ed.) Advances in Substance Abuse, Volume 1. Greenwich, CT: 1 AI Press, 1980. 13. 1 lall SM, Tunstall C, Rugg D, Jones RT, Benowitz N. Nicotine gum and behavioral treatment in smoking cessation. J. of Consulting and 42 43

Clin. Psych. 1985; 532:256-258. 14. Hatsukami DK, Hughes JR, Pickens RW. Characterization of tobacco withdrawal: Physiological and subjective effects. Pp. 56-67 in J Grabowski, S Hall (Eds.) Pharmacological Adjuncts in Smoking Cessation (NIDA Research Monograph 53). Rockville MD: National Institute on Drug Abuse, 1985. 15. Hatsukanu DK, Hughes JH, Pickens RW, Svikis D. Tobacco with- drawal symptoms: An experimental analysis. Psychopharmacology 1984; 84:2331-2336. 16. Hill RG. The status of naloxone in the identification of pain control mechanisms operated by endogenous opioids. Neurosci. Letters 1981; 2:217-222. 17. Hughes JR. Psychological effects of habitual aerobic exercise: A critical review. Prev. Med. 1984; 13:66-78. 18. Hughes JR, Hatsukami DK, Pickens RW, Krahn D, Malin S, Luck- nic A. Effect of nicotine on the tobacco withdrawal syndrome. Psy- chopharm. 1984; 83:82-87. 19. Jarvik M. Biological factors underlying the smoking habit. Pp. 122- 146 in M Jarvik, J Cullen, E Gritz, T Vogt, L West (Eds.) Research on Smoking Behavior (NIDA Research Monograph 17). Rockville, MD: National Institute on Drug Abuse, 1977. 20. Jarvik M. Biological influences on cigarette smoking. Pp. 7-45 in N Krasnegor (Ed.) Behavioral Aspects of Smoking (NIDA Research Mono- graph 26). Rockville, MD: National Institute on Drug Abuse, 1979. 21. Jarvik M, Schneider N. Degree of addiction and effectiveness of nico- tine gum therapy for smoking. Am. J. of Psychiat. 1984; 141:790-791. 22. Kaakaola S, Ahtee, L. Effect of muscarinic cholinergic drugs on morphine-induced catalepsy, antinociception and changes in brain dopamine metabolism. Psychopharmacology 1977; 52:7-15. 23. Kawamura M, Domino EF. Differential actions of m and n cholinergic agonists on the brainstem activating system. Internat. J. of Neuropharm. 1969; 8:105-115. 24. McCarty ME Nutritional support of central catecholaminergic tone may aid smoking withdrawal. Med. Hypotheses 1982; 8:95-102. 25. McNabb ME, Ebert RV, McCusker K. Plasma nicotine levels produced by chewing nicotine gum. J. Am. Med. Assoc. 1982; 248:865-868. 26. Millan M, Emrich H. Endorphinergic systems and the response to stress. Psychother. and Psychosomatics 1981; 36:43-56. 27. Morgan R, Gildincr M, Wright G. Smoking reduction in adults who take up exercise: A survey of a running club for adults.J. of the Cana- dian Assoc. for Health, Physical Ed., and Recreation 1976; 52:39-43. 28. Novack DH, Allcn-Rowlands CF. Pituitary-adrenal response to cigarette smoking (abstract). Psychosomatic Med. 1985; 47:78. 29. Olpe HR, Jones RSG, Steinmann MW. The locus coeruleus: Actions of psychoactive drugs. Experientia 1983; 34:242-249. 30. Pechacek T Specialized treatment for high anxious smokers. Paper presented at the Annual Meeting of the Association for the Advance- ment of Behavior Therapy, New York, December, 1976. 31. Pert A, Maxey G. Asymmetrical cross-tolerance between morphine and scopalamine induced antinociception in the primate: Differen= tial sites of action. Psychopharmacologia 1975; 44:139-145. 32. Pomerleau OF. Underlying mechanisms in substance abuse: Exam- ples from research on smoking. Add. Beh. 1981; 6:187-196. 33. Pomerleau OF, Fertig J, Seyler LE, Jaffe J. Neuroendocrine reactivi- ty to nicotine in smokers. Psychopharm. 1983; 83:61-67. 34. Pomerleau OF, Fertig J, Shanahan SO. Nicotine dependence in cigarette smoking: An empirically-based, multivariate model. Pharm., Biochem., and Beh. 1983; 19:291-299. 35. Pomerleau OF, Pomerleau CS. Ncuroregulators and the reinforce- ment of smoking: Towards a biobehavioral explanation. Neurosci. and Biobeh. Rev. 1984; 8:503-513. 36. Pomerleau OF, Scherzer HH, Grunberg NE, Pomerleau CS, Judge J, Fertig J, Burleson J. The effects of acute exercise on subsequent cigarette smoking. J. of Behav. Med. 1987 (in press). 37. Pomerleau OF, Turk D, Fertig J. The effects of cigarette smoking on pain and anxiety. Add. Beh. 1984; 9:265-271. 38. Redmond DE. New and old evidence for the involvement of a brain norepincphrine system in anxiety. Pp. 153-203 in WE Fann, I Kara- can, AD Pokorney, RL Williams (Eds.) Phenomenology and Treatment of Anxiety. New York: S.P. Medical and Scientific Publishers, 1979. 39. Robbins TW. Cortical noradrenaline, attention, and arousal. Psycho- logical Med. 1984; 14:13-21. 40. Rose JE, Tashkin DP, Ertle A, Zinser MC, Lafer R. Sensory block- ade of smoking satisfaction. Pharm., Biochem., and Beh. 1985; 23:289-293. 41. Russell MAH. Smoking problems: An overview. Pp. 13-33 in M Jar- vik, J Cullen, E Gritz, T Vogt, L West (Eds.) Research on Smoking Be- havior (NIDA Research Monograph 17). Rockville, MD: National In- stitute on Drug Abuse, 1977. 42. Russell MAH, Feyerabend C. Cigarette smoking: A dependence on high-nicotine boli. Drug Metabolism Rev. 1978; 8:29-57. 44 45

43. Russell MAH, Jarvis MJ, Devitt G, Feycrabend C. Nicotine intake by snuff users. Brit. Med. J. 1981; 283:814-817. 44. Russell MAH, Jarvis MJ, Feyerabend C, Ferno O. Nasal nicotine so- lution: A potential aid to giving up smoking? Brit. Med. J. 1983, 286:683-684. 45. Russell MAH, Raw M, Jarvis MJ. Clinical use of nicotine chewing- gum. Brit. Med. f 1980, 280:1599-1602. 46. Schachter S. Pharmacological and psychological determinants of smoking. Ann. of Int. Med. 1978; 88:104-114. 47. Schneider NG, Jarvik ME, Forsythe AB, Read LL, Elliott ML, Schweiger A. Nicotine gum in smoking cessation: A placebo- controlled double-blind trial. Add. Beh. 1983; 8:253-261. 48. Seyler LE, Pomerleau OF, Fertig J, Parker K, Hunt D. Pituitary hor- mone response to cigarette smoking. Pharm., Biochem., and Beh. 1986; 24:159-162. 49. Shiffman S, Jarvik M. Smoking withdrawal symptoms in two weeks of abstinence. Psychopharmacologia 1976; 50:35-39. 50. Spitzer RL (Ed.) Diagnostic and Statistical Manual (Third Edition). Washington DC: American Psychiatric Association, 1980. 51. Stern MJ, Cleary P. The National Exercise and Heart Disease Project: Long term psychosocial outcome. Arch. of Int. Med. 1982; 142:1093-1097. 52. Svenson TH, Engberg G. Effect of nicotine on single cell activity in the noradrenergic nucleus locus coeruleus. Acta Physiologica Scandinavi- ca 1980; 479:31-34. 53. Tripathi J, Martin B, Aceto M. Nicotine-induced antinociception in rats and mice: Correlation with nicotine brain levcls.JJ of Pharm. and Exper. '77zee 1982; 221:91-96. 54. van Rcc JM, de Wied D. Brain peptides and psychoactive drug ef- fects. Pp. 67-105 in Y Israel, F Glaser, H Kalant, R Popham, W Schmidt, R Smart (Eds.) Research Advances in Alcohol and Drug Problems. New York: Plenum Press, 1981. 55. Volle RL, Koelle GB. Ganglionic stimulating and blocking agents. Pp. 565-574 in LS Goodman, A Gilnian, (Eds.) The Pharmacological Basis of Therapeutics (Fifth Edition). New York: Macmillan, 1975. 56. Wesnes K, Warburton DM. Smoking, nicotine, and human perfor- niance. Pharmacological Therapeutics 1983; 21:189-208. 57. Wcsnes K, Revell A. The separate and combined effects of scopola- mine and nicotine on human information processing. Psychopharm. 1984; 84:5-1 l. 58. Wise RA. Brain neuronal systems mediating reward processes. Pp. 46 405-483 in JE Smith, JD Lane (Eds.) The Neurobiology of Opiate Re- ward Processes. New York: Elsevier Biomedical Press, 1983. 59. Wise RA, Bozarth MA. Action of drugs of abuse on brain reward systems: An update with specific attention to opiates. Pharrn., Biochem., and Beh. 1982; 17:239-243. 60. Wurtman RJ, Wurtman JJ. Nutrition and the Brain. New York: Raven Press, 1986. 61. Yao M, Narita M, Baba S, Kudo M, Kudo T, Oyama T Effects of cigarette smoking on endocrine function in man. Masui, 1985; 34:1105-1114. 47

Psychosocial Factors in Smoking and Quitting: Health Beliefs, Self-efficacy, and Stress Saul Shiffman, Ph.D. Assistant Professor of Psychology University of Pittsburgh Pittsburgh, Pennsylvania Introduction My charge on this panel is a tough one. This is a conference on the pharmacologic aspects of nicotine dependence and its pharmacological treatment, and my task is to represent the psychosocial and behavioral aspects of smoking. It's a bit like being the token Freudian at a Behaviorist convention. In considering the relationship between psychosocial and pharmaco- logical factors, it would be very easy to set up a false dichotomy and com- petition between them. It is ingrained in our Cartesian dualism to set up a contradiction between physiology and psychology. All of us who are involved in smoking research or treatment undoubtedly are accosted at cocktail parties for quick consultations, where the single most common question is, "Well, is it a physical dependence or just psychological?" That is not an answerable question. There is no contradiction between the two. It's the interaction between them that's important. It makes no sense to ad- dress the question of why people smoke and why they don't quit without mentioning rucotine-that would be like talking about why objects fall without mentioning gravity. Let me begin, then, by acknowledging the very important role of nico- tine dependence in cigarette smoking. I will note in passing also that de- pc:ndence itself has an important psychosocial dimension. Even in opi- ate addiction - the classical "physical" dependence-there's evidence that physiological factors alone are not enough to explain dependence. For example, it's clear that not everyone who repeatedly uses opiates becomes addicted (21,13). Although we don't generally focus on non-addicted smokers, it's clear that the same thing can be said of tobacco use-not all users are addicts (18). Only some are vulnerable to dependence, and the vulnerability depends on psychosocial factors, including the social context of the drug use (21). So, dependence and psychosocial factors are not diametrically opposed concepts; they work together. We should not fall into the trap of considering behavior divorced from pharmacology or pharmacology divorced from psychology. To have a chance at even a cursory review of psychosocial factors in smoking, I will have to restrict myself to just a few highlights. I'll con- centrate on two issues: why people continue smoking, and the role of stress in smoking and cessation. "Why do people smoke?" or "Why don't people stop smok- ing?" The key question in smoking behavior research is "Why do people smoke?" Rather than address this in terms of the psychology of smok- ing, about which we still have much to learn, let's address this question in terms of the psychology of quitting, about which we know even less. Let's turn the question on its head to ask: "Why don't people stop smok- ing?" I'll approach that question by focusing on the very earliest stage of cessation: the stage at which a person who has been a regular smoker considers quitting smoking and perhaps embarks on a cessation effort. Jim Prochaska and Carlo DiClemente (5) have called this the "contem- plation phase" of quitting. This is a phase that doesn't come to our at- tention much clinically: the person is still smoking, but contemplating quitting. Yet, obviously, this phase is crucial to cessation. Health beliefs. "Why don't people quit smoking?" It seems easy to understand why many people do stop smoking: they're concerned about their health. In our own clinic data, we find that although other reasons (expense, social pressure, and so on) are also cited, literally everyone cites health concerns among their top three reasons for quitting. So, it's easy to understand why people do quit. The task is to under- stand why so many don't quit. Do they not care about their health? They do, I think, but their health concerns are often not motivating enough to make the difference. Ignorance and rationalization weaken the motivat- ing power of health concerns. It is tempting to believe that the millenni- wn of health information has arrived - that all smokers know what they need to know about the health risks of smoking, and, as a consequence, are highly motivated to quit. "It ain't necessarily so." Let me summarize some data from several recent (1978-1980) polls concerning American smokers' knowledge of the health impact of smoking (10). The figures are discouraging: • Over one-third of smokers (37% of all smokers; 40% of heavy smokers) were unaware that smoking causes heart disease, and a majority were unaware that smoking is a major cause of heart disease. 48 49

• Although most (though not all) smokers knew that smoking was linked to lung cancer, halfdid not know that smoking causes most lung cancer. Smokers also underestimated the fatality oflung cancer. • Many women were unaware of the impact of smoking on preg- nancy. A majority of women smokers (54-58`%o) were unaware that smoking is associated with miscarriage and stillbirth, and only one-third were aware of its relationship with low birth weight. Lest one think that this ignorance is restricted to the un- educated, college students were even less well-informed on these matters than was the general population. • Over 40"/0 of Amcrican adults did not know that smoking would shorten the life expectancy of a person smoking in his or her 30s. (This is the basis for the "I'll quit later" defense which I'll discuss later.) Arnong those who acknowledged that smok- ing resulted in a loss of life expectancy, the average person esti- mated the loss of life to be 2-4 years, underestimating by four years. (In fact, smoking results in a loss of 6-8 years of life.) • A key facet of health information about smoking is its relative importance in the contexts of other risks. A recent Louis Harris poll of 1,254 Americans (6) asked respondents to rate the impor- tance of many disease-preventing behaviors, and compared their responses to those of a panel of health experts. Consider the ten top-ranked actions chosen by each group. The experts were unanimous-not smoking is the single best thing one can do for one's health. In the public's eye, however, non-smoking was at the bottom of the list, below practices of questionable value such as taking vitamins and minerals. In an earlier poll (10), a third of all smokers thought that air pollution, rather than smoking, was the major cause of lung cancer. The public may be aware that smoking is unhealthy, but the quality of this knowledge is poor. • I should add that in almost every category in almost all the polls, smokers showed less awareness of the health risks than did non- smokers, and heavy smokers less awareness than light smokers. Perhaps the "knowing" smokers are already ex-smokers. Or perhaps smokers' ignorance is part of the cognitive defense they use to rationalize continued smoking. In suni, smokers may not have the knowledge we assume they have. What does this mean for smoking cessation? The implication is that often peoplcs beliefs about smoking and health may riot he sufficient to motivate them to quit. Beliefs about smoking and health, if inaccurate or distort- ed, could actually provide a rationalization for continued smoking. Let me illustrate this with some older data collected by McKennel and Thomas (9) in England. In this survey, smokers were asked how much smoking was safe-how much one can smoke before incurring excess risk of heart disease, cancer, bronchitis, etc. Figure 1 shows that about one-third of the smokers thought one could smoke over a pack per day without incurring any excess risk for the noted diseases-and these data were collected before the days of low tar and nicotine cigarettes. You may be tempted to think that this is only a quaint bit of history, but more re- cent American data compiled by the U.S. Federal Trade Commission (10) show that 40% of current smokers thought that one could smoke up to a pack a day without incurring any excess risk. As you might expect, the respondents who actually smoked less than 20 cigarettes per day were especially likely (50`%) to believe that this level of consumption was harm- less. In other surveys, 36% of smokers asserted that smoking low tar and nicotine cigarettes carried no risk at all, and another 32% were unsure (10). These people are engaging in defensive rationalization. So long as their knowledge of the health threat is abstract, vague, and distorted, it will not motivate behavior change. Behavior change is motivated only when the person experiences a concrete personalized health threat. This process is illustrated dramatically by other data from McKennel and Thomas (9). Figure 2 shows the health reasons cited as motivating a cessation effort. Far more cessation is motivated by awareness of actual current, though minor, illnesses -colds, sore throats, etc. -than is motivated by fear of lung cancer and other major but distant, illnesses. Obviously, not everyone who has a cold is thereby motivated to quit smoking, nor is someone who thinks that smoking causes only colds likely to be motivated. But these minor illnesses have the virtue of bringing the threat home to the smoker. People are more likely to respond to immediate and personal threats than to abstract and distant ones. In sum, American smokers' health information is crucially inaccurate and incomplete, largely as a result of cognitive distortions and misrepresentations used to justify continued smoking. People often continue to smoke because they do not recognize or acknowledge how dangerous their smoking is to them. Let me digress briefly here to address one implication of these ideas for pharmacological smoking cessation interventions. Besides its specific effect on the smoker, Nicorette has a powerful side effect: with proper intervention, it can mobilize the medical system, especially physicians, to be active in counseling patients to quit. The physician, if trained and motivated, can then play an important role in patient education to cor- rect the patient's misconceptions about snroking and health. Perhaps more important, the physician can persotralize the health information. The health 50 51

Figure 1 BELIEFS ABOUT HEALTH RISK: THE NUMBER OF CIGARETTES PER DAY CONSIDERED SAFE BY ONE THIRD OF RESPONDENTS Smokers 0 Ex-Smokers Cough Short Breath Bronchitis Lung Cancer Colds Heart Disease Figure 2 HEALTH REASONS CITED FOR STOPPING SMOKING % of Respondents 60 Smokers-No Wish to Quit 0 Smokers-Wish to Quit 50  Smokers-Have Tried to Quit I  Ex-Smokers 40 10 30 20 wl- Feared Feared Lung Cancer Other 111ness After McKennel & Thomas (1967) After McKennel & Thomas (1967) Actual Minor 111ness 53 52

risks can be presented in connection with the patiew's smoking pattern, t{ic patient'.c blood pressure, the patient's family history, and so on. So the advent of pharmacological treatments may mobilize physicians to play a more active role in educating patients so that their knowledge is ade- quate to provide motivation for cessation. Self-efficacy. 13esides motivation, there's another factor that's terri- bly important in enabling cessation, and that is self-efficacy (1). "Self- efficacy" is a fancy word for specific self-confidence, our belief in our own ability to accomplish a very specific goal, such as quitting smoking. (Psy- chologists make their living inventing new words for old concepts.) Carlo DiClcmente and Jim Prochaska (4) and numerous other investigators have demonstrated how important self-efficacy is to smoking cessation. In their Figure 3 PREDICTION FROM SELF-EFFICACY SCORES OF ONEYEAR SMOKING STATUS FOR BASELINE CONTEMPLATORS Baseline Self-Efficacy data, confidence in quitting distinguished those who succeeded in quit- 90 ting from those who failed, those who tried to quit from those who did not, and those who gave up from those who remained interested in 80 quitting. In this study, people who were thinking about quitting ("contempla- 70 tors"), rated their efficacy or confidence in being able to succeed. One year later, they were re-surveyed to determine how they had changed their 60 smoking status. Figure 3 shows that one year later, some smokers are still in the same place-contemplating quitting. These "procrastinators" arc 50 people who, at the beginning of the year, were not feeling very confi- dent. Others, in contrast, have quit smoking by the end of the year. These "successes" are the people who started the year with the highest efficacy scores. A third group of respondents have regressed-a year later, they are no longer even contemplating quitting, they have become "inuno- tive" These "backsliders" had started the year with the lowest confidence Immotives Contemplator- Immotives of all the conternplators. As you can see, they were actually not very differ- After oiCiorneoro & Prochaska (1983) ccut in efficacy from initial "immotivcs"-those who were not interested in quitting smoking initially. An important barrier to quitting is a sense of incapacity, a lack of con- fidence in one's ability to succeed. Lou Penner, a social psychologist col- Icaguc, once gavc the following as the first law of social psychology: "No- body likes to look like an ass." We often don't attempt that which we don't believe we can accomplish. To mobilize smokers to quit, one has to give them confidence that they can succeed. That, by the way, can be another positive sidc-effcct of Nicorcttc-it can help to boost pcoplc's confidence in their ability to quit and make them willing to try. To sunrmarize, I've suggested two answers to the question, "Why don't people quit smoking?" First, people sometimes don't quit because their understanding of the hcalth risks is riot sufficient to motivate them to Contemplator- Contemplators Contemplator- Quitters 54 55

quit. Second, smokers don't attempt quitting until they feel somewhat confident that they can succeed. Neither health knowledge nor self- efficacy is a panacea, but given more mobilizing health information and a perceived self-efficacy, more smokers will quit. Stress and smoking. Let's turn now to the issue of stress and smok- ing, which Dr. Pomerleau has already introduced. Stress is inextricably intertwined with smoking -and with inability to quit smoking. I'll fo- cus here mostly on the beginning and end of the natural history of smok- ing, on smoking initiation and smoking cessation. Although studies of smoking initiation have focused on peer pressure as the key culprit, stress is also involved in smoking from its very be- ginnings. Tom Wills (19,20) has recently completed a study on seventh graders' initiation to smoking. The striking, and somewhat surprising, finding is that the likelihood that a seventh grader takes up smoking is related to the youngster's experience of major negative life events in the last year, minor everyday hassles in the last month, and to recent feel- ings of subjective distress. This holds especially true for girls, support- ing the idea that women are especially likely to smoke in response to stress. Stress is involved in smoking from its inception. Typically, we think of stress as a series of environmental challenges or threats. To complement this notion, one should also consider internal fac- tors that govern the individual's vulnerability to stress. Smoking also seems related to these factors. A study by Cherry and Kiernan (2) looked at smoking initiation in relation to Neuroticism, which is a measure of emo- tional volatility or reactivity. They first measured Neuroticism when the respondents were 16 years old, and then followed them for 9 years. They found that youngsters who started smoking earliest were those with the highest Neuroticism score-those most emotionally volatile. In contrast, the respondents who had not started smoking by age 25-which essen- tially means they're likely to stay nonsmokers (7)-had significantly lower Neuroticism scores. Extrapolating, one can conclude that those who arc most stressed and most vulnerable to stress are most likely to take up ~ smoking. L•vl In adult smoking, stress reduction has long been considered a major ~ ~ N ~ 0® W motive for smoking. McKennel and Thomas (9) compared the relative strength of several stated motives for smoking. Two things stand out in their findings: 1) "nervous irritation smoking"-snioking when stressed - is the most strongly endorsed motivc for smoking; 2) among the vari- ous motives, stress relief is most closely related to the experience of craving to smoke. Now, one could argue that this is only because the most ad- dicted smokers experience tension (induced by withdrawal) when they're 56 not smoking, but the data may support a different conclusion: craving to smoke may arise in part out of the need to control stress. The connection between smoking for stress reduction and smoking cessation is even stronger. Figure 4 (9) contrasts several groups of smokers on two motives for smoking - stress reduction and social motives. The key comparison is between ex-smokers, who have succeeded in quitting, and those who are still smoking but say they are motivated to quit-i.c., those who have not quit despite a wish to do so. The two groups share about the same level of social motivation to smoke. What sets the failed quitters apart from the successes is a high likelihood of smoking in response to stress. In a similar vein, Ovide Pomerleau has also found that those who smoke in response to stress arc the poorest risks for main- tenance of nonsmoking (12). Whatever one's motives for smoking, the relationship between stress and smoking is so strong that the amount of stress a smoker reports when quitting smoking is a good predictor of his or her success for as much as a year later. Sue Curry and Alan Marlatt (3) assessed stress by focus- ing on everyday, small-scale stressors or "hassles" (8), like having a domes- tic spat or losing a set of keys, rather than major stressors like losing a loved one. Both the frequency and severity of these hassles at baseline predicted the person's success in quitting one year later. As I suggested in discussing stress and initiation, one should look not just at stress, but at the vulnerability of the individual who is being inr pacted by stress. An example of this approach is a study by Judy Ockene and her colleagues (11), which found not only that people under greater stress were more vulnerable to relapse, but also that smokers with lower "personal security" were more likely to relapse. Again, both external stres- sors and personal vulnerabilities are important. Another way to look at the role of stress in relapse is to look at its in- fluence on specific relapse episodes or relapse crises. A "relapse crisis" is a situation which threatens abstinence-it either leads to smoking or at least brings the ex-smoker to the brink of smoking. I've studied relapse crises by analyzing ex-smokers' reports to a relapse-prevention hotline (15,17). Fully 71 "/<, of the relapse crises occurred in the context of nega- tive affect. Table 1 shows the variety of feelings that people reported preceding relapse crises. (The presence of both anger and anxiety, on the one hand, and depression on the other, may reflect the bi-phasic or para- doxical effects of nicotine. That is, some people sought nicotine when they were over-aroused, others when they were under-arouscd.) Negative affect was not only a common antecedent of relapse crises; typically, it was the factor that actually triggered the episode. Its role was 57 i,;

Figure 4 SMOKING MOTIVES AND SMOKING STATUS % of Respondents with Motive 80 r 70 50 40 30 ~ Sedative Smoking 0 Social Smoking 60 All Current Smokers Motivated Smokers to Quit Source: McKennel & Thomas (1967) Ex-Smokers Table 1 The roles of stress in relapse: Affective antecedents and precipi- tants of relapse crises and stress-reduction as a motivator of relapse. Affect preceding Precipitant of Effect sought from episode % episode % smoking % Positive 28.8 Affect, stress 52.0 Stress reduction 43.0 Negative 71.2 Smoking stimuli 31.8 Reduced craving 20.0 Anxiety 41.7 Food or drink 20.9 Self-indulgence 12.0 Anger 25.8 Relaxation 10.8 Stimulation 10.0 Dcpression 21.5 Other 8.1 Enhanced positive experience 8.0 Other 16.0 Note: FrouT Shiffirnan (15, 16). Figures may add to more than 100% because of multiple responscs. specifically motivational, as well. When people undergoing relapse crises were asked what it was they sought when they were tempted to smoke- "What did you want to get out of it?"- a plurality, 43%, said they were seeking tension reduction (16). Surprisingly few people-only 20%- said that they were trying to satisfy or reduce craving, and those who did were especially likely to weather the crisis without smoking. In con- trast, those who said they were trying to achieve tension-reduction were especially likely to smoke. Relapse to smoking is motivated by tension reduction more than by any other motive. As Dr. Pomerleau pointed out, nicotine is used to control tension and not just to reduce craving. Coping. I emphasized earlier that smokers attempting cessation need to have a sense of their ability to successfully quit and stay off cigarettes. The data on relapse crises are clear: i.e., smokers can successfully fend off temptation. The telephone hotline study investigated different ways people coped with the temptation to smoke (15,17). A simple way to classify how people coped was to put coping into one of two categories: behavioral coping consisted of behaviors the person actually did-eating and keeping busy were typical examples. In cognitive coping, all the activity was mental- thinking about the health consequences of snioking or about how dis- appointed you would be if you smoked are typical examples.. The results were clear: used alone, either type of coping was effective in preventing relapse. A single coping response reduced the probability of relapse by 50%. Combinations of behavioral and cognitive coping were even more effective. It was the smokers who did nothing at all to cope- about one out of five-who were most likely to relapse. Relapse can be prevented, but it requires action. How does this relate to Nicorette and to pharmacological treatment? Dualistic thinking again tempts us to consider the relative virtues of phar- inacological vs. behavioral treatment. Another view is to think of nico- tine supplementation as simply one part-a powerful part-of the smok- er's armamentarium for dealing with the challenges of maintenance. Nicotine gum is not by itself a magic bullet. Many of the studies on its effectiveness have evaluated it in the context of a broader program that includes behavioral components such as coping skills. My colleagues at UCLA (14) specifically investigated nicotine's efficacy (vs. placebo) in the context of either an intensive behavioral program or in a dispensary con- dition where the gum, with instructions, was simply given to patients. Figure 5 shows the results. The key finding is that the effect of nicotine gum is most prominent when it is used in conjunction with a behavioral treatment. Now, perhaps the intensive treatment simply motivated people to use the gum more, but the results are at least suggestive. 58 : 59

Figure 5 EFFECTS OF NICORETTE & CLINIC TREATMENT ON SMOKING CESSATION % Abstinent 100.00] 75.00 -1 50.00 -1 25.001 0.00-i 30 Days 90 Days 180 Days LEGEND Nicotine & Clinic Placebo & Clinic N Nicotine & No Clinic  Placebo & No Clinic  Quitting smoking requires more than nicotine -it requires motivation, commitment, planning, effort, activity, and persistence. Nicotine and will- power are not enough. One tnight say: "Where there's a will, you still need a way." Health information can help motivate smokers to try to quit and pharmacological and behavioral treatments can enable smokers to quit, resulting in substantial numbers of healthier ex-smokers. Let's keep these treatment components working together. References 1. Bandura A. Self-efficacy: Toward a unifying theory of behavior change. Psych. Rev. 1977; 4:191-215. 2. Cherry N, Kiernan KE. A longitudinal study of smoking and per- sonality. Pp. 12-19 in RE Thornton (Ed.) Smoking behaviour: Physio- logical and psychological influences. New York: Churchill Livingstone, 1978. 3. Curry S, Marlatt GA, Gordon J. Stress and smoking cessation maintenance. Unpublished manuscript, University of Washington, 1985. 4. DiClemente CC, Prochaska JO. Self-efficacy and the stages of self-change of smoking. Paper presented at the annual meeting of the American Psychological Association, Los Angeles, August, 1981. 5. DiClemente CC, Prochaska JO. Processes and stages of change: Cop- ing and competence in smoking behavior change. Pp. 319-344 in S Shiffman, TA Wills (Eds.) Coping and substance use. New York: Aca- demic Press, 1985. 6. Harris poll, The Prevention Index: A report card on the nation's health. Em- maus, PA: Rodale Press, 1984. 7. Kandel DB, Logan JA. Patterns of drug use from adolescence to young adulthood: 1. Periods of risk for initiation, continued use, and dis- continuation. Am. J. Pub. H. 1984; 74:660-666. 8. Kanner AD, Coyne JC, Schaefer C, Lazarus RS. Comparison of two modes of stress measurement: Daily hassles and uplifts versus major life events. J. Beh. Med. 1981; 4:1-39. 9. MeKennel AC, Thomas RK. Adults' and adolescents'smoking habits and attitudes. London: British Ministry of Health, 1967. 10. Myers ML, Iscoe C Jennings C, Lenox W, Minsky E, Sacks A. Staff report on the cigarette advertising investigation. Washington, DC: Federal Trade Commission, 1981. 11. Ockene JK, Benfari RC, Nuttall RL, Hurwitz I, Ockcne IS. Relation- ship of psychosocial factors to smoking behavior change in an in- tervention program. Prca,~ Med. 1982; 11:1:13-28. 12. Pomerlcau 0, Adkins 1), Pcrtschuck M. Predictors of outcome and 60 61

recidivism in smoking ccssation trcatment. Add. Beh. 1978; 3:65-70. 13. Robins LN, HclzcrJE, Davis DH. Narcotic usc in Southeast Asia and afterward. Arch. Gc>n. Psy. 1975; 32:955-9fi1. 14. Schneider NG, Jarvik ME, Forsythe A13, Rcad LL, Elliott ML, Schwiger A. Nicotine gum in smoking cessation: A placebo- controlled, double-blind trial. Add. Beh. 1983; 8:253-261. 15. Shiffinan S. Relapse following smoking cessation: A situational anal- ysis. J Consult. and Clin. Psych. 1982; 50:71-86. 16. Shiffman S. Cognitive antecedents and sequelac of smoking relapse crises. J. Appl. Soc. Psych. 1984; 14:296-309. 17. Shiffinan S. Coping with temptations to smoke. Pp. 223-242 in S Shiffman, TA Wills (Eds.) Coping and substance use. New York: Aca- demic Press, 1985. 18. Shiffinan S. Tobacco "chippers": A study of individual differenccs in depen dence. Paper presented at the annual meeting of the American Psy- chological Association, Washington, DC, August, 1986. 19. Wills TA. Stress, coping and alcohol and tobacco use in early adoles- cencc. Pp. 67-94 in S Shiffinan and TA Wills (Eds.) Copin~ and suh- stance use. New York: Academic Press, 1985. 20. Wills TA. Stress and coping in early adolescence: Relationships to substance use in urban school samples. Unpublished manuscript, Albert Einstein College of Medicine, 1986. 21. Zinbrrg NE, Jacobson RC. Thc natural history of "chipping." Arn. J. Psychiat. 1976; 33:37-40. 62 Discussion: Smoking Behavior and Tobacco Dependence Chair: Murray Jarvik, M.D. heterans Administration Medical Center, Los Anqeles Presenters: Jack E. Henningfield, Ph.D. IVational Institute on Drug Abuse, Baltimoir• Ovide Pomerleau, Ph.D. University of Mich~qan School of Medicine, Ann Arbor Saul Shiffman, Ph.D. University of Pittsburgh, Pittsburqh DR. JARVIK: Can I have your attention, please. I would be interested if any of the members of the audience have any questions about what they've heard, and I'll try to field the questions as best I can. Yes? FLOOR: Could you comment on the use of lobeline for nicotine de- pendence? DR. JARVIK: Could I have your name, please? FLOOR: Sachs. DR. JARVIK: Thanks. Lobeline is a very interesting drug, because in- deed it is a relative of and an analog of nicotine, and it has nicotinic ef- fects. The man who did most of the research was a Swede who did half of his work in Sweden, half in New York. And there were some con- trolled studies in which placebos were used, but done by others, to see if lobcline in its various forms would be useful, and to date there has been no study that indicates that it is useful. I think it's a drug that is worthy of further study. It's obviously not identical to nicotine in its actions, but it certainly resembles it. It does seeni to be a cholinergic nicotinic agonist, and one of its biggest effects is on the carotid cheruoreceptors. It's been used to some slight extent in medicine. I don't think it's been subject to the well-controlled studies that we are now very sensitive to. On the other hand, it's been marketed for a long time in a variety of forms. Nicoban I think is one. You may be able to buy it at your drug stores still. And I don't know how popular this substance is. Usually when something works, the marketplace shows you that it works. 63

Drug addicts and smoking FLOOR: This is to Jack Henningfield. I'm wondering about the effect of using drug addicts in studies to assess the drug effects of nicotine. It seems to me a drug addict will like anything that's hitting him. Maybe other people don't respond quite the same way. I hope in the future you will look into other subjects. DR. HENNINGFIELD: We work with a variety of populations, rang- ing from heavy drug users, heavy smokers, to people who have never smoked and never abused a drug. And we use different populations to address different kinds of questions. The question of euphoria is not "Is nicotine always euphoriant?" but "Can it be a euphoriant under condi- tions in which drugs known to produce dependence are euphoriants?" So that's the use of that sort of population for that sort of study. DR. JARVIK: Well, there's a related observation to that, which is perhaps somewhat of a mystery: why is it that other drug abusers, such as heroin addicts or alcoholics, tend to be such heavy smokers? The incidence of smoking in these groups is incredibly high. Is it a behavioral carry-over effect or is there some pharmacological process underlying this? Alcoholics and smoking DR. POMERLEAU: If I might comment briefly, that has been a theme that has been of some interest to several of us. As a matter of fact, in a very early collaboration with llr. Jaffe, I looked at the smoking patterns of non-alcoholics and alcoholics, and though we only accumulated a rela- tively small data base, there were definitely some characteristics that seemed to differentiate the two populations. The alcoholics were uniform- ly taking in higher or obtaining higher plasma nicotine levels. They're also quite responsive in other respects. And though we did not accumu- late a sufficiently large nurnber of patient versus non-patient compari- Son studies, which as you all know, are somewhat tricky when one takes into account the various factors that need to be controlled, such as age and state of health and so on, we did find one particular hormonal differ- ence that we thought was quite interesting. It looks as if, at this point, it will not be published so I share it with you for whatever it's worth. One hormonal system with the alcoholics was quite difficult to get a good baseline, and that was antidiuretic hormone, its older name. It's now called argenine vasopressin. The interesting thing about this is that we hydrat- ed our subjects considerably, had them drink half a liter of water. Specific gravity of the urine at that point was 1.01, which certainly meant that we should have been able to get a very low baseline of this hormone because it responds to a lack of fluid. We were not able to get the same low baselines that we were able to get successfully with the non-alcoholic subjects. That was one aspect of it. But the other interesting thing was that the response to nicotine in the alcoholic subjects was muted. The non-alcoholic subjects produced linear dose response relationship with respect to nicotine stimulation of arge- nine vasopressin. Now, those of you who haven't been following the neu- ropeptide and neurohormone game, you may be wondering what the relevance of this is to motives for smoking. Argenine vasopressin has been shown, independently of nicotine related studies, to serve as a neuromod- ulator for memory. Putting a large one and one together and ending up perhaps with a generous three, we wondered if the memory deficits that alcoholics have might in some way be mediated by some kind of defi- ciency in this antidiuretic hormone system. And, of course, ethanol is a powerful diuretic agent. And we wondered if this insufficiency might also account for the higher self-dosing. That is: some deficiency in a sys- tem that produces a desired effect and requiring, therefore, a larger dose to produce the effect. I offer it to you not because there's any real proof from our data, which we accumulated, I think, in a study with 15 alco- holic patients versus 15 non-alcoholic patients, but more because it represents a testable hypothesis. It's something that one could pursue and explore, I would hope with more resources than we had available, to get a sufficient sample and to address the kind of control issues that are re- quired. There are a lot of variables like this that can begin to be addressed in a very productive way, given the availability of new technologies for measuring biochemical substances in plasma. DR. SHIFFMAN: I would add that the issue is just not related to alco- holics, but rather there is very strong data concerning perfectly normal social drinkers. There's an acute effect that when people drink they're more likely to smoke. So that one shouldn't attend to just permanent organis- mic factors that may differentiate alcoholics, but rather to the interaction between the drugs. The evidence is overwhelming. There's a review by Matarazzo and colleague in "Psychological Bulletin" that is very convinc- ing. And if you prefer, you can go down to just about any bar and collect the data directly. There's clearly an acute link, even among social drinkers. One of the things I puzzle about is whether that might not be related to people liking to smoke after eating. That is: we know that af- ter drinking, people like to smoke. We know that after eating, people like to smoke. We don't generally consider eating to be a pharmacological in- tervention. But, of course, it is. One's blood chemistry and neurochetnistry is quite changed by eating. 64 1 65

DR. JARVIK: I've been quite interested in the postprandial cigarette nry- self. I might say something about your comment, that, of course, there is a link and thcre's almost a causal link between drinking and eating and drinking and smoking. Drinking seems to stimulate smoking. On the other hand, if you go to meetings of Alcoholics Anonymous, they're not drinking, but they sure are smoking. So, there may be several effects that arc at work. Smoking cessation and eating DR. POMERLEAU: The subject of interactions among substances of abuse or acquired drives and physiological drive reduction is of some con- siderable interest. And obviously, the interaction that is of tremendous concern to treatment of smoking is that between eating behavior and nico- tine. There is someone here in the audience whose work I have been fol- lowing for some time and I would like to invite Dr. Neil Grunbcrg to the microphone briefly to give a one or two minute summary of sonie of the recent findings having to do with the effects of nicotine on ener- gy regulation, on eating behavior, and on taste preferences specifically. I think it is an issue that has to do with the understanding of the effects of Nicorette, nicotine replacement strategies, and really is a large issue that has to do with treatment outcome. Neil, would you describe some of your recent findings, please. DR. GRUNBERG: Ovide [Pomerleau] is referring to work that we've been doing on both animals and humans on the effects of nicotine and smoking, basically motivated by the question: why do people who quit smoking gain weight? Why do they get fat? This is clearly a concern about smoking cessation, particularly among women. Our animal and human studies indicate that nicotine works in a dose-response relationship to decrease body weight. After cessation of habitual use of nicotine or chronic adnunistration in animals, there's a dose-response increase in body weight, and interestingly enough, relevant to the question at hand, it ap- pears that a major factor in this body weight gain is an increased con- sumption of sweet-tasting foods in particular. Most important is the sweet taste, but also carbohydrate content from glucose. This works in a num- ber of ways. With respect to the question of alcoholism, while we're cit- ing other people, there's an interesting paper by Jack Henningficld with (Jeorge I3igelow and IZoland Gritiiths, which cited a commonly report- ed clinic finding with alcoholics. In alcohol rehab programs, oftentimes alcoholics are given access to sweet-tasting foods to help relieve with- drawal. I have also spent some time interviewing cx-heroin addicts who also show a draiuatic increase in consumption of sweet foods. Some nutritionists have problems in methadone maintenance clinics when the dosage gets quite low-ex-heroin addicts literally fill their coffee cups one-half full of sugar. This is all sirnilar across nicotine, heroin, and al- cohol, and in addition we're pursuing work on the neuroendocrine mechanisms. That's it briefly. Nicotine decreases pain and anxiety? DR. GRUNBERG: But let me ask you two questions, Ovide. One is: you concluded that smoking decreases pain, yet your operationalization of pain, using the cold pressor test on the arm, of course, could allow for the alternative interpretation that it's simply a peripheral vasoconstric- tion as imposed by the nicotine. I was wondering if you have other data on pain, or would you comment on that. The second question is another one we've discussed. You also suggested strongly or concluded, whichever you prefer, that smoking, through nico- tine, probably reduces or decreases stress and anxiety. Let's take the anxi- ety. Yet you did not run nonsmoker controls. You did have a sentence I noticed at the end of your presentation which implied that you were confident that the relationship was a decrease in anxiety rather than the obvious counterargument of an increased withdrawal. But I was won- dering if you could comment on these two issues: the pain effect with the operationalized peripheral vasoconstriction as a possible mediator; and secondly, whether or not the anxiety actually was decreased or whether wc're still open on the paradox. DR. POMERLEAU: Those are very thoughtful questions and certainly concerns that we have had over the years of being involved in this research. With respect to the peripheral vasoconstriction, it has been an issue that we have examined to some extent. I'in not quite sure how to come down on it by virtue of the mode of pain induction. There are several things that I think add up to supporting the conclusions I have drawn, though they are not fully satisfactory. So we are beginning to replicate some of this work using ischemic pain, which will simply give us a completely different pain modality to take care of this. And I should point out, as another direct aspect of it, the fact that we get similar effects with anxi- ety reduction, we fcel, indicates that we are looking at, in a sense, be- havioral markers of either cholinergic activity known to produce pain reduction and anxiety reduction or beta endorphinergic or endogenous opioid activity having similar effects. With respect to the technical aspects of it, one could argue both sides of that coin. The fact that we are reducing the temperature of the skin by immersing the arm into very cold water, in effect, could be seen as 67 66

reducing the threshold for the pain stimulus. That is: we are inducing vasoconstriction, thus, lowering the temperature. The mechanism by which the pain is believed to operate is by a vasoconstrictive mechanism and, therefore, the antinociceptive effect of the nicotine would, presumably operating in this particular set-up, create a bias against showing pain reduction by virtue of this. On the other hand, as you correctly point out, it's not entirely clear using this kind of modality. And so, we're really not fully satisfied that this is as tight a demonstration as we would like, and therefore, we're using ischemic pain as an alternative approach. I should point out, however, that there is a considerable literature indepen- dent of ours that does show reduction of pain, for example, Stanley Schac- ter's work with aversive shock. And there is an animal literature that also shows nicotine administration producing pain reduction in animals, and at least the inference of anxiety reduction. So, taking the larger pattern, looking at the literature as a whole, I think that one can probably feel reasonably confident that this is tapping that common phenomenon. Nonsmoker control DR. POMERLEAU: Nonsmoker control, of course, is something that we've been discussing for some years, Neil, and you've finally caught me in a public physicians' forum so that I now have to respond to it. My prin- cipal constraint has been the kinds of ethical considerations that are in- terpreted by human subject review committees in the United States. They are not terribly comfortable with the idea of administering nicotine, par- ticularly in the form of cigarettes, to people who have never smoked. And as a matter of fact, one of our real concerns in dealing with ex-smokers is that we might be subject to the accusation that we contributed to the resumption of smoking in this population. Therefore, we chose to ad- minister nicotine in a forrn with which these people had no previous his- tory. One of the controls that we used involved querying at a six-month ~ point, six months from th4 research, asking these people whether they ~ had resumed smoking or had any interaction with nicotine in any form. '~ We were relieved to find that in no cases was there a resumption of tobacco I~y use. ®G 13 ~d 1 h bf1 l1 b t t f asc on ia , we mrg t now egm to ce a rtt e rt morc com ort- able with the idea of approaching a human subject committee with a proposal to administer nicotine, though I doubt, given the current rules in this country, that it would be possible to use cigarettes in either ex- smokers or nonsmokers. And thus, we chose to administer snuff, which has a very sharp plasma nicotine rise time, rather than infuse nicotine by injection or sonie other modality. The reason for Neil's asking about the never-smoker, of course, is srm- ply the concern that these people are different- that smokers are differ- ent by virtue of exposure to nicotine. And I agree that that's a very inr portant question. It may well be, though, that it is the sort of question that Neil's own research is best set up to answer at the animal level. And so, I would turn the question back to Neil and ask him what his never- exposed-to-nicotine data looked like that would be any different than what we obtain from organisms that have a history of chronic exposure to nicotine. Are you seeing differences, Neil? DR. GRUNBERG: We don't have data relevant to that question right now. What about nonsmokers not smoking? DR SHIFFMAN: I'm not sure that the context of tl:is discussion is clear. There was some very nice work in the 1960s by Heimstra looking at things like performance in a driving simulatory task and anxiety under stress and so on. Now, if you just look at two conditions, smokers smok- ing and smokers not smoking, you would conclude, indeed, smoking is a stress reducer and a performance enhancer. That is: smokers smoking did better at driving simulatory tasks, and were less stressed. In his case he had a control group of nonsmokers not smoking. When you add them into the equation, it becomes clear that we were deceived by the first com- parison. The smokers smoking don't do any better than nonsmokers not smoking. The effect is due to the fact that smokers when not smoking show what seem to be withdrawal effects and a decrement in performance and an increase in stress. The point is: to interpret the data you really need a nonsmoker's not smoking control. DR. POMERLEAU: The problem that I should identify, and there is a context in the literature, is that we were not able to show any increases in anxiety as a result of very brief deprivation intervals in the smokers. And the ex-smokers, by definition, were not undergoing any pharmaco- logical withdrawal. We could measure plasma nicotine levels to cor- roborate the fact that they had not had any cigarettes for an extended peri- od of time. Therefore, it did not seem to us reasonable to conclude that the reduction of nicotine withdrawal explained the phenomenon. It still leaves open the question that Dr. Grunberg is raising, which is that by virtue of chronic exposure to nicotine, smokers are different in some way. The one direct point of evidence that I would introduce are some studies by Wesnes and Warburton in which nonsmoker controls were used and compared with smokers smoking. The performance data there on a driv- ing simulator seem to indicate that the smokers smoking were at an ad- vantage with respect particularly to being able to maintain a high level of alertness-they were at an advantage over nonsmokers not smoking. 69 68

Now, we're looking at relatively subtle effects, I should point out. These arc not dramatic major changes, and, thus, I think the difficulty of mak- ing an adequate demonstration. And the Heimstra study, as Dr. Shiff- man points out, does not reach that same conclusion, and thus the con- troversy. That's what makes our lives interesting. Nicotine dependency DR. JARVIK: I think there was a question .... FLOOR: Do all smokers suffer nicotine dependency? DR. SHIFFMAN: I think there's substantial variability. One exciting area of research right now is measuring smoking dependence, and Dr. Fagerstrom, who's sitting about two chairs down from you, has devel- oped a very useful scale in that regard. But it's very clear and implicit in some of the data I showed that there arc people who seem to smoke for motives other than withdrawal relief or stress relief, who have little trouble quitting and suffer no withdrawal. Finally, there'.s a phenome- non which we have tended to ignore altogether. If you had never seen people smoking, if you were from Mars and only learned about it through the literature, you would have no idea that there are people who smoke a few cigarettes a day, maybe a few cigarettes a week, for long periods of time. In our research we systematically exclude them, but I think they're very important in terms of casting some light on the vast majority who don't seem to be able to do that. So it's clearly not universal and I don't think we understand what determines dependence at all. DR. JARVIK: I guess this is what is referred to as chipping. DR. SHIFFMAN: Yes. I wish I had data to report. I have a study go- ing on of those folks, and they do exist, they do inhale. We don't have enough data to be able to address what is special about them. DR. HENNINGFIELD: If I could just makCe an additional point, which I started out trying to make and Dr. Shiffman made much more cloquent- ly, I think there is often a mistake of trying to arbitrarily separate what's commonly called physiologic withdrawal or physiologic dependence and dependence in general or compulsive drug-seeking behavior. A lot of these performance effects, cognitive effects, and so forth, arc due to ac- tions of nicotinc at the receptor in the central nervous system. I think this is part of the dependence process. And again, if wc go back to the co- caine model, we see that if you ask people why they use cocaine, and if you look at some of the data, you find that they are not, by and large, using cocaine to avoid withdrawal, nor do a lot of people report that they're simply using cocaine to become stoned out of their minds. A lot of the people are using cocaine to enhance their performance. I think to understand the dependence process is to understand a lot of these factors. Vulnerability factors DR. HENNINGFIELD: It also brings up another issue that we came on very closely in the last part of this discussion, that is, vulnerability factors, which I think we're now realizing are as much a part of the tobacco dependence process as a part of the drug dependence process in general. For instance, are people who become smokers more likely to be people who benefit from smoking, as far as their performance, mood, and af- fect are concerned? And again, some of the data Dr. Shiffman showed suggest this is possible. Unfortunately, in the area of tobacco dependence, as in the area of drug dependence studies, in general, vulnerability fac- tors have not been carefully studied. This is an area of study that wc're pursuing under the direction of Dr. Jaffe right now, not only with tobacco, but with other forms of drug dependence the question is equally valid. DR. SHIFFMAN: I think if you look at the initiation process, in some way that's where you see it most clearly. There seems to be a very quick self-screening where many teenagers try smoking; a substantial propor- tion of them go on to smoke, but in some ways it would be a reasonable interpretation of the data to say that by the third or fourth or fifth cigarette, teenagers have sorted themselves into those who are destined to become smokers for a long term and those who aren't. We really have very little understanding of what that sclf-screening is about. It's quite plausible that some of that has to do with vulnerability, either physiologically or be- cause some are ttnder more stress than others. DR. HENNINGFIELD: And I would venture that that's part of the dependence process. Differences in smoking DR. JARVIK: Dr. Glassman wants to ask a question. DR. GLASSMAN: You know, there are a couple of remarkable things about smoking. It's both a stimulant and a sedative or it reduces dysphoric feelings and perhaps is also a euphoriant, and also that the individual can so exquisitely modify the rate at which he takes the drug in. I wonder if you look within a smoker and look at either the total level of nicotine that he takes or the rate that he reaches that level, would you see differ- ences depending upon what he wants to do with his smoking. Has any- body really looked at whether somebody who wants to reduce tension smokes in a different way than somebody who wants to improve 70 1 71

performance or memory or so on? Can you distinguish that in a given person, using the person as their own control with conditions or goals? DR. JARVIK: Which one of you wants to take that? DR. POMERLEAU: I guess it's been a headache that I've assumed for myself, so I can respond to it. I can't say there's really a satisfactory data base to answer the question that you pose. Some of us have accumulated subdata that are suggestive, but hardly represent proof. In the literature and in my own work we've been able to induce stress, for example, us- ing mental arithmetic. Subtracting 13s serially is aversive as a procedure for most people. And subjecting people to this weve been able to demon- strate very clearly an increase in smoking, comparing it with a completely neutral activity, such as reading a magazine or watching television. However, the next step is the critical one, and that is to demonstrate that there is resulting differential dosing from that kind of precipitation of smoking or an increase in probability of smoking or nicotine self- administration. The argument would be that the smoker has the ability to fine-tune the dose by virtue of nicotine's very rapid absorption. What needs to be done is to demonstrate, for example, that under conditions of performance demand, which involves purely a need to increase alert- ness but has no anxiety or aversive component, one would get a small rise in self-administered plasma nicotine. That should be compared with a highly aversive situation which does not involve any performance de- mand or alertness-enhancing needs, but is strictly aversive. In this situa- tion, smoking reduces anxiety exclusively and one would find a much larger nicotine dosage being administered so that one then is able to get the biphasic eflect- the initial alerting followed by the calming or anxiety- reducing effect. The experiment that you were calling for clearly has not been done. As I describe it in this way, one can see some of the problems that are at hand. There are both psychological design problems- definition of a stimulus- and there is also inadequacy of the technology to measure the various complex of responses that would be engendered, which clearly, for an adequate demonstration, would involve more than simply the measurement of plasma nicotine over time. But I think it's something that necds to be done, and I'm glad that you've pointed out the issue in such a clear way. \C Tobacco withdrawal M-+ DR. JARVIK: How about Dr. Hughes over there? DR. HUGHES: Yes, I'd like to give a data base answer to the question about the prevalence of tobacco dependence and withdrawal. We recently completed a study in which we used DSM-111 criteria for tobacco dependence and Dr. Fagerstrorris criteria for dependence in a population based sample of middle-aged men. We found that using the DSM-111 criteria we had 90 percent behavioral dependence, if you will. And us- ing Dr. Fagerstrom's criteria, 36 percent. So, I'm afraid the answer is no answer. It depends on how you define it. Using the DSM-111 criteria, we asked these people: if you had quit for at least 24 hours, did you have certain withdrawal symptoms? Using DSM-111 criteria, we found that 21 percent had tobacco withdrawal and at some point, therefore, were physically dependent by that definition. And if we used our own criter- ia that we've developed, which is a little broader, we found that 46 per- cent of the population of middle-aged men had withdrawal when they quit. So, I'm afraid that at this point the simple question ofwhat propor- tion of people are physically or behaviorally dependent on tobacco real- ly can't be answered until we do more work on clarifying our definitions. Doctor-patient relationship DR. JARVIK: Let me go round clockwise to the three microphones. CALVIN FERMAN: Dr. Calvin Ferman from Baltimore, Maryland. I'm a practicing lung specialist who got involved in this issue of trying to help people quit. Not that I want to put myself out of business, but thought that I could get started. I have some very practical concerns as a person who sees people with cancer, heart disease, and so forth. Dr. Shiffman, I think you can help me. I can predict who's going to be suc- cessful in a $240 smoking cessation program of a referred base; these are people sent to me by other physicians. I'd like to know, is anyone look- ing at this ethereal funny thing called a doctor-patient relationship in terms of its power or strength in helping people to get off the weed? DR. SHIFFMAN: I will comment on that very briefly, making it clear that I'm commenting not from a base of hard data, but rather from this general conclusion that people are more motivated by immediate per- sonalized information. I think the physician is in a special position to per- sonalize the information-that is: the physician saying smoking is bad for you is not all that effective. I think the physician can personalize the information. If the patient has hypertension or a family history of hyper- tension, if the patient has some reduction in pulmonary capacity, et cetera, hooking the information onto that person's specifics will be a way to en- hance the power of the counseling. That's my sense of where we need to go. Let me get Judy Ockene to talk from data. DR. OCKENE: Actually I don't have very much data on the study we're now doing, which is looking at residents in family practice and internal medicine and their impact on patients' smoking behavior. But we did find 72 1 73

ill sonlc preliminary studies that perhaps about 40 percent of patients reported that physicians never advised them to stop smoking. Now, this, of course, is somewhat biased data because it's the patient's report. But we looked further at which factors affected whether or not patients were advised to stop smoking. We found that, as if the patient had no disease and had smoking-rclatcd symptoms such as coughing or shortness of breath, these had very little effect on whether or not they were advised to stop. But the factor that seemed to have the most impact was whether or not the patient was already diseased. So, I think the first step seems to be whether or not physicians are actually practicing primary prevention-at least by their patients' rcport-with regard to smoking intervention. We next surveyed all of the physicians at our medical school and in the community to find out what type of smoking intervention they did with their patients and what factors they felt affected their intervention with these patients. Most physicians are reporting that yes, they do intervene with their patients, but they're also reporting that they do not feel very confident in their ability to do this well. They report that this belief affects their intervention with patients. So, again we get to this idea of self-effi- cacy, but self-efficacy now from the physicians' perspective. And that is: if he or she doesn't believe that the intervention will have an impact on pa- ticnts, then there's very little effort made to intervene with patients. Right now, our present study is looking at whether or not training physicians to intervene with patients has an effect on the smoking outcome. So, we've trained physicians to counsel patients and we also have them just providing advice to patients, and we're looking at the effects of each of these. The very promising result right now-and we don't have any outcomes for patients, but we do for physicians-is that physicians are very educable in learning how to counsel their patients. And it doesn't take a very in- volved program to train them to counsel patients. Our program is about two hours, and physicians' increase in skills is very great from baseline to completion of the program. I think this is very promising, and it's cer- tainly promising in the context of using nicotine gum in a behavioral pro- grain because I think what many of us are saying is that use of nicotine gum by itself is perhaps not that effective, but that if used within a be- havioral context, that it could be effective. We're hoping to show this in our present study. Nicotine dependence DR. JARVIK: Mr. Pinncy says that we will have to break for lunch early. However, I think the four people who are standing around the microphones certainly should have a chance to ask their questions. And let me take it in a counterclockwise direction. Keep your questions short, please, and keep your answers short. Dr. Fagerstrdm. DR. FAGERSTROM: Thank you. I would like to make some comments on nicotine dependence. To me it's absolutely clear that there is a con- tinuum from very strong dependence to no dependence to pharmaco- nicotine at all, similar to what may be found with ethanol. Most of us here use ethanol and may be termed to some degree dependent, but not very dependent. But even among the smokers who may be classified as dependent and motivated to smoke for the nicotine, I think it might be valuable to distinguish between those dependent smokers who are nega- tively reinforced for smoking and those who are positively reinforced. By negative reinforcement I mean those who have to smoke to get rid of their withdrawal, which is the classical definition of dependence. But then we have smokers who don't smoke to get rid of withdrawal symp- toms, but more to get something they value, for example, a better ability to concentrate and alter emotional states, like euphoria. And in my clin- ical experience in Sweden where, as here, more women are taking up smoking than are men, I find that a larger proportion of women seem to be motivated for a positive reinforcement. They are not good candi- dates for nicotine substitution; they may find it difficult to take, to tolerate. But nevertheless, in some circurnstances or, let me say, in some emotional states they are very, very dependent on the nicotine. I think to recognize this difference in negative and positive reinforcement might have some implications for treatment. Craving DR. JARVIK: Thank you. Dr. Rose. DR. ROSE: I have a question for Saul Shiffinan, although the others may want to comment, too. Would you feel comfortable defining craving as the perception of the immediate reinforcing consequences of smoking? And, if so, how can you separate the generic craving from specific im- mediate consequences, like tension reduction and improved concentra- tion and so forth? DR. SHIFFMAN: I'm going to have trouble keeping my answer short. I think craving is very complex. I wouldn't feel comfortable with that definition. For example, the complexities include the fact already men- tioned that nicotine gum does not seem to have a substantial impact on craving, which is a puzzle. Jack also presented some data that smoking behavior changes even when craving hasn't been affected. In addition, people who smoked in a relapse crisis are very unlikely to have cited 74 1 75

t craving as a motivation. To my mind this reflects the very complex na- ture of craving. We may need more than one word for it. On the one hand it has an implication ofproprioceptive cues, physiologically driven mo- tives and impulses. On the other hand, I think there's a very big cogni- tive overlay. My interpretation of the data I just cited from my own studies is that in order for a person to label their experience as craving, they have to have a sense of deprivation. That is, if you're tempted to smoke and you go ahead and smoke, you don't say: gee, I craved it. Craving is a label for the experience of not smoking while you're tempted. Very similarly, there's a study done with opiate addicts which shows that craving rises when people know cognitively that the opiate is available and tends to drop when they're aware that it's not available. So, again, I think there's a very big cognitive overlay, and I'd be reluctant to accept a simple defi- nition of craving. I think that research on craving which helps to disen- tangle these aspects would be a very exciting prospect. Relapse DR. JARVIK: Mr. Pinney has told me that the meeting is over, but I'm going to take the last three speakers. DR. VERIGO PUENTE: I would like to raise the question of the doctor-patient relationship. I would say that it is not just a matter of doctor-patient relationship; it is also the therapist-patient relationship be- cause frequently it is our psychologists, not physicians, who work out the problems. For instance, in Mexico we started a quit smoking pro- gram with chronic cardiorespiratory patients with very poor socioeco- nomic status and very low academic level. We found that when the rela- tionship is good and when we established the right setting with the right tools, people do respond to the program. For instance, even with people who do not know how to read and to write, we used self-monitoring charts with drawings and colors, and we found that they were able to keep the monitoring system with not too many problems. Now, I would like to raise a point about minor stressful events as a cause to relapse. I do agree, but I think that we must also consider minor events which produce depression and anxiety in people who smoke could work as a precipitant to produce the smoking again. From our data in Mexico with psychiatric standardized interview we have found an im- portant proportion of people who have depression and do have anxiety, not only stress. Of course, it's very difficult to draw a clear-cut line where, you know, it's stress or depressed mood or when it's depression or anxi- ety. But we have found in our data that depression and anxiety are very common problems. DR. JARVIK: Dr. Shiffman, I think that was addressed to you. DR. SHIFFMAN: I would have to concur with you on your last point, that, first of all, depression is very important as an affective antecedent of relapse, and one we haven't looked at very much. And secondly, that it's useful sometimes to think of a person's smoking as self-medication for a substantial clinical problem with depression and anxiety. I think those of us who do treatment are aware that some of the patients experience an unmasking of an underlying, relatively serious problem. I don't think they're the majority, but I do think that that's an important issue. DR. JARVIK: I think you're next. Ireatment issues DR. KRISTELLER: I'm a psychologist at University of Massachusetts Medical Center. This question is addressed initially to Dr. Henningfield and then to the whole panel. There were a lot of parallels made between the use of other addictive substances and nicotine and tobacco smoking. What I'm curious about, partly for clinical and partly for theoretical rea- sons, is any evidence that bears on the treatment of smoking in multi- drug users. That's become an issue for me partly clinically because in- creasing numbers of cardiologists refer patients who are alcohol abusers, but their acute medical problem is related to their tobacco smoking. It's also of interest because of the very high number of alcoholics who are smokers. I think we're recognizing that as a medical area that's going to need to be addressed. DR. HENNINGFIELD: Your question raises many more questions than could be answered, but it gets to this whole issue of commonalities of people, vulnerability factors. It's recommended when people quit smok- ing that they avoid high-relapse situations, such as drinking, for instance. And we know that other drugs as well are setting stimuli for relapse to smoking. Here, again, we can do a little bit of a turnaround. To what degree in treatment of other kinds of drug dependence is smoking a relapse factor? Alternately, to what degree is nicotine a forni of medica- tion for those people? And I think these are some legitimate issues. It may be that there are different people in different kinds of functional relations. But clearly, this is a wide-open area. DR. JARVIK: Dr. Sachs. DR. SACHS: Yes. Actually, with all the discussion about physician in- tervention, I have to make a brief comment on the doctor-patient rela- tionship. My own belief-and I say belief because I do not have control data-is that the doctor-patient relationship is something added on to the therapist-patient relationship. The physician can certainly be 76 1 77

therapist, and I think a physician is far more effective when he or she is also a therapist. But I think even throughout our society, and if you go back to primitive cultures, there's something almost mystical about the medicine man-patient or the doctor-patient relationship. I think that when the physician in the white coat, who is known, almost in a very profound psychological sense, as having powers to snatch somebody from the jaws of death, says, "You should stop smoking; that carries added weight. I would pose this as a question that is worthy of systematic investigation, and you may, in fact, be doing that. The question that I have, although stimulated by Dr. Pomerleau, is ad- dressed to Dr. Henningfield. Dr. Pomerleau raised some interesting points about the nicotine spike issue and the fact that perhaps more effective pharmacologic substitutes would replace that spike. That may or may not be, but I think that raises another interesting question that I'd like Dr. Hcnningfield to respond to, and that is whether or not he thinks that the relative absence of the spike with nicotine gum therapy may be one of the factors behind the decreased abuse liability of nicotine gum in con- trast to the cigarettes that the patients are giving up? DR. HENNINGFIELD: I think that plausibly explains that and also explains some of the problems we have to overcome when using the gum in treatment, that is, how do we get the bolus effect? Is it possible to get a little bit more of a bolus effect than, for instance, people get by chew- ing slowly for 30 minutes, as they're told to do? The other thing is get- ting a high enough dose period. We know that, for instance, the 2 mg dose, which is available in this country, does not produce under usual conditions the kind of nicotine levels that people get from smoking cigarettes. So there are a number of possibilities here, one is alternate for- tnulations. Possibly though, we can do a lot with what we've got to work with just by alternate instructions. DR. JARVIK: O.K. Dr. Henningfield suggested to me that the speak- ers might like to make a 15-second summary of what they've said. That's a good challenge. DR. HENNINGFIELD: I think the important thing is that drug de- pendence is a confluence of psychological, pharmacological, and social kinds of factors. To understand tobacco dependence, it might be helpful to look at other kinds of dependencies, like cocaine dependence. DR. POMERLEAU: I'd like to emphasize the fact that the nicotine replacement model has focused largely on the nicotine withdrawal, the effects of smoking or nicotine dosing. I think that we need to pay more attention to nicotine as a coping response, particularly those aspects of the complex of effects which improve performance and affect. We will have to find both pharmacological substitutes, which may include nico- tine replacement, or other kinds of drugs, but ultimately for a drug-frec existence, nonpharmacologieal procedures which add to the behavioral repertoire of the smoker. DR. SHIFFMAN: I will use that as a jumping-off point. We have tended to focus on nicotine gtun as a specific agent and have thought of behavioral treatment as a nice supplement. As a devil's advocate I would propose that we think about coping training as the important therapeutic element and nicotine gum as one of an armamentarium of coping responses which the person can exercise. DR. JARVIK: Thank you very much. 9 78 1 79

III: Development of Nicorette: Its Uses and Limitations

Overview: Development of Nicorette, Its Uses and Limitations John R. Hughes, M.D. University of Vermont College of Medicine Burlington, Vermont Abstract In this section, Dr. Russell frrst outlines the rationale for the development of nicotine gum. Dr. Benowitz then describes pharmacological variables that must he consid- ered in the actions of nicotine. Dr. Sachs discusses the impact of nicotine gum on physicians. Finally, I discuss problems with nicotine gum. In this section overview, I suggest the efficacy of nicotinegum may change sever- al policies about the treatment of smoking. First, it may increase the efforts of phy- sicians and health care institutions to recognize and treat smoking. Second, it sug- gests smoking is a drug dependence. This may cause treatment and research efforts to focus more on dependence processes. Third, the gum is cost-effective. This may encourage remuneration for smoking treatment. Fourth, future studies of the use of nicotine f r nonsmoking disorders (e.g., aggression, dementia, pain, parkinsonism, sleep apnea, ulcerative colitis) may change the approved indications for the gum. Fifth, nicotine gum can produce behavioral and physical dependence; thus, legal policy classifying nicotinegum as a dependence-producing drug may need to be reevaluated. Introduction The four papers of this section illustrate the broad range of consider- ations that were necessary in the development of Nicorette. Dr. M.A.H. Russell is the foremost expert on nicotine dependence. Dr. Russell presents a brief overview of the empirical support for the nicotine dependence the- ory. He points out that advances in new areas such as CNS nicotine recep- tors may prove crucial to this theory. Then Dr. Russell clearly states the rationale for the use of nicotine gum. Finally, he goes on to discuss other routes of administration that could be used for therapeutic purposes. Dr. Preparation qf this article was supported by a grant (DA-04066) and a Research Scientist Development Award (DA-00109) from the National Institute on Drug Abuse. I thank Merrell Dow for permission to use their telephone survey data. Russell's article is another example of his ability to mix hard science, solid logic and clinical acumen into a very readable paper. Dr. Neal Benowitz probably knows more about the clinical pharma- cology of nicotine than the rest of us combined. Dr. Benowitz illustrates how pharmacological principles and data can impact the use and effica- cy of nicotine gum. He first presents a clear description of the principles of pharniacokinctics and what we know about the pharrnacokinetics of nicotine. Dr. Benowitz then goes on to illustrate the important need for dose response studies on nicotine gum. He then shows his own data, how there may be more flexibility in adjusting dose with the 4 mg gum, how smokers may be titrating their level of nicotine by swallowing nicotine, and finally how the lac_k_ of an effect of nicotine on heart rate may indi- cate that the gum is relatively safe. Dr. Benowitz does an excellent job of clearly describing pharmacokinetic principles to the uninitiated and lie shows that pharmacological data has clinical implications. Dr. David Sachs has a wealth of first-hand experience with nicotine gum not only in terms of patients but also physicians and adnvnistra- tors. Dr. Sachs indicates how nicotine gum will cause physicians to change their thinking about smoking cessation and how it can help reeducate physicians on the new evidence about why people smoke. He reviews crucial evidence that nicotine gum is effective in real world situations. I)r. Sachs' presentation is enthusiastic; however, he carefully documents the scientific evidence to support his enthusiasm. I review the probletns of nicotine gum. Contrary to clinical impres- sions, side effects from nicotine gum rarely limit its use; however, be- havioral and physical dependence on the gum can occur. I then empha- size that a major problem of nicotine gum originates in inappropriate use of it by physicians and patients, and difficulty in making the best selec- tions of patients, adjunctive treatments, doses, and durations of treatment. In the remainder of this paper, I will show that because nicotine gum is the first successful and generalizable pharmacological treatment of smoking, it introduces several policy questions about smoking treatment. I will also discuss how recent evidence about the gum may change its use. Impact ofnicotine gum on the recognition and treatment of smoking New medical treatments have an impact not only on health care poli- cies but also on financial, legal and research policies. This is especially true when the treatment is significantly more effective than existing tre.at- ments or brings treatment to a previously unserved population. 82 1 83

Several pharmacological treatments for smoking have been proposed and tested (see Dr. Russell's paper). Nicotine gum is the first successful pharmacological treatment (10,20,22,30). Perhaps more importantly, for some smokers, the gum is more acceptable than the more time-consuming behavioral therapies. As outlined in Dr. Sachs' paper, physicians fail to treat smokers for sever- al reasons (19). One of the most frequently cited reasons is that physi- cians don't believe smoking treatment is their responsibility nor do they believe they can effect smoking cessation. Dr. Sachs and others suggest that the availability of an effective medical treatment for smoking (i.e., prescribing nicotine gum) will increase physician involvement in smok- ing cessation. Although this makes sense, there are reasons why the gum might not increase physician involvement in smoking treatment. First, physicians might be discouraged by the relatively low success rate of the gum in medical practice (10%-20%). However, when success rates with the gum are compared with those for other chronic diseases, physicians may be less apt to belittle the gum's efficacy. Second, physicians often attribute success in smoking cessation to the patient and failure to themselves. When physicians are made aware of this no-win attitude, their attributions for success in smoking cessation may become more realistic. Third, the un- reliability of reimbursement for smoking cessation often deters physi- cians. Possible solutions to this problem are discussed later. And, fourth, even armed with the gum, many physicians feel inadequate to undertake smoking cessation "counseling" When physicians are shown that even brief protocols for cessation "advice" ( < 10 minutes) are effective (e.g., 32), their willingness and confidence in advising smokers increases. Nicotine gum may also impact the behavior of institutions toward the diagnosis and treatment of smoking. For example, several multi-physician clinics and HMOs have begun or expanded programs for smoking ces- sation now that an effective treatment is available. Finally, nicotine gum may impact medical training on smoking ces- sation. Students will need to be taught about nicotine gum in pharma- cology or drug abuse courses. This exposure may increase use of the gum in their practices. Impact of nicotine gum on future treatment research Dr. Russell argues in his paper that prior methods for smoking cessa- tion have been unsuccessful because they did not treat the dependence on nicotine that many smokers develop. The nicotine dependence the- ory has been endorsed by many health researchers (e.g., 11,28) and clearly labels smoking as a drug abuse disorder (1). Finally, the success of nico- tine gum is a clinically significant confirmation of the nicotine depen- dence theory. A central assumption in Dr. Russell's argument is that nicotine depen- dence is a factor in cessation failure for a large proportion of smokers. Although many laboratory studies have documented nicotine dependence among selected groups of smokers, the prevalance of such dependence among all smokers and whether it prevents cessation has only begun to be studied (13,16). Another phenomenon which may make Dr. Russell's argument more valid in the future is that a selection process may be oc- curring in which the less dependent smokers are now quitting and the more dependent smokers are continuing to smoke (16). Thus, future populations of smokers may include higher and higher proportions of dependent smokers. If this is true, then greater numbers of future smokers will require dependence-based treatments. The success of nicotine gum has also prompted several research pro- grams for other pharmacological treatments of smoking (see Glassman, Jarvis, Rose, this volume). Pharmacological treatments for dependence disorders as a rule have been less helpful than expected (10). In my paper in this section I suggest this is due not to the pharmacological charac- teristics of the drug, but rather to nonpharmacological factors suth as phy- sician and patient noncompliance and ignorance about how best to use the drug. Thus, 1 suggest we must address these problems before any pharmacological treatment can be successful. Impact of Nicotine Gum on Financial Reimbursement for Smoking Cessation Even with the most conservative estimates, nicotine gum is cost- effective (see Oster, this volume). At present, nicotine gum is paid for in several ways. In England, it has the distinction of being one of the few drugs not on the National Health Service formulary and thus patients must pay for it. In the U.S., 45% of patients pay for the gum (25), while others have their gum paid for by private or government (e.g., Medicaid) plans. Health maintenance organizations vary widely in their approach to payment; some have patients pay for the gum and then reimburse pa- tients if quitting is verified by biochetnical tests. An empirical policy about payment for nicotine gum can be formu- lated by comparing the cost-efficacy of the different treatment schemes. For example, if giving free gum increases the cessation rate and if the sav- ings in health care costs exceed the costs of the gum, then free gum would make economic sense. However, the premise of this argument may be 84 1 85

incorrect; i.e., one could hypothesize that having to pay for the gum is beneficial in that it eliminates the non-serious quitters and, thus, increases the success rate with nicotine gum. Our group is presently testing the effects of having to pay for nicotine gum on the use and efficacy of the gum. Impact of Nicotine Gum on Disorders Other Than Smoking Once a drug is on the market it is often used for indications other than those approved by the FDA. Although most pharmacological textbooks list no therapeutic indications for nicotine, several possibilities do exist. Nicotine gum has been tested as a treatment for musculoskeletal disorders (23), sleep apnea (9), and ulcerative colitis (27). Other studies suggest nico- tine might relieve aggression (6), dementia (35), anorexia (11), anxiety (4,14), pain (24), and parkinsonism (3). A major impediment to the treatment of such disorders with nicotine has been concerns that nonsmokers might become dependent on nico- tine gum. The lack of case reports of abuse of nicotine gum by non- smokers and laboratory evidence that nicotine gum is not a reinforcer in nonsmokers (33) tentatively suggest the gum has little abuse liability in nonsmokers. This issue will have to be carefully examined in future studies. Finally, whether other routes of nicotine administration (e.g., transdermal or aerosols, see Rose and Jarvis this volume) are better suit- ed for nonsmokers remains to be seen. Adverse Effects of Nicotine Gum At present, nicotine gum is not scheduled as a dependence-producing drug; i.e., there are no federal controls or monitoring of its use. Yet one risk of nicotine gum is behavioral dependence (15,20) and physical de- pendence (17,34). Although some believe the incidence of dependence is low, it is greater than that of narcotic analgesics and benzodiazepines (29). Thus, one public health policy may need to be reevaluated; i.e., the scheduling of nicotine gum as a dependence-producing substance. Some clinicians believe that in setting policy on this issue, the risk of dependence must be weighed against smoking relapse. They point out that chronic use of nicotine gum produces one-third to one-half the blood nicotine concentration from smoking. They also note that nicotine gum does not expose patients to the tar and carbon monoxide thought to cause the cancer and heart disease from smoking. Clinicians and researchers riot supportive of the gum point out problems with the use of other phar- macologic treatments for drug dependence (e.g., methadonc; 7) and 86 believe that physicians should not encourage dependence on any drug or substance. They also note that weaning from the gum does not always lead to relapse and that the long-term risk of the gum is unknown. Be- fore policy on the scheduling of nicotine can be set, studies in several areas are needed: e.g., the incidence of relapse back to smoking after forced ter- mination of the gum, and the strength and duration of dependence on nicotine gum. References 1. American Psychiatric Association. Pp. 176-178 in Diagnostic and Statistical Manual, Third Edition. Washington, DC: American Psy- chiatric Association, 1981. 2. Benowitz NL. Increased 24-hour energy expenditure in cigarette smokers. N. Eng. J. Med. 1986; 314:1640-1641. 3. Bharucha NE, Stokes L, Schoenberg BS, Ward C, Ince S, Nitt JG, Eldridge R, Calne DB, Mantek N, Duvoisin R. A case-control study of twin pairs discordant for Parkinson's disease: A search for environ- mental factors. Neurology 1986; 36:284-288. 4. Brodsky L. Can nicotine prevent panic attacks? Am. J. Psychiat. 1985; 142:524. 5. Blum A. Nicotine chewing gum and the medicalization ofsmoking. An. of Int. Med. 1984; 101:121-122. 6. Cherek DR. Effect of smoking different doses of nicotine on human aggressive behavior. Psychopharm. 1981; 75:339-345. 7. Dole VP, Nyswander ME. Methadone maintenance treatment: A ten- year perspective. J. Am. Med. Assoc. 1976; 235:2117-2119. 8. Fagerstrom K-O. Effect of nicotine chewing gum and follow-up ap- pointments in physician-based smoking cessation. Preventive Med. 1984; 13:517-527. 9. Gothe B, Strohl KP, Levin S, Cherniack NS. Nicotine: A different approach to treatment of obstructive sleep apnea. Chest. 1985; 87:11-17. 10. Grabowski J, Hall SM. Tobacco use, treatment strategies and phar- macological adjuncts: An overview. Pp. 1-14 in J Grabowski and SM Hall (Eds.) Pharmacological Adjuncts in Smoking Cessation, National In- stitute on Drug Abuse Monograph 53. Washington, DC: DHHS Pub. No. (ADM)85-1553, 1985. 11. Grunberg NE. Nicotine, cigarette smoking and body weight. Br. J. Add. 1985; 80:369-377. 12. Hollon S, Beck AT. Psychotherapy and drug therapy: Comparisons and conclusions. Pp. 437-490 in SL Garfield, AE Bergin Handbook 87

of Psychotherapy and Behavior Change: An Empirical Analysis. New York: John Wiley and Sons, 1978. 13. Hughes JR. Identification of the dependent smoker: Validity and clin- ical utility. Beh. Med. Abstracts 1984; 5:202-204. 14. Hughes JR. Nicotine gum to treat panic attacks? Am. J. Psychiat. 1986; 143:271. 15. Hughes, JR, Gust SW, Keenan S, Skoog KP, Pickens RW, Ramlet D, Healey M. Efficacy of nicotine gum in general practice. Annual meet- ing of the American Psychological Association, Washington, August, .1986. 16. Hughes JR, Gust SW, Pechacek T. Prevalence of tobacco dependence and withdrawal. Am. J. Psychiat., in press. 17. Hughes JR, Hatsukami D. Physical dependence on nicotine gum: A placebo-substitution trial. J. Am. Med. Assoc. 1986; 255:3277-3279. 18. Hughes JR, Hatsukami DK, Mitchell J, Dahlgren L. Prevalence of smoking among psychiatric outpatients. Am. J. Psychiat. 1986; 143:993-997. 19. Hughes JR, Kottke T. Doctors helping smokers: Real world tactics. Minnesota Medicine 1986; 69:245-250. 20. Hughes JR, Miller S. Nicotine gum to help stop smoking. J. Am. Med. Assoc. 1984; 252:2855-2858. 21. Jamorzik K, Vessey M, Fowler G, Wald N, Parker G, Van Vunakis H. Controlled trial of three different antismoking interventions in general practice. Brit. Med. J. 1984; 288:1499-1503. 22. Kozlowski LT. Pharmacological approaches to smoking modification. In JD Matarazzo, SM Weiss, JA Herd, NE Miller (Eds.) Behavioral Health: A Handbook of Health Enhancement and Disease Prevention. New York: John Wiley and Sons, 1984. 23. Lees AJ. Hemidystonia relieved by nicotine. Lancet 1984; 2:871. 24. Milgrom-Friedman J, Penman R, Meares R. A preliminary study on pain perception and tobacco smoking. Clin. & Exper. Pharm. and Phys- iol. 1983; 10:161-169. 25. Merrell I)ow Pharmaceuticals. Nicorette Long Term Users Survey. Cin- cinnatti: Merrell-Dow Pharmaceuticals, 1984. 26. Package insert, Nicorette, 1984. 27. Perrara DR, Janeway CM, Fcld A, Ylvisaker JT, Belic L. Smoking and ulcerative colitis. Brit. Med. J. 1984; 288:1533. 28. Pollin W. Why People Smoke Cigarettes. Washington, DC: U.S. Depart- ment of Hcalth and Human Services Pub. No. (PI-iS)83-50195, 1983. 29. Porter J, Jick H. Addiction rare in patients treated with narcotics. N. Eng. J. Med. 1980; 302:123. 30. Raw M. The treatment of cigarette dependence. Pp. 441-485 in Y Israel, FB Glaser, H Kalant, RE Popham, W Schmidt, RG Stnart (Eds.) Research Advances in Alcohol and Drug Problems, Volume 4. New York: Plenum, 1978. 31. Rounsaville BJ, Klerman GL, Weissman MW. Do psychotherapy and pharmacotherapy for depression conflict? Arch. Gen. Psychiat. 1981; 38:24-29. 32. Russell MAH, Merriman R, Stapleton J, Taylor W. Effect of nico- tine chewing gum as an adjunct to general practitioner's advice against smoking. Brit. Med. J. 1983; 287:1782-1785. 33. Strickler G, Hughes JR, King D. Nicotine as a reinforcer among never-smokers and ex-smokers. Presented at the Annual Meeting of the American Psychological Association, Washington, August, 1986. 34. West RW, Russell MAH. Effects of withdrawal from long-term nico- tine gum use. Psychol. Med. 15:891-893, 1985. 35, Whitehouse PJ, Martino AM, Antuono PG, Lowenstein PR, Coyle JT, Price DL, Kellar KJ. Nicotine acetylcholine binding sites in Alz- heimer's disease. Brain Research. 1986; 371:146-151. 88 1 89

Conceptual Framework for Nicotine Substitution M.A.H. Russell, M.R.C.P. Institute of Psychiatry London Introduction This large and well attended conference is a credit to John Pinney and his colleagues at the Harvard University Institute for the Study of Smok- ing Behavior and Policy who were responsible for it. But it must be es- pecially gratifying for Ove Ferno of AB Leo, in Helsingborg, Sweden, who I am pleased to see is in the audience today. It was he, with colleagues at the University of Lund, who first developed nicotine-containing chew- ing gum in the late 1960s. This product is now widely accepted as the first effective form of nicotine substitution therapy for smokers. It has also heightened awareness of the role of nicotine in smoking, stimulated research, and is beginning to motivate physicians to become more in- volved in intervention against smoking. Indeed, had it not been for Ove Ferno and the development of nicotine gum, it is doubtful whether this conference would have taken place or that we would be assembled here today. Trying to help dependent smokers has for many years been a rather thankless task. The availability of nicotine chewing gum, if used correctly, has turned it into an effective and rewarding one. It has also proved a useful tool for research. Its clinical efficacy, compared with placebo, in aiding cessation, alleviating withdrawal symptoms, inhibiting ad libitum smok- ing et cetera, has provided important new evidence for the role of nico- tine in smoking. In addition, nicotine gum has stimulated a new wave of clinical trials in smoking cessation research in which more rigour has been applied to important methodological issues such as the considera- tion of statistical power to avoid false negative results, biochemical vali- dation, and success criteria. Only recently, for example, has a clear Acknowledgements I thank the Medical Research Council for fundinA my research continuously since August, 1969, and Pamela Hancox for typing this manuscript. distinction been made between abstinence at one year follow-up and lapse-free abstinence throughout the year. The logic behind the development of nicotine substitution and other pharmacological treatments as aids to smoking cessation is clear to those who regard smoking as a form of drug dependence and who fully ap- preciate the crucial role of nicotine in generating and maintaining that dependence. This might seem clear to most researchers at this confer- ence, but it is not always understood by the public or even by all health professionals and other scientists. Indeed many people strongly oppose the notion that smoking is a form of drug dependence and that smokers need sympathy and more effective help rather than harassment and fur- ther restrictions_ This is o_n_e of t}ie reasons for holding this conference. It is to update ourselves on the role and effects of nicotine and to share experiences with a view toward developing better treatment methods to help smokers. Although the conference coincides with the publication of a NIDA Monograph on "Pharmacological Adjuncts in Smoking Ces- sation" (13), most of the chapters in that volume were written almost two years ago and are consequently somewhat outdated. There is now strong evidence showing that smoking is a form of drug dependence; the drug of course being nicotine. Measurements of the con- centrations of nicotine, and its metabolite cotinine, in the blood oi''smokers have demonstrated the extent to which smoking is a drug-taking activi- ty. Rapid absorption through the lungs enables the smoker to get an intravenous-like shot of nicotine after each inhaled puff. By varying puff rate, puff volume, and depth of inhalation, smokers regulate their nico- tine intake and have literally finger-tip control over the concentrations of nicotine in their brain. Nicotine has been shown to act as a primary reinforcer in animals and many of its pharmacological effects are poten- tially rewarding. It induces tolerance and smokers experience physical and psychological effects when it is withdrawn. There has been a recent surge of interest in the structure, function and regulation of nicotinic recep- tors in the brain. Nicotine induces changes in the number of nicotinic cholinergic receptors in the brain and this is one possible mechanism un- derlying tolerance. This is not the place to review all this material. I shall, however, outline some of the recent evidence which has in my opinion greatly clarified our understanding of the role of nicotine in smoking. I shall also indi- cate the extent to which smokers are addicted and need help with quit- ting, rather than more harassment to simply get on with it. Finally, I shall discuss why and how nicotine substitution might help them. The potential of other pharmacological approaches does not fall within my brief here. 90 1 91

Role of nicotine The role of nicotine in smoking is reviewed in this volume by Jack Hen- ningfield and Ovide Pomerleau, and the psychosocial and behavioral aspects are reviewed by Saul Shiffman. It is clear from these papers that people smoke cigarettes for many reasons -social, psychological, sen- sory, behavioral, and pharmacological. But of all these, the pharmaco- logical motives are, in my view, the most powerful and the most fun- damental. If tobacco contained no nicotine, there would be no problem. People wouldn't smoke it, nor would they snuff it or chew it. Although people begin to smoke for social and psychological reasons, pharmacological motives gradually take over as the smoker learns to in- hale and as a regular dependent smoking pattern becomes established. Other factors such as taste, smell, sensory irritation, and behavioral com- ponents such as handling also become important. This is mainly through frequent and close association with the pharmacological effects of nico- tine. In other words, they are secondary. Without the presence of nico- tine few people would develop a strong taste for tobacco. We know about the efficiency of the modern cigarette in delivering nicotine to the brain, about the rapid absorption of nicotine through the lungs after each inhaled puff and the few seconds it takes for the post- inhalation bolus to reach the brain. The implications of this for the phar- macological effects produced and for learning and conditioning do not need to be repeated. We know that in smoking doses nicotine has numerous pharmacolog- ical effects in the brain and throughout the body. It stimulates, it sedates, it does many other things. Besides its classical primary action on the nico- tinic receptors of acetylcholine, it affects the release of virtually every known neurotransmitter including noradrenaline, dopamine, 5HT and beta-endorphin. It also stimulates the release of hormones such as adrena- line, cortisol, vasopressin, growth hormone and prolactin, and it appears to enhance performance of various mental and psychomotor tasks. Many of these effects could be rewarding and reinforcing, but for many years we lacked the crucial evidence that nicotine can act as a primary rein- forcer. Unlike other addictive drugs, it had not been clearly shown that animals would self-administer nicotine at a higher rate than saline con- trol. Recently, however, Steve Goldberg and his colleagues have shown in a series of studies that several species of animal will self-inject nico- tine, responding at high rates comparable to those produced by cocaine (15,29). These studies demonstrate unequivocally and for the first time that nicotine can act as a primary reinforcer, and this greatly strengthens the evidence that people smoke for nicotine. Self-regulation of nicotine intake It is 15 years since Heather Ashton (1) and Chris Frith (5), working independently, first showed that smokers modify their puffing behavior in response to changes in the tar and nicotine yields of their cigarettes. Numerous studies have replicated and amplified these findings and there are several reviews on the issues of compensatory smoking, nicotine titra- tion, etc. (19). Although a number of direct studies have shown that the self-regulation of smoke intake is determined primarily by nicotine rather than other components in the smoke, it was unclear from these earlier studies whether or not the regulation of smoking behavior was controlled simply by the aversiveness of excessive nicotine. There was no evidence that people would smoke harder to obtain an optimal dose of nicotine (17). This crucial evidence has now been provided by Neal Benowitz (2). hl a Schachter-type study, he has shown that when nicotine excretion is enhanced by urinary acidification smokers increase their nicotine in- take by smoking harder but do not raise their blood nicotine to their usual smoking levels. This is the first study to show directly that self-regulation of nicotine intake is not controlled only by avoidance of the aversive barrier. In summary, there is now sound evidence that self-regulation of smoke intake is controlled by nicotine independently of other factois such as tar, CO or other gas phase components of the smoke. Down-regulation of nicotine intake to avoid a toxic aversive barrier is very precise and in this respect nicotine is unlike other addictive drugs. But up-regulation is less sensitive and is usually incomplete. This may in part be due to the aversiveness of intensive puffing and inhalation. On balance, smokers seem to tolerate a drop in blood nicotine to roughly two-thirds of their usual levels with some initial loss of satisfaction but they adapt gradually to the lower level over two or three weeks. Greater reductions occur when fuller up-regulation is mechanically difficult or impossible (e.g. on ultra- low yield cigarettes) and generate increasing craving and loss of satis- faction. However, other withdrawal symptoms only become prominent when blood nicotine levels fall below about one-third of the usual smok- ing levels. Thus nicotine chewing gum, which with normal ad libitum clinical use produces blood nicotine levels averaging about one-third of the usual smoking levels, alleviates many withdrawal symptoms but has less effect on reducing craving for cigarettes. Tolerance and withdrawal Tolerance and a specific withdrawal syndrome are inherent parts of most drug dependence processes. Drug tolerance, however, is not an essential 92 1 93

component of the dependent or addictive state, nor does its presence necessarily produce or imply dependence. It has been known for many years that nicotine induces tolerance of at least two types-an acute toler- ance which develops rapidly (within minutes) but disappears as soon as nicotine has been eliminated from the body, and a chronic tolerance which develops over several days and persists for at least three months after nico_ tine has been withdrawn. Table 1 Main Features of the Tobacco Withdrawal Syndrome Subjective effects Craving for tobacco Increased tension Restlessness Irritability Aggressiveness Depression Decreased alertness Difficulty with concentration Increased hunger Physical and objective effects EEG changes Drop in heart rate and blood pressure Increased peripheral circulation Drop in urinary adrenaline, noradrenaline and cortisol Sleep disturbances Increase in weight Decreased performance on: Vigilance tasks Simulated driving Selective attention tasks Memory tasks tFrom Russell, 1985 (18).] The existence of a clear-cut and specific tobacco withdrawal syndrome is now widely recogndsed (18); its main features are listed in Table 1. It is clear that in addition to subjective effects, tobacco withdrawal produces physiological changes and objectively measurable impairment ofperfor- mance on a number of psychomotor and mental tasks. However, apart from the uncontrolled intravenous nicotine studies of Johnston in the 1940s (11), there was no clear evidence that any of these withdrawal ef- fects arc related to nicotine. Now, over the past two years, placebo- controlled studies from three different centres have shown that nicotine gum alleviates some of the effects of tobacco withdrawal, providing clear- cut evidence for the first time that these effects are nicotine-related (7,27,30). Thc recent surge of interest in and research on the location, structure, fiInction and regulation of nicotinic receptors in the brain is already yield- ing results that have implications for increasing our understanding of the inechanisms underlying nicotine tolerance. It has been shown, for ex- ample, that nicotine induces changes in the number of nicotinic cholinergic receptors in the brain (28). Treatment of rats with nicotine for ten days increases the number of nicotinic cholinergic receptors in most areas of the brain (up-regulation) in contrast with cholinesterase inhibitors which reduce the "density" of these receptors (down-regulation). Furthermore, these changes are substantial: the ten-day nicotine treatment doubled the mimber of receptors. Longer treatment may well have induced further increases. There T~'^~^ ' =s no reason to -,xpect that human smokers would not show similar changes. A key question is how long after quitting it would take for the number of receptors to return to normal. Would the changes still be evident after the three-month period that chronic tolerance has been shown to persist? It is evident from all this that, as with other addictive drugs, when nico- tine is missing its effects are missed by the smoker (psychological depen- dence) and that the nervous system adapts and changes in response to nicotine so that it too functions differently when nicotine is missing (phar- macological dependence). Regular smokers and their ncrvous systems are different than they would have been had they never smoked and even more different when they stop smoking (18). Smoking as a form of drug dependence The discussion above illustrates that smoking is a drug-taking activi- ty and that nicotine has all the hallmarks of an addictive drug; nicotine has many pharmacological effects, it can act as a primary reinforcer, it induces tolerance, and psychological and physical effects occur when it is withdrawn. In this section I shall discuss how addictive smoking is and what proportion of smokers appear to be dependent on nicotine. In terms of its intractability, the tendency to relapse after short-term cessation, and the proportion of users who become dependent, cigarette smoking is on a par with other forms of drug addiction. Unlike alcohol use, smoking is seldom a take-it-or-leave-it activity practised in moder- ation on appropriate occasions. Over 80%) of people who smoke at all do so regularly on a daily basis (26). Just how easily and almost inevita- bly people become addicted if they smoke only a few cigarettes has been demonstrated by two major studies, one in the UK many years ago (12), the other in Ireland recently (14). They showed that of those teenagers ~ 94 1 95

who smoked more than one or two cigarettes, 80-85% went on to be- come regular smokers as adults. Surveys in several countries have shown that at least three out of four smokers report that they want to stop smoking or have tried to stop, yet only one in four men and one in three women who smoke succeed in stopping permanently before the age of 60 years (9). Thus most people smoke not because they want to but because they cannot easily stop. In other words, it is compulsive and they smoke because they are depen- dent. How powerful dependence on cigarettes can be is illustrated by a study in which about 70% of smokers who survived a heart attack relapsed to smoking within a year, despite advice and warnings against it from their physician (3). Even after undergoing surgery for lung cancer about 50% of smokers who survive resume the habit (4). No one needs to be reminded of the poor results obtained with numer- otis treatment methods over the past 20 years. Just as the dependency fac- tor blocks the response of most smokers to treatment, so it also blocks their capacity to respond more effectively to mass media campaigns, re- strictions and high taxation on tobacco. Apart from the difficulty in quit- ting, smokers also have problems with smoking fewer cigarettes and switching to lower tar cigarettes. Fewer than 20% seem able to switch to brands with machine-smoked yields below about 0.6 mg nicotine (10) and the minority who do switch appear to achieve this only by smok- ing more intensively to compensate for the lower yield, thereby under- inining much of the potential health benefits (25). In essence the terms "dependence" and "addiction" refer to a state in which the urge or need for something is so strong that the individual suffers or has great difficulty in doing without the drug (or object) and in extreme cases cannot stop using it when it is available. I use the terms interchangeably; although many reserve the term addiction for the more severe cases or for those in whom pharmacological factors predominate. However, the dependent state is not an all-or-none phenomenon. Smokers are dependent to varying degrees; how high a degree of dependence is required before the condition is labelled a "dependence disorder" or "ad- diction" is somewhat arbitrary. Likewise, the contrasting of"psycholog- ical" versus "physical" or "pharmacological" dependence creates a false dichotomy. Most addicts suffer from both to varying degrees, with the two factors inextricably interwoven (17). The blood nicotine levels of smokers vary widely, from below 5 ng/ml to over 70 ng/ml, with an average level for heavy smokers of about 35 ng/ml. The distribution of blood nicotine concentrations just after a cigarette is shown in Figure 1 for a sample of heavy smokers (22). 96 35 10 20 30 40 50 60 Plasma Nicotine (ng/ml) 70 90 t~!] 80 Figure 1. Distribution of peak plasma nicotine concentrations in a sample of393 heavy smokers (250 women, 143 men) with a mean cigarette con- sumption of 30 per day. Blood was taken two minutes after completing a cigarette during the afternoon of a day of usual smoking. The average plasma nicotine concentration was 35.8 ng/ml (SD 13.7) and was not sig- nificantly different between men and women. [From Russell and Jarvis, 1985 (22).] ) 97

Although the curve for smokers in the general population would be some- what to the left, measurable pharmacological effects are produced with blood nicotine levels of 10 ng/ml or less. It is thus apparent that most regular smokers inhale and absorb sufficient nicotinee to produce phar- macological effects. Table 2. Perceived Dependence According to Daily Cigarette Consumption in a Population of Normal Smokers Attending Their Family Physicians e 5 5-14 15-24 25 + Total Base 196 494 571 239 1500 (%) (%) (%) (%) (`%) Would find it "very difficult" to stop: for one week .4 8 6 4 0 altogether 3.0 24 49 68 38 Crave "frequently" 10 31 60 79 47 or "always" when unable to smoke JAbstracted from Russell, 1978 (16).] Most studies of withdrawal symptoms have been carried out in rela- tively small samples of heavy smokers. An indication of the incidence of craving and perceived dependence in smokers in the general popula- tion is gained from Table 2. In this study, almost 50% of smokers reported that they experienced craving "frequently" or "always" when they were unable to smoke, almost 40% believed it would be "very difficult" for them to give up smoking and 30% said they would find it "very difficult" to stop smoking for a week. As expected, the more heavily people smoked the more dependent they appeared to be. However, it is also clear that many lighter smokers experienced frequent craving and anticipated great difficulty with quitting, suggesting that they too might benefit from an effective aid to cessation. Altogether, only 11'% of the total sample reported that they "never" experienced craving and would find it "very easy" to quit permanently (16). This indicates that roughly nine out of ten smokers were to some degree dependent, and is similar to the proportion who absorb suffrcient nicotine to obtain pharmacological effects. Rationale for nicotine substitution The foregoing discussion has sought to establish two main points. First, smoking is highly addictive, so that only a minority of smokers will suc- ceed in stopping without assistance. This establishes the need for an ef- fective aid to cessation. Second, for the majority of smokers nicotine is an important component of their dependence. This suggests that replace- ment of the nicotine obtained from cigarettes with nicotine from an al- ternative source should facilitate the task of quitting by alleviating those aspects of the withdrawal syndrome that are nicotine-related. This should enable the smoker to tackle the problems of cessation in two stages. In the first stage, the smoker is able to focus on overcoming the behavioral and psychological components of dependence without at the same time having to cope with nicotine withdrawal. The dependence on nicotine can then be overcome at a later stage when the urge to smoke has declined. The capacity of a given nicotine substitute to relieve nicotine-related withdrawal symptoms is likely to depend on how closely it mimics the nicotine intake from cigarette smoking. The blood nicotine profiles of smokers vary widely (Figures 2 and 3) so that different substitutes might be best suited to different types of smokers. Some may require rapid ab- sorption, others may be satisfied with slower absorption. Less effective nicotine substitution might satisfy smokers who take in small amounts of nicotine, while smokers with high blood nicotine levels may require a more effective substitute capable of providing similarly high blood nico- tine levels. Another consideration is the dependence-producing potential of the substitute itself. The more closely it reproduces the smoking nicotine pro- file of a given smoker, the more effective it might be in assisting that smok- er to quit, but it is also possible that the smoker will then have greater difficulty giving up the substitute. The best approach might be to aim for a compromise by having a partial nicotine substitute which relieves the worst aspects of the withdrawal syndrome without providing total nicotine replacement. Indeed, nicotine chewing gum could be said to pro- vide just such a comprotnise. Further considerations are safety, social acceptability, and the capaci- ty of a substitute to provide, in addition to nicotine, substitution for some of the non-pharmacological components of the smoking habit. Nico- tine chewing gum, for example, also provides a nicotine taste as well as some form of oral substitution. The potential risks of nicotine substitu- tion have at times seemed an issue of somewhat hysterical overconcern. It is difficult to conceive how any practical form of nicotine substitution could even approach the harmfulness of smoking. But, that aside, as a 98 1 99

60 50 0 30 Z~ 0 20 0 m 10 Cigarette smoked 09 10 11 12 13 14 15 16 T T Time (hours) T T T T T T Cigarette (1.2 mg nicotine) Q v I I I I I I I i t 60 F 50 ~ S 40 c 0 30 z c0 20 ro 0 1 2 3 4 . . . . . . . . . . . . . (•=cigarette smoked) Time (hours) 5 . 6 7 . . Figure 3. Blood nicotine levels of a heavy smoker, smoking three cigarettes an hour. [From Russell and Feyerabend, 1978 (20).] Figure 2. Blood nicotine levels of an inhaling smoker, smoking one cigarette an hour. [From Russell et al, 1976 (2l).]

general principle it is obviously desirable that the blood nicotine levels should be kept as low, and the rate of absorption as slow, as is compati- ble with clinical efficacy in the interests of both safety from possible nicotine-related health hazards and the dependence-producing potential of the substitute itself. Finally, it is likely that more satisfaction and therefore greater therapeu- tic success might be gained from those forms of nicotine substitution which also provide a sensory stimulus and/or socially acceptable be- havioral component to act as substrates for conditioning as acquired secondary reinforcers. For example, the smell, the sharp taste, the local irritancy to the mouth, even the sight of the pack, the feel and the snap- ping sound made by pressing a piece of nicotine gum out of its package can become a source of satisfaction to the experienced user. Routes and vehicles for nicotine substitution There are many potential routes for administering nicotine. Besides the rate of absorption and other issues relating to bioavailability, the ther- apeutic potential of a particular route will also depend on the efficiency of the devices or vehicles of administration available, their acceptability to the clients and other factors discussed previously. A number ofroutes can be immediately excluded on practical grounds. In the case of ingestion, absorption is slow and most of the nicotine is metabolised to pharmacologically inert metabolites before gaining ac- cess to the general circulation. Nicotine suppositories and pessaries could no doubt provide a slow build-up of adequate blood nicotine levels, but would be inconvenient to use and would probably be unacceptable. In- jections would need to be repeated too frequently to be practical and would not be feasible for widespread use. There is also a risk that solutions in- tended for subcutaneous or intramuscular injection might be abused and used intravenously. Long-lasting subcutaneous mini-osmotic pumps would be safe but expen- sive. However, they could maintain adequate steady-state blood nicotine levels for long periods and would seem a possibility for the future. Trans- dermal delivery using skin patches has been widely used to administer coro- nary vasodilator drugs and also has potential in the case of nicotine. It is best suited for the slow build-up and maintenance of steady-state blood nicotine levels but would have more flexibility than mini-osmotic pumps for providing fluctuating levels under subject control. The exciting poten- tial of this approach is discussed by Jed Rose (this volume). It is my be- lief, however, that the lack of a sufficient sensory stimulus or sensori- motor ritual and the relatively slow absorption of nicotine will limit its capacity to do more than alleviate some withdrawal symptoms. It may prove less effective in reducing craving and providing any positive satis- faction. There remain the three routes-lungs, buccal and nasal mucosae- which have been used for nicotine absorption from tobacco for over 500 years. Due to their large surface area, the rate of absorption through the lungs is far and away the most rapid. Only by this route could the nico- tine intake from smoking be closely matched. While it is technically pos- sible to produce aerosols with particles small enough to reach the alveo- li, to my knowlege no satisfactory nicotine aerosol has yet been developed. We have tested four. They have been too clumsy or too irritant, or have simply failed to produce potentially useful blood nicotine concentrations. The potential of this route awaits further research. Some years ago a device was developed in the form of a vaporiser shaped like a cigarette (8). In- halation through this device enabled nicotine vapour to be taken into the lungs. It was not excessively irritating and was capable of producing ther- apeutically useful blood nicotine concentrations. The commercial product, Favor smoke-free cigarettes, is currently being market tested in the United States. But, unlike its prototype, the draw resistance is too great for a whole inspiration to be sucked through the device. It must therefore be puffed like a cigarette. We have not yet tested it, but it is my impression that the nicotine content in a 50 ml puff is rather low and that, unlike a puff of smoke, most of the nicotine in the vapour is deposited in the mouth and throat before reaching the lungs. It will be interesting to see what blood nicotine concentrations it can produce. The historical fact that tobacco has been chewed and taken as "wet" snuff in the mouth and "dry" snuff in the nose suggests that absorption of nicotine by either route is sufficient to produce some satisfaction. Blood nicotine concentrations from use of both "wet" and "dry" snuff build up to levels equivalent to those produced by cigarette smoking (6, 23) but absorption is slower and there are of course no sharp blood nicotine peaks and no post-inhalation high-nicotine boli. Various nicotine-containing lozenges and tablets have been produced from time to time, but they have never been systematically tested and in most cases their nicotine content has been inadequate for therapeutic value. Nicotine chewing gum has of course been extensively tested. The rate of nicotine absorption from the gum is relatively slow, but with repeated doses the 4 mg gum produces and maintains plasma nicotine levels comparable to heavy smoking. Lack of sharp peaks and post-inhalation boli limit its capacity to match the Positive pleasure and satisfaction of inhaled smoking so that its main ef- fect is to relieve nicotine-related withdrawal symptoms. 102 1 103

The blood nicotine profile of repeated gum use suggests that it would be a more effective nicotine substitute for the type of smoker shown in Figure 3 than for the one shown in Figure 2. The oral activity of chew_ ing is no doubt an additional help for many smokers. A major probletn with nicotine gum is its acceptability to smokers. Many reject it quickly and others require considerable encouragement and motivation to per- sist long enough to get used to it and come to like it. On the other hand, some smokers take to it immediately and find it very helpful. About 8% of heavy smokers transfer their dependence from cigarettes to the gurn. Little is known of the factors which determine its acceptability and ef- ficacy among different smokers. It does seem, however, that it is more acceptable and effective with heavier smokers and those with higher blood nicotine levels. They use more of it and it has more impact on enhanc- ing their success rate. It is not my purpose here to discuss at length the promises and problems of treatment with nicotine gum. This is covered elsewhere in this volume. It is the first form of nicotine substitution to have been applied and test- ed on a large scale. There is no doubt that it works and that its develop- ment has been a major advance in smoking cessation. Yet, it is a slow and relatively ineff cient form of nicotine substitution. During normal clinical use blood nicotine levels average about one-third to one-half of the levels obtained from smoking. The fact that it can work well despite its limitations is encouraging and suggests that better substitutes might give even better results. One such possibility is nasal nicotine solution for use as a kind of liquid nasal snufl: It appears that absorption of nicotine through the nose is much more rapid and efficient than through the buccal mucosa. A single drop of the solution containing 2 mg nicotine produces a peak plasma nico- tine concentration within ten minutes, which is about the time it takes to smoke a cigarette (24). This too has been developed by Ove Ferno and the research on it to date is described by Martin Jarvis (this volume). Fi- nally, a nasal nicotine spray, if developed, could prove a better device for nasal administration. Summary and conclusions 1) Recent research has increased our understanding of the role of nico- tine in smoking and the self-regulation of nicotine intake by smokers. It is in this context that pharmacological treatments, especially nicotine substitution, become logical and relevant. The rationale for nicotine sub- stitution and the various potential routes and vehicles for nicotine ad- ministration are discussed. 2) Smoking is a drug-taking activity, and it is highly addictive. The drug, nicotine, has numerous pharmacological effects and can act as a primary reinforcer. It induces tolerance, and with chronic doses there is an up-regulation in the number of nicotinic cholinergic receptors in many parts of the brain. Physical as well as psychological effects occur on with- drawal and most smokers are to some extent dependent on nicotine. 3) Although most smokers want to quit and try many times, only one in three succeed in stopping permanently before age 60. 4) Smokers need more help overcoming their dependence rather than further harassment to quit. 5) The development of nicotine chewing gum has been a major ad- vance and it is currently the most eiiective pharmacological aid to smoking cessation. Better forms of nicotine substitution such as nasal nicotine so- lution may produce even better results, possibly at the cost of generat- ing greater dependence on the substitute itself. Less efficient substitutes such as transdermal administration may have a role in weaning those who have become dependent on more efficient substitutes. 6) The prospects are bright for developing a range of nicotine substi- tutes to use separately or in combination as the clinical situation demands. With such aids, the smoker who is nicotine-dependent but wants to quit should be able to obtain the kinds of help needed to succeed. °. References 1. Ashton H, Watson DW. Puffing frequency and nicotine intake in cigarette smokers. Brit. Med. J. 1970; 3:679-681. 2. Benowitz NL, Jacob P. Nicotine renal excretion rate influences nico- tine intake during cigarette smoking.J. Pharm. and Exper. Ther. 1985; 234:153-155. 3. Burt A, Illingworth D, Shaw TRD, Thornely P, White P, Turner R. Stopping smoking after myocardial infarction. Lancet 1974; 1:304-306. 4. Davison G, Duffy M. Smoking habits of long-term survivors of sur- gery for lung cancer. Thorax 1982; 37:331-333. 5. Frith CD. The effect of varying the nicotine content of cigarettes on human smoking behaviour. Psychopharmacologia 1971; 19:188-192. 6. Gritz ER, Baier-Weiss V, Benowitz NL, Van Vunakis H, Jarvik ME. Plasma nicotine and cotirune concentrations in habitual smokeless tobacco users. Clin. Pharm. and 7her. 1981; 30:201-209. 7. Hughes JR, Hatsukami DK, Pickens RW, Krahn D, Malin S, Luknic A. Effect of nicotine on the tobacco withdrawal syndrome. Psychophar- macology 1984; 83:82-87. 104 1. 105

8. Jacobson NL, Jacobson AA, Ray JP. Non-combustible cigarette: Al- ternative method of nicotine delivery. Chest 1979; 76:355-356. 9. Jarvis MJ. Gender and smoking: I)o women really find it harder to give up? Brit. J. Add. 1984; 79:383-387. 10. Jarvis MJ, Russell MAH. Tar and nicotine yields of UK cigarettes 1972-1983: Sales-weighted estimates from non-industry sources. Brit. J. Add. 1985; 80:429-434. 11. Johnston LM. Tobacco smoking and nicotine. Lancet 1942; 2:742. 12. McKennell AC, Thomas RK. Adults' and Adolescents' Smoking Habits and Attitudes. Government Social Survey. HMSO, London, 1967. 13. National Institute on Drug Abuse. Research Monograph 53. J Grabowski, SM Hall (Eds.) Pharmacological Adjuncts in Smoking Ces- sation. Rockville, MD: Department of Health and Human Services, Public Health Office, DHHS Pub. No. (ADM)85-1333, 1985. 14. O'Connor J, Daly M. The Smoking Habit. Dublin, Ireland, Gill and MacMillan, Ltd., 1985. 15. Risner ME, Goldberg SR. A comparison of nicotine and cocaine self- administration in the dog: Fixed-ratio and progressive-ratio sched- ules of intravenous drug infusion. J. of Pharm. and Exper. Ther. 1983; 224:319-326. 16. Russell MAH. Smoking addiction: Some implications for cessation. Pp. 206-226 in JL Schwartz (Ed.) Progress in Smoking Cessation. New York: American Cancer Society, 1978. 17. Russell MAH. Tobacco dependence: Is nicotine rewarding or aver- sive? Pp. 100-122 in NA Krasnegor (Ed.) Cigarette Smoking as a De- pendence Process NIDA Research Monograph 23. Washington, DC: Department of Health, Education and Welfare, Public Health Serv- ice, DHEW Pub. No. (ADM)79-800, 1979. 18. Russell MAH. Smoking, nicotine dependence and its treatment with nicotine chewing gum. Pp. 7-42 in Excerpta Medica Asia Pacific Con- gress Series No. 39 Nicorette in Smoking Cessation. Princeton, NJ: Ex- cerpta Medica, 1985. 19. Russell MAH. Nicotine intake and its regulation by smokers. In WR Martin, GR Van Loon, ET Iwamoto, DL Davis (Eds.) Tobacco Smok- ing and Ilealth: A Neurobiologic Approach. Lexington, KY, in press. 20. Russell MAH, Feycrabcnd C. Cigarette smoking: A dependence on high-nicotine boli. Drug Metabolism Rev. 1978; 8:29-57. 21. Russell MAH, Feycrabend C, Cole PV Plasma nicotine levels after cigarette smoking and chewing nicotine gum. Brit. Med. J. 1976; 1:1043-1046. 22. Russell MAH, Jarvis MJ. Theoretical background and clinical use of 23. 24. 25. nicotine chewing gum. Pp. 110-130 m J Grabowski, SM Hall (Eds.) pharmacological Adjuncts in Smoking Cessation NIDA Monograph 53. Washington, DC: Department of Health and Human Services, Public Health Office, DHHS Pub. No. (ADM)85-1333, 1985. Russell MAH, Jarvis MJ, Devitt G, Feyerabend C. Nicotine intake by snuff users. Brit. Med. J. 1981; 283:814-817. Russell MAH, Jarvis MJ, Feyerabend C, Ferno O. Nasal nicotine so- lution: A potential aid to giving up smoking? Brit. Med. J. 1983; 286:683-684. Russell MAH, Jarvis MJ, Feyerabend C, Saloojee Y. Reduction of tar, nicotine and carbon monoxide intake in low tar smokers. J. Epid. & Comm. Health. 1986; 40:80-85. 26. Russell MAH, Wilson C, Taylor C, Baker CD. Smoking habits of 27. men and women. Brit. Med. J. 1980; 281:17-20. Schneider NG, Jarvik ME, Forsyth AB. Nicotine versus placebo gum in the alleviation of withdrawal during smoking cessation. Add. Beh. 1984; 9:149-156. 28. Schwartz RD, Kellar KJ. Nicotinic cholinergic receptor binding sites in the brain: Regulation in vivo. Science 1983; 220:214-216: , 29. Spealman RD, Goldberg SR. Maintenance of schedule-controlled be- havior by intravenous injections of nicotine in squirrel monkeys. J. of Pharm. and Exper. Ther. 1982; 223:402-408. 30. West RJ, Jarvis MJ, Russell MAH, Carruthers ME, Feyerabend C. Effect of nicotine replacement on the cigarette withdrawal syndrome. Brit. J. Add. 1984; 79:215-219. 106 1 107

Clinical Pharmacology of Nicotine Gun1 Neal L. Benowitz, M.D. Division of Clinical Pharmacology and Experimental Therapeutics 1 40 University of California, School of Medicine, San Francisco Introduction The basis for optimal therapeutics is an understanding of the clinical pharmacology of a drug. In examining the clinical pharmacology of the drug, the first consideration is the desired drug effect, which for nico- tine gum is relief of withdrawal symptoms so that successful cessation of smoking can occur. One must also consider the relationship between the concentration of nicotine in the body and its effects, and the dose of nicotine needed to achieve a desired (`therapeutic") concentration. Blood Nicotine Levels From Smoking The science that relates the dose of a drug to blood concentration is called pharmacokinetics. After a cigarette is smoked there is a rapid rise and fall in blood nicotine levels (Figure 1). The rise reflects absorption of nicotine across the lung. To characterize the pharmacokinetics of nico- tine, we have performed studies with intravenous administration of nico- tine in which it was infused over 30 to 60 minutes to achieve blood lev- els similar to those observed during cigarette smoking (1). From these studies we have determined that after its initial decline (i.e., nicotine mov- ing from blood into tissues) the half-life (the time it takes for nicotine level to fall by one-half), is about two hours. What does a nicotine half-life of two hours mean with respect to hu- man use of nicotine-containing products? It suggests that with regular dosing (regular smoking or regular gum chewing), nicotine will accumu- late over three or four half-lives, i.e., six to eight hours. Thus, nicotine will accumulate in the body of a regular user over the day. It also means that if there is a significant level of nicotine at bedtime, significant levels Supported in part by U.S. Public Health Service Grants DA02277, CA32389, HL29476, and DA01696. Studies were carried out in part in the General Clini- cal Research Center (RR-00083) with support of the Division of Research Resources1 National Institutes of Health. 30 20 10 0 Cigarette Gum Cigar 0 10 20 Time (minutes) ~ 30 40 50 60 70 Figure 1. Plasma nicotine concentrations after chewing one piece of 4 mg nicotine gum, smoking a cigarette, and smoking (non-inhaling) a large cigar. [From Russell and Jarvis, 1980 (6)] 108 1 109

of nicotine (about 5% or 10% of thc level of the night before) will be present even at waking in the morning after several hours of abstinence, Persistence of nicotine in the body for 24 hours a day may have implica_ tions for the pathogenesis of smoking-related diseases. On average 1300 ml of blood are cleared of nicotine per minute in habitual smokers. Clearance occurs primarily by liver metabolism. The clearance concept is important because it relates a given dose ofnicotine to a given level of nicotine in the body. A smoker with rapid clearance needs to smoke much more frequently to maintain a particular nicotine level than does a smoker with slow clearance. We have found that smokers with faster metabolic clearance rates consume greater amounts of nico- tine while maintaining the same level of nicotine in the body compared to smokers with slower metabolism. Metabolism may be one of the de- terminants of individual differences in smoking behavior. As predicted by the pharmacokinetic data on regular smoking, nico- tine levels rise over six to eight hours to an average level of about 30 nano- grams per ml (2). Carbon monoxide has a half-life of three to four hours with normal physical activity during the day, and seven to eight hours during sleep at night (2). With regular smoking, CO levels also increase throughout the day. In the morning, before smoking the first cigarette, CO levels average 4% and increase to 8% by the end of the day (2). 35 J 30 Blood nicotine levels from gum use ~ z 10 0 0 I I •••*••• CIGARETTES -t-NICOTINE GUM (4 mg) -o- NICOTINE GUM (2 mg) In a recent study, we asked volunteers to chew 2 mg nicotine gum, one 1 O piece per hour, for 12 hours per day. While chewing gtun, nicotine 1ev- I m 5 cls plateaued at around 8 nanograms per ml, less than a third of the level the same subjects achieved while smoking cigarettes (Figure 2) (3). Another group chewed 4 mg nicotine gum in the same manner and their nicotine levels plateaued at about 15 nanograms per ml, about 60% of the level observed with smoking. As with smoking, there was accumu- lation of nicotine over several hours and persistence of nicotine in the blood overnight. Remarkably, while chewing 2 mg gum, with levels of only 8 nano- O! I I I I I I I 0900 1300 1700 2100 0100 0500 0900 CLOCK TIME grams per nil, our subjects reported that they were relatively unaffected Figure 2. Mean blood nicotine concentrations throughout the day while by not smoking. With the 4 mg gum, subjects reported feeling fairly smoking cigarette ad libitum or chewing nicotine gum, 2 or 4 mg, one satiated, and did not feel a strong need to smoke a cigarette. The data from piece hourly from 0900 to 2000 (12 pieces). N= 14 for smoking; N= the 2 mg gum subjects suggest that even low levels of nicotine have con- 7 for each gum group. siderable pharmacologic effect. Examination ofdose-response relation- ships for nicotine gum and therapeutic benefit is an important area for future research.

Ni,cotine intake Pharmacokinetic techniques have been used to quantitate the daily in- take of nicotine from different products. The level of nicotine in the blood is not per se a measure of intake of nicotine as that level is determined NICOTINE INTAKE both by intake and, as I noted previously, by the rate of metabolism of DAILY PER CIGARETTE nicotine. Since there is considerable individual variability in the rate of NICOTINE INTAKE OR PIECE OF GUM I metabolism ofnicotine, intake canvary widely yet resultin the same nico- l l . eve an --4 -r I t tine I 1.5 To determine intake ot nicotine from cigarettes or nicotine gum, we studied volunteer smokers on a research ward. On day one, after over- CIGARETTES T night abstinence from cigarette smoking, subjects received an intravenous ~. G U M infusion of nicotine. Knowing the dose administered and measuring blood 30 T levels of nicotine during and after infusion, clearance (rate of metabolism) was determined (4). For the next three days subjects smoked cigarettes , or chewed nicotine gum. Over one 24- hour period blood levels were , 1 0 ! ~ a t, h~ T I I I I I lia,a~. .. ~ ... . .... ..... ~ ...... ............ b .b .. ... ~ - chewmg gum. t5lood levels sampled throughout the day and clearance d (~ CV Or) J I 2 data were used to compute the daily intake of nicotine (4). C'~ " While smoking, the average daily consumption of nicotine was 33 mg 2 T (Figure 3) with a range from 15 to 56 mg/day (3). Intake per cigarette was measured by dividing the 24-hour intake of nicotine by the num- ~-- 0.5 ber of cigarettes smoked. On the average, intake was about one milli- ram per cigarette the same as the average smokin machine ield , g y . , n J ~ ~, ,~ ~, ~ g However, tnere was no correlation between intaKe ot nicotine per cigarette and the machine-determined yield. Subjects' average intake from chewing 12 pieces of 2 mg gum per day was about 10 mg nicotine; from the 4 mg gum it was about 15 mg (3). 0 Nicotine intake from chewing gum was much less than intake from smok- n l ~ 2MG 4MG 2MG 4MG GUM GUM GUM GUM mg ctgarettes. However, subjects smoked an average of 35 cigarettes per day with no limits on number of cigarettes smoked, while gum use was limited to 12 pieces. Whether nicotine intake would have been greater with unlimited availability of gum is not known. The average intake of nicotine was 0.9 mg per cigarette and 0.8 mg per piece of gum for the Figure 3. l)aily intake of nicotine and average intake of nicotine per 2 mg gum study group. Intake averaged 1.2 mg per cigarette and 1.4 mg cigarette or piece of gtun while smoking ad libitum or chewing nicotine per piece of gum for the 4 mg gum group. gum, 12 pieces per day. N- 7 for each group. We hypothesized that heavy smokers (i.e., those with greater habitual nicotine intake) would consume greater amounts of nicotine from gum than lighter smokers. However, there was no correlation between daily intake of nicotine while smoking and intake while chewing gum. There was a significant correlation (r = 0.75) between intake of nicotine per 112 ~ 113

piece of gum and intake per cigarette. This observation suggests that habitual smokers regulate intake of nicotine per item (i.e., per cigarette or piece of gum) such that the more nicotine they consume per cigarette, the more they consutne per piece of gum. We also found that the correlation between intake per piece of gum and intake per cigarette was higher for the 4 mg than 2 mg gum. Presurna., bly, this is because there is more flexibility in adjusting intake with the 4 mg, that is, dosage can be better titrated. Whether greater flexibility in regulating nicotine intake (i.e., use of 4 mg vs. 2 mg gum) improves outcome of smoking cessation therapy needs to be studied further. Sites of nicotine absorption from nicotine gum Seventy percent of nicotine is converted to cotinine, its major metabolite. We found that blood cotinine levels in subjects using nicotine gum were higher than predicted (considering the differences in nicotine levels) from blood cotinine levels measured while the same subjects were smoking cigarettes (3). What does this observation mean? To understand our observation, one must consider the concept of first-pass metabolism. As one smokes, nico- tine enters the lung and then enters the systemic circulation. In every cir- culation, about 30% of blood flow enters the liver, where nicotine is metabolized. Nicotine contained in the other 70% of the blood goes to various body tissues or is recirculated. When chewing nicotine gum, some nicotine is absorbed directly through the buccal mucosa and enters the systemic circulation, similarly to inhaled nicotine. But in addition, some of the nicotine may be swallowed and enter the GI tract, to be absorbed through the intestine. This nicotine enters the portal circulation, which then goes into the liver. All of the nicotine in the portal circulation is sus- ceptible to metabolism by the liver before it reaches the systemic circu- lation. This is what is termed first pass metabolism. With extensive first pass metabolism, levels of metabolites in the circulation relative to the levels of the parent drug will be higher compared to when the drug is given systemically (or inhaled). Thus, swallowing nicotine could explain unexpectedly high cotinine compared to nicotine levels. We used continine-nicotine blood level data and the estimate of 30'%, bioavailability of nicotine from the intestine to estimate buccal versus gas- trointestinal absorption (3). Bioavailability refers to the fraction of a dose of a drug that enters the systemic circulation. After inhalation, the bio- availability of nicotine is 100%. After swallowing, for reasons discussed above, bioavailability is 30%. For the 2 mg gum, the daily systemic "'- take of nicotine as estimated from blood levels averaged 10.2 mg; buccal absorptioti was 9.5 mg; 2.6 mg were swallowed. For the 4 mg gurn, the average daily systemic dose was 14.7 mg; buccal absorption was 10.9 mg; 12.5 mg were swallowed. These observations, coupled with the idea that there is a correlation between the intake per piece of gum and cigarette smoking, suggest that people can control intake of nicotine from the gum by controlling how much nicotine is swallowed versus how much they hold in their mouths. This is analogous to the smoker regulating intake by changing cigarette puffing behavior. Cardiovascular effects of nicotine gum Let us next consider the relationship between the nicotine concentra- tions in the blood and its effects. The primary cardiovascular effects of nicotine, seen after smoking or infusion of nicotine, are increased heart rate and blood pressure and decreased skin temperature, the latter reflecting constriction of blood vessels in the skin (1). Studies with intravenous nico- tine show that early in the course of infusion, when nicotine levels are relatively low, heart rate accelerates but then plateaus despite rising nico- tine levels. After the infusion is terminated, as nicotine levels decline, heart rate falls slowly. The data indicate sensitivity to low levels of rucotine with an apparent ceiling to the response. Skin temperature decreases steadily as nicotine levels rise, and returns to baseline as nicotine levels fall. Fur- ther analysis of the relationship between cardiovascular eflects and nicotine levels over time using hysteresis plots suggests substantial but incoin- plete development of tolerance to heart rate acceleration, no tolerance to skin temperature changes, and partial tolerance to blood pressure elevation. How do these observations relate to the effects of cigarette smoking and chewing nicotine gum? While smoking, the heart rate is higher than during abstinence. Consistent with a plateau in effects at low levels of nicotine, heart rate acceleration was no different over a fourfold range in nicotine levels (5). Heart rate acceleration persisted overnight, cor- responding to the overnight persistence of nicotine levels in the body. We recently completed a study cornparing thee cardiovascular effects of cigarettes and nicotine gum (5). The data from 2 and 4 mg gum chewers were analyzed together because cardiovascular responses were statisti- cally indistinguishable. Data were based on measurements taken by nurses five times a day for two days while subjects were smoking cigarettes or chewing the gum. As expected, blood pressure and heart rate increased during smoking as compared to abstinence. Chewing nicotine gttm in- creased blood pressure but had no effect on heart rate (Figure 4). We 114 1 115

BLOOD PRESSURE (mmHg) r-130 F-110 . i- 80 r 70 r- 90 I- 70 Z~ [ I I I L S G A 4- 60 HEART RATE (BPM) I I I DOUBLE PRODUCT (mmHg • min-') .$ r 9000 T ~- 8000 1- 7000 L -1_-1 S G A S G A Figure 4. Cardiovascular effects of smoking cigarettes (S) or chewing nicotine gum (G) compared to abstinence from tobacco or gum (A). Data represent average of five measurements (every 4 hours) per day for two days per subject per treatment. Measurements were made while subjects were recumbent. Double product, the product of systolic blood pressure and heart rate, reflects myocardial work and oxygen demand. * = Significantly different from abstinence. N= 14. hypothesize that the discrepancy in heart rate response to smoking ver- sus gum use is due to differences in the pharmacokinetic profile of nico- tine taken by smoking and nicotine taken by gum. Smoking produces repeated peaks and troughs of nicotine; chewing nicotine gum produces nucotine levels that gradually rise and fall. When nicotine levels are ris- ing or are sustained, tolerance develops. When nicotine levels fall, toler- ance is lost. With smoking, tolerance may be repeatedly gained and lost but overall some effects persist. With nicotine gum use, sustained nico- tine levels may result in the full development of tolerance. Similarly, there are likely to be different psychological responses to the rising and fall- ing concentrations of nicotine while smoking compared to the relative- ly constant levels of nicotine while chewing nicotine gum. Health risks of nicotine gum The health risks of long-term use of nicotine gum are not yet estab- lished. For cigarette smokers, though, the major health consequences of smoking are lung cancer, chronic lung and cardiovascular disease, includ- ing premature atherosclerosis, acute myocardial infarction and sudden death. Nicotine may contribute to the cardiovascular effects of smok- ing: it increases heart rate and blood pressure, thereby increasing my- ocardial oxygen demand; it may induce platelet aggregation and-hyper- coagulability; and it releases catecholamines which may contribute to arrhythmias. Theoretically, nicotine gum could have similar effects on users. However, since nicotine gum increases heart rate less than cigarette smoking, the effect on myocardial oxygen demand should be less, perhaps resulting in a lower risk of myocardial infarction, angina or sudden death. Based on our data, nicotine gum appears to be a reasonably safe alterna- tive to smoking and is probably less hazardous than smoking, even for someone with coronary heart disease. Other potential nicotine-related hazards such as peptic ulcer disease and reproductive disorders should be considered. At present we do not know to what degree the ulcer-promoting effect of smoking is due to local effects of nicotine on the stomach or duodenum and to what degree it is due to systemic effects. Some people complain of gastrointestinal upset when chewing nicotine gum. Our data suggest that substantial amounts ofrucotine arc swallowed by subjects while chewing nicotine gum, which may explain some of the gastrointestinal upset when chewing nicotine gum. Patients with a history of duodenal ulcers may be a high risk popu- lation for therapy with nicotine gum. Smoking is a risk factor for low birth weight and consequently fetal 116 1 117

morbidity and mortality. Possible effects of nicotine on the fetus include reduction of uterine blood flow and/or a direct effect on fetal function. Nicotine and its principal metabolites have been found in the umbilical cord blood and urine of newborn infants of smoking mothers, as well as in amniotic fluid, indicating transplacental passage. While the impor_ tance of nicotine in causing reproductive disorders is unknown, insofar as nicotine may be involved, nicotine gum is potentially hazardous. However, as is true for coronary heart disease, nicotine gum may be safer than smoking for the pregnant smoker who cannot quit smoking be_ cause the level of exposure to nicotine while chewing nicotine gum is less, because substantial tolerance develops to many cardiovascular ef- fects of nicotine, and because exposure to other toxins in cigarette smoke is avoided. Recommendations for future research Our data and the data of other investigators indicate that blood levels of nicotine from nicotine gum are substantially lower than those observed during cigarette smoking. Thus nicotine gum may not really be "substi- tution" therapy, as originally intended, particularly for heavier smokers. Inadequate levels of nicotine may explain some of the therapeutic failures with nicotine gum. To examine this possibility and optimize the ther- apeutic potential of nicotine gum, the relationship between doses of nico- tine gum and smoking cessation rates must be systematically studied. Specific questions to be answered include: is 4 mg gum better than 2 mg gum; is there a relationship between the nicotine levels while the gum is used and outcome; can we use pretreatment nicotine or cotinine levels and nicotine levels during treatment to predict outcome or to monitor dosing? Further information about toxicity with long-term use, partic- ularly by patients with cardiovascular or peptic ulcer disease is needed. Through careful exploration of dose-response and dose-toxicity rela- tionships, with appropriate documentation of nicotine blood levels, the pharmacotherapy of tobacco addiction can be individualized and its benefits optimized. References 1. Benowitz NL, Jacob P III, Jones RT, Rosenberg J. Interindividual variability in the metabolism and cardiovascular effects of nicotine in man. J. Pharm. Exp. Ther. 1982; 221:368-371. 2. Bcnowitz NL, Kuyt F, Jacob P III. Circadian study of blood nico- tine concentrations during cigarette smoking. Clin. Pharm. Ther 1982; 118 32:758-764. 3. Benowitz NL, Jacob P III, Savanapridi C. Determinants of nicotine intake while chewing nicotine polacrilex gum. Clin. Pharm. Therr in press. 4. Benowitz NL, Jacob P III. Daily intake of nicotine during cigarette smoking. Clin. Pharm. Ther. 1984; 35:499-504. 5. Benowitz NL, Kuyt F, Jacob P III. Influence of nicotine delivery on the cardiovascular and hormonal effects of cigarette smoking. Clin. Pharm. Ther. 1984; 36:74-81. 6. Russell MAH, Raw M, Jarvis MJ. Clinical use of nicotine chewing gum. Brit. Med. J. 1980; 280:1599-1602. 119

Nicotine Polacrilex: Clinical Promises Delivered and Yet To Come David P. L. Sachs, M.D. SmokinA Cessation and Research Institute Palo Alto, California and Stanford University School of Medicine Stanford, California Introduction In 1975, when I began my research career in smoking cessation, no other pulmonary specialist was involved in the field. Fortunately, that has changed. In the course of my last six and a half years on the faculty of University Hospitals of Cleveland, as a pulmonary physician and a critical care medical specialist, I kept a log of the patients I would not have had to see in the Medical Intensive Care Unit and the Pulmonary Consult Service had they never smoked. That figure ran consistently in the 70% range. I realized through data such as these, that I owed my liv- ing as a pulmonologist and intensivist to the American and British tobacco industries. They provide me and my colleagues with the patients that we see in the practice of pulmonary medicine. Even though tobacco dependency causes most of the illnesses pulmo- nary clinicians treat, smoking cessation research has yet to become an academically accepted and respected endeavor in academic pulmonary medicine divisions. (With the recent infusion of NIH funds in this treat- ment area, these attitudes may change.) Nonetheless, the attitudes of prac- ticing physicians, though, is quite different. They want to know the results of current smoking cessation research and how they can use it to treat this country's number one preventable killer - diseases caused by cigarette smoking. I believe that this awakening in the medical practice commu- nity is fiindamentally due to the introduction into clinical medical prac- tice of a medication to help treat tobacco dependence: nicotine polacrilcx (Nicorette). Medicalization of Smoking Thus, the first and probably most important promise on which nico- tine gtnn has delivered has been the medicalization of smoking. A very distinguished medical journal editor wrote several years ago that he was Flgure 1 10 8 N = 1,567 p < 0.001 5.1% 3.3% 1.6% O 0.3% ~~ 6 4 2 0 A B C D Sixty to 120 seconds of simple, firm, unequivocal physician advice to pa- tients in a primary care, general practice setting, produced a seventeen- fold increase in one-year sustained abstinence, vs. the no-intervention con- trol group. (N = 1,567; p < 0.001). Patients were randomly assigned to each of the four conditions, A-D. Patients in Group A were the control group and were only asked by the physicians whether or not they smoked. Patients in Group B received the same treatment as those in Group A, except they were also given a 33-item questionnaire about their smok- ing habits and their health beliefs regarding smoking. Patients in Group C received identical treatment to patients in Group B but were also given 60-120 seconds of firm, simple, unequivocal physician advice to stop smoking. Patients in Group D received the same treatment as patients in Group C except they were also given a health education booklet and were advised by the physician that he would be following up on their smoking behavior. The total physician contact time in either Groups C or D was 60-120 seconds. [Adapted from Russell et al, 1984 (2).]

very concerned about the medicalization of smoking (1). I think it is the other way around: the medicalization of smoking, both because of the appearance and development of nicotine gum and also because of the ap- pearance, at about the same time, of research showing the effectiveness of physician delivered smoking cessation interventions, has had the almost circular effect of making practicing physicians increasingly aware of treat- ment interventions that can be directed toward cigarette smoking. Thus, I see the medicalization of smoking as positive and not something to be concerned about. As I shall discuss below, one effect of this "medicali- zation" has been the increasingly colleageal interaction between both prac- ticing psychologists and physicians and also medical and behavioral research investigators. The second promise, which has yet to be fulfilled, is that nicotine gum is hoped to be only the first in a series of medication that will enhance tobacco dependency treatment. Physician As Catalyst. Returning to the medicalization of smoking, Russell was the first researcher to show that physicians can serve as a catalyst in helping people to stop smoking (2). He showed that 60-120 seconds of simple, general practitioner advice, accompanied by an educa- tional leaflet and the comment that the physician would follow-up on the patient's smoking status, helped the patients stop smoking and re- main nonsmokers for a one-year period (2). Figure 1 summarizes the results. Although the absolute magnitude of sustained abstinence might be viewed as low, it still represented a seventeenfold increase over the con- trol group's sustained abstinence rate of 0.3 % (p < 0.001). Interestingly, although the average sustained abstinence among patients of the 28 general practitioners involved in the study was 5.1 %, some physicians effected greater abstinence rates than others: the sustained, one-year abstinence rates ranged from 0% to 11% (2). This original study was not designed to identify those factors related to why some physicians had better ef- fects than others when delivering smoking cessation advice. It appears not to have been basic demographics or socioeconomic status, since these were relatively constant across all 28 physicians involved in the study. Rather, it was more likely attributable to the individual style that some physicians used. Current research at a number of sites across the United States should shed important light on how physicians can deliver sim- ple, even more effective advice in primary care outpatient settings. Other studies published since 1980 confirtn the unique, catalytic role which physicians can have in changing a patient's behavior from smok- ing to nonsmoking (3,4). 122 physician As Direct Provider. Although most physicians are very adept at talking to patients, and although physicians are involved daily in changing patient behavior (e.g., taking a medication four times a day, entering the hospital for surgery), few physicians, other than psychiatrists, regard themselves as skilled in changing such a profound behavior as cigarette smoking. Studies have documented that although physicians be- lieve there is nothing more important they could do than help their pa- tients stop smoking, the overwhelming majority are exceedingly pessimis- tic about their ability to change smoking behavior and help patients quit. In fact only 4% believe they can change smoking behavior (20). Conse- quently, even this simple tool-clear, succinct, and unambiguous advice to stop smoking -has not been used by the majority of physicians. With the introduction of nicotine polacrilex (more popularly known as nicotine gum) in the United States in 1984, American physicians were suddenly given a tool to use in treating tobacco dependency with which most feel much more at home: a prescription medication. Thus the phy- sician's role shifted from that of being a catalyst to being a direct provider of smoking cessation treatment. All physicians are accustomed to writ- ing prescriptions. Indeed, it is a traditional part of the fundamental, med- ical socialization process that begins in medical school. The development and availability of nicotine gum has forced physicians to take more of a role in treating tobacco dependency. In October, 1984, four colleagues and I presented a day-long post- graduate course for members of the American College of Chest Physi- cians entitled, "The Scientific Basis of Smoking Cessation Intervention." We repeated the same course one year later in October 1985. During that intervening year the physicians attending the course seemed to have gained a substantially higher level of insight and sophistication. Particularly, many physicians had become quite familiar with the behavioral medicine liter- ature and with the drug abuse literature. I believe the fundamental rea- son for this was that they realized that to properly use nicotine gum, they had to become knowledgeable in these behavioral areas. Thus, I believe availability of nicotine gum and the need to use it correctly provided the catalyst for these physicians to delve into the behavioral sciences litera- ture for the first time. Additionally, the availability of nicotine gum leads to physicians' awareness of the need to combine behavioral and pharmaco- logic management to optimally treat tobacco dependency. Natural History of Tobacco Dependency Recent psychological research has shown that tobacco dependency has a natural history that is very similar to diseases with which physicians 123

Table 1 Natural Histories of Tobacco Dependency and Asthma Phases Tobacco Dependency Initiation: Person Begins Smoking Age of Onset: Adolescence to - 21 Etiology: Peer pressure; Rebellion against authority; Assertion of autonomy; Social pressure Treatment Block Etiologic Agent, i.e., Prevention: Change peer dynamics, etc. Maintenance: Person Continues Smoking Etiology: Nicotine dependency: Mood and Affect Modulation; Environmental cues. Tceatment: Nicotine substitution/ replacement; Alteration of reinforced behavior patterns; Stress manage- ment; Minimize exposure to known environmental Asthma Airways Become Hyperreactive Childhood and/or Adulthood Viral/bacterial bronchitis; Inflammation; Tobacco smoke (?); Air pollution; Allergy; Industrial futnes%aerosols Block Etiologic Agent, i.e., Prevent exposure to etiologic agents Person Continues to Have Shortness of Breath, Wheezing, Cough, and/or Sputum Production Exposure to any of the following (partial list): Allergens; Stress; Cold air; Exercise; Tobacco smoke; Smoke; Air pollution; Industrial fimies; Viral infection. Beta-2-adrenergic agonists; Theophylline; Steroids; Stop smoking; Stress reduction; Minimize exposure to known etiologic agents. agents, etc. Cessation: Person Stops Smoking Asthma Symptoms Stop Etiology: Generated by Maintenance Generated by Maintenance Phase Treatment Phase Trcannent Relapse: Person Resumes Smoking Asthma Symptoms Recur Etiology: Stress; Noncompliance with Stress; Noncompliance with medication (nicotine nudication(s); Loss of polacrilex); Loss of motivation; Concurrent motivation, etc. illness; Intrinsic worsening of airway hyperreactivity; Resum- ing smoking Treatmcnt: Basically same as for Basically same as for Maintenance Phasc Maintenance Phasc are more familiar (5,6,7). This should further facilitate physician's involve- ment in diagnosing and treating tobacco dependency. The natural his- tory of tobacco dependency is contrasted with a "classic" medical disorder, asthma, in Table 1. Although I am changing some of the specific names that the original investigators have used so as to better fit the traditional medical model, the concepts are fundamentally unchanged. This table shows not only the similarities in the basic phases, but also the striking similarities in the treatment interventions that can be directed at each phase. For example, Maintenance Phase Treatment for either tobacco dependency or asthma involves both pharmacologic management and behavioral management. Relapse, in either disease, can be triggered by many of the same processes. These concepts are ones with which physicians are familiar, so most can readily extend them to the new field of tobacco de- pendency. Basically any chronic medical disease could be substituted for asthma in the right hand column of Table 1. With all chronic medical disorders, physicians are always attuned to the fact that relapse is a very real possi- bility against which they must be ever vigilant. With the proper guidance, physicians should be able to extend existing skills they use day in and day out in the management of chronic medical disorders to the manage- ment of tobacco dependency, which is also a chronic disease. , . Research Ideas. Another value in conceptualizing tobacco dependency as a chronic medical illness, as shown in Table I and Figure 2, is that it focuses attention on additional treatment interventions-intcrventions in which nicotine polacrilex was the pioneer. Since the etiologic factors at each phase are generally different, the treatment interventions that would be directed at each phase would also be different. For example, a theoretically highly effective intervention would be directed at the first force, peer pressure, by discouraging children and adolescents from tak- ing up the smoking habit. While variants of traditional preventive medi- cine approaches could be used, the most effective may be counter- advertising, such as has been practiced by the Australian government (Bit- toun, International Update, this volume) and by DOCs (Doctors Ought to Care), in which cigarette smoking is deglamorized and robbed of its adult mystique. By the time the stnoker is approximately 10 years into his natural his- tory, he is most likely a fully addicted smoker. These are the individuals that physicians, psychologists, and other health professionals are most apt to see for a specific treatment intervention. These are the individuals who started smoking in adolescence because it was "cool" and thought they could stop at any time they wanted to. When they first began 124 1 125

Figure 2 FORCES DRIVING SMOKING t o0°i° 50% Peers Social Cur osity cultural factors i5 20 first cigarette Usefulness as a psycf7ological . toot in arousal and mood control ~ AG E increasing frequency of smoking 25 regular dependent smoking 30 This schematic illustration shows the relative importance of various forces driving smoking behavior, as a function of age. Arbitrarily, for the pur- poses of this illustration, the age of first cigarette is defined as 15. While age is plotted on the horizontal axis, the relative magnitude of the given forces is represented by the vertical axis. For example at age 15, 100% of the motivating force comes from such factors as peer pressure, curi- osity, adolescent rebellion, etc. Two years later only about 10% of the driving force conies from that source, while 50% comes from more general social and cultural factors, about 20% from smoking's useful- ness as a psychological arouser and mood regulator, and the final 20% from physical dependency (nicotine dependency). Roughly ten years into the smoker's natural history, approximately half of the driving force comes from physical dependency and the other half from smoking's "usefulness" as a psychological tool controlling arousal and mood states. trying to stop in their early 20s, they began to realize that they were com- pletely addicted to tobacco products and that the issue of smoking was something they did not have control over. In treatment clinics I have run, I have found that young professionals, business men, and business women in particular, who exercise a high degree of control over events, budgets, and political affairs, find this especially frustrating: that they cannot control their own smoking behavior. Pharmacologic management can be directed at either the physical dependency force, the arousal and mood control force, or both. Nicotine polacrilex predominantly affects the physical de- pendency force, but it also has some impact on the mood and arousal con- trolling force. A scrotonin re-uptake antagonist, which also increases neu- ropeptides such as beta-endorphin, would act predominantly to affect the arousal and mood control force and would probably have minimal direct impact on the physical dependency force. This line of conceptu- alization, in turn, raises the idea of employing conjoint or sequential phar- macologic management to minimize or ameliorate both major forces driv- ing smoking in regular dependent cigarette smokers. Other pharmacologic agents, such as serotonin re-uptake antagonists, may func- tion as treatments for two phases, either simultaneously or sequentially. For example, such an agent might function initially most effectively as a maintenance phase treatment, leading to the cessation phase, and then be a useful adjunct in preventing the relapse phase from occurring be- cause of the demonstrated effect of some serotonin re-uptake antagonists to decrease high carbohydrate food intake. Behavioral modification tech- niques are aimed predominantly at the mood and arousal controlling force. Their use is critical to successful, pharmacologic management of physi- cal dependency, and will most likely remain of critical importance no mat- ter what additional pharmacologic agents are added to fill out the paradigm outlined above. Thus, a model such as illustrated by Table 1 and Figure 2 points out a number of future promises, originating from the use and understanding of nicotine polacrilex's mechanisms of action, which might ultimately be delivered. Physician Delivered Smoking Cessation Interventions Earlier we reviewed some of the data showing that simple, clear-cut physician advice helped patients to stop smoking and remain nonsmokers for at least a one-year period (2,3,4). Recent data show that nicotine Polacrilex, 2 mg, multiplies the effect of simple physician advice (8,9). In one study, Russell replicated his earlier physician advice study (2) but added the new intervention condition of physician advice + education leaflet + follow-up warning + nicotine polacrilex, 2 mg (8). He found 126 1 127

Figure 3 10 8 2 0 A B C D E F To simplify comparison of two Russell studies, the original study data (Groups A-D) appear again in the left hand portion of this figure, while two of the conditions from the subsequent study are in the right hand portion, Groups E & F. (Note that the second study was carried out utiliz- ing a totally different group of general practitioners with a totally different patient base.) Patients in Group E received the identical treatment as pa- tients in Group Ill from the earlier study (see Figure 1 legend). Whether or not the patient accepted the offer of a nicotine gum prescription, all patients randomized to Group F were included in the data analysis. Fur- ther analysis of the data, in the original article, showed that those patients who used more than one box of nicotine gum, 2 mg, had sustained one- year abstinence that was nearly threefold higher than all patients in that group: 24%>. (N = 1,354; p < 0.001). [Adapted from Russell et al, 1983 (8).1 that the sustained one-year abstinence, or true cure rate, was effectively doubled even though nearly half of the people assigned to that group declined the offer of the nicotine gum prescription + initial instruction (see Figure 3). Also impressive in these results is that there was not specific follow-up to determine whether patients were using the gum correctly, chewing it correctly, or experiencing side effects. Fagerstrom found similar results (9) and also showed that regular follow-up increased one-year sustained abstinence (see Figure 4). This study was also carried out in a general practice setting with motivated patients randomly assigned to one of four conditions. In the short follow- up condition, patients received simple physician advice with one follow- up visit two weeks after initial visit. In the long follow-up condition, the initial advice was the same; however, the physician then telephoned the patient at one week to find out how he or she was doing. The two-week follow-up appointment was the same as in the short follow-up condi- tion. The patient had an additional return appointment one month after the initial advice. Finally, three months after the initial advice the physi- cian sent a personal letter to the patient to provide encouragement and remind him or her to call for an appointment if he or she was having difficulty controlling smoking. Simply adding the extra telephone call and follow-up visit within the first month (along with the threC-month letter) boosted the sustained one-year abstinence, objectively confirmed by exhaled air carbon monoxide, from 3% to 15%. Adding nicotine gum, 2 mg, (Groups B and D) along with proper initial and follow-up instruc- tions in gum use boosted sustained, one-year abstinence over sevenfold in the short follow-up condition and almost twofold in the long follow-up condition (9). Multi-Component Interventions with Nicotine Polacrilex Figure 5 shows a conceptual schematicc of treatments aimed at the two major limbs of the tobacco dependency equation: nicotine dependency and psychological dependency. In order to provide optimal treatment for tobacco dependency, the interlinked dependencies must be treated. The illustration is also designed to show the probable neurochetnical over- lap between these two dependencies. Nicotine, once absorbed into the serum, triggers the release of various central neuropeptides, such as beta- endorphin, which improve mood and affect. The setting or cue in which this mood change occurs becomes the conditioned stimulus which leads to the psychological dependency, e.g., pairing of smoking a cigarette with feeling angry. Thus, this hypothetical schematic in Figure 5 shows the feedback between these two dependencies. 128 1 129

Figure 4 30 25 U .F20 15 10 5 0 A B C D Patients coming to see their general practitioner were randomly assigned to one of two basic conditions: short follow-up (Groups A and B) or long follow-ttp (Groups C and D). Patients randomly assigned to Groups B and D also received nicotine polacrilex, 2 mg. Self-reported nonsmok- ing status was objectively confirmed by exhaled carbon monoxide meas- urement. This study would support the conclusion that both regular phy- sician follow-up and use of nicotine polacrilex, 2 mg, are independent multipliers of the effectiveness of basic, physician advice. (N = 145; p < 0.05). [Adapted from Fagerstrom, 1984 (9).] Figure 5 I Self Help Nicotine polacrilex Brief M.D. advice ? Nasa/ Nicotine Solution (NNS) ? 4' Counseling ? Transdermal Nicot ne Patch (TNP) ? M D counseling & regular follow-up ? Nicotine Aerosol (MDI) ? Comprehensive multicomponent ? Clorndine 9 behavior modification This figure illustrates the interlinked nature of the two major dependencies driving tobacco dependency: nicotine dependency and psychological de- pendency. 130 1 131

The probable biochemical interlinkage between these two dependen- cies is illustrated pictorially by the crossover of the black and white dots. Evidence is accumulating that when nicotine is ingested, it stimulates the release of a number of neuropeptides, which in turn modulate psycho- logical dependency (18). A serotonin re-uptake antagonist which increases circulating beta-endorphin levels might offer the first pharmacologic route for assisting in the management of Psychological Dependency that could also feed back onto the Nicotine Dependency side, decreasing nicotine requirements in treatment. Also, an agent of this class which decreased carbohydrate craving, such as fluoxetine has been reported to do, might have an important role in preventing relapse by minimizing weight gain after smoking cessation. (See reference 18 for one study of such a phar- rnacologic agent.) At the present time, the only agent which clinical outcome data show to be effective in treating nicotine dependency is nicotine polacrilex. Other papers in this volume discuss different potential routes of nicotine ad- ministration that are in various stages of development and evaluation, in- cluding nasal nicotine solution (NNS), transdermal nicotine patch (TNP), and nicotine aerosol. While clorudine has been shown in one, 8-hour study (10) to relieve craving for tobacco, compared with placebo, it has not yet been tested in clinical outcome smoking cessation trials. Even though there is a good theoretical rationale to support its use in treating tobacco de- pendency (10,11), at present we do not have evidence that clonidine would, in fact, be useablc as a smoking cessation aid. Moreover, since known side effects of clonidine include somnolence and depression, it could, in sustained use for treating tobacco dependency, aggravate other of the tobacco withdrawal symptoms. Thus, its use in clinical practice at this time to treat tobacco dependency probably should be confined to specific research protocols. At the present time there are no behavioral treatments available for the nicotine dependency side, although aversive condition- ing techniques may well act at this site. Because of the persistent, consistent, and pioneering research efforts of behavioral scientists over the last 15 years, we have available a wide range of behavioral interventions with which to treat the psychological dependency component. At one end are minimal intervention strategies, such as self-help guides, and brief, simple physician advice. As behavioral intervention becomes more extensive and intensive (including compre- hensive psychological counseling, extensive physician counseling and scheduled follow-up, and use of comprehensive, multi-component, be- havior modification programs) sustained abstinence increases. The ar- row alongside these interventions in Figure 5 expands toward the arrow- head not only to reflect the increased, sustained, one-year abstinence rates but also to reflect the increased time involved in providing such treat- ments. There are no pharmacologic agents of proven efficacy to combat psychological dependency. For example, if you set psychological dependency treatment to 0, which the dispensary model effectively does, then nicotine polacrilex, 2 mg, produces no more effective results than placebo gum (12). This dispen- sary model is equivalent to a physician responding to a patient's telephone inquiry about obtaining a prescription for nicotine polacrilex by merely phoning a prescription to the pharmacy without providing any instruc- tion, counseling, or follow-up. On the other hand in a separate study, when all patients were given a comprehensive six-week group counseling program with half randomly assigned to receive nicotine polacrilex, 2 mg, and the other half to receive placebo gum, marked and highly significant differences in sustained, one- year abstinence resulted: 38% (2 mg gum) vs. 16%0 (placebo gum), p < 0.01 (13). A highly relevant point in considering the implications of such a study in the physician practice environment is that this study was car- ried out in a specialized smoking cessation clinic, without the benefit of the individual's personal physician also serving as a motivating factor. No data exist at present to show quit rates that might be obtained in'a medi- cal practice setting using a similar kind of program or with physicians pr'Scribing nicotine gum but then referring the patients to a comprehen- sive, psychological counseling program, such as the above study em- ployed. My bias is that the addition of personal physician advice, follow- up, and nicotine gurn prescription will serve as a multiplier producing even further enhanced results than those reported by Jarvis, et al. (13). Nicotine Polacrilex Combined with Multi-Component, Aversive Conditioning Strategies Several recent studies have shown the incremental effect that can be produced by combining nicotine gum treatment with comprehensive, multi-component behavioral treatment involving aversive conditioning, skills training, and relapse prevention. A particularly unique study em- ployed a multi-component, aversive, smoke-holding procedure during the first week of the cessation, or induction phase. Subjects were then randomly assigned to one of three maintenance phase conditions for six weeks: 1) nicotine polacrilex, 2 mg, on an ad lib basis; 2) skills train- ing/relapse prevention, only; 3) combined nicotine polacrilex, 2 mg, ad lib + skills training. 132 1 133

Figure 6 Follow Up Point Aversion K N = 64 o All Subjects (N = 64) • Nicotine Gum (2 mg) + Skills Training (N = 22)  Nicotine Gum (2 mg), alone (N - 22) I • Skills Training. alone (N = 20) ~ 0 6 wks 15 wks Nicotine Gurn, 2 rng N- 45 IntlucUon Phase t wk Maintenance Phase 6 wks Followup Phase (No Interventlon) 9 rnos 100 80 60 40 20 0 Unusually good long-term abstinence of 50%, ten and a half months after the end of the induction phase or nine months after the end of all intervention, was achieved by combined use of nicotine polacrilex, 2 mg, + behavioral skills training, compared with 23-30% for each compo- nent, alone. This study was also one of the first to have a clearly defined induction phase -a daily, aversive, smoke-holding procedure (one week), followed by the six-week maintenance phase condition. Although the sample size was too small to provide sufficient power to enable discrimi- nation of significant differences between the three maintenance phase con- ditions, all three showed statistically significant results at the ten-and-a half month follow-up point, compared with the start of treatment (N = 64; p < 0.02). [Adapted from Killen et al, 1984 (l4).] Eighty-three percent stopped smoking at the end of the one-week in- duction phase (14). After the end of the six-week maintenance phase, there was no further treatment of any kind. Specifically note that nicotine gum was not available at all after the end of the six-week maintenance phase. ( This study was carried out before the prescription availability of the medi- cation; hence, subjects could not have gone to their own physician, for example, to obtain further nicotine polacrilex.) Ten and a half months after the beginning of the maintenance phase, or nine months after the end of all treatment intervention, abstinence rates, most likely sustained ab- stinence, were as follows: nicotine polacrilex alone, 23% vs. skills train- ing alone, 30% vs. nicotine polacrilex + skills training 50% (14) (see Figure 6). This study from the Stanford Heart Disease Prevention Program is im- portant for a number of reasons. Not only does it integrate two of the most powerful smoking cessation strategies, aversive conditioning and nicotine substitution therapy; but also, this study is one of the first to con- ceptually separate the initial cessation phase (induction phase) from the maintenance phase (or relapse prevention phase). This distinction has be- come increasingly important in conceptual design of smoking cessation studies. While this approach is critically important from a behavioral and methodologic standpoint, it also provides another area of common ground that physicians can easily appreciate and understand. In cancer chemotherapy, another chronic disease like tobacco depen- dency, a very similar approach has been used. Chemotherapy for induc- tion phase of a neoplastic disease typically involves one set of agents, designed to act biochemically in a certain way, while maintenance phase chemotherapy involves use of different agents, with different biochenu- cal mechanisms of action, designed to prevent recurrence of disease. Fi- nally, if the cancer does recur, then either the same initial chemotherapeutic agents or a different set will be used to attempt to achieve re-induction. Obviously, the importance of regular follow-up in the management for any chronic medical disorder -asthma, diabetes, hypertension, renal failure, or neoplastic diseases, for example-is well established in medi- cine. Once physicians achieve an even better understanding of the universal Principles of tobacco dependency management, they should find this very compatible with their management approach to other medical diseases. Interdisciplinary Research. Finally, nicotine polacrilex has catalyzed interdisciplinary research in tobacco dependency in other ways. For ex- ample, Hall randomly assigned subjects to one of three treatment con- ditions: 1) multi-component, normally paced, aversive smoking alone; 2) nicotine polacrilex, 2 mg, alone; or 3) combined multi-component, 134 1 135

Figure 7 100 0 r ~ • Combined (N = 35)  Nicotine Gum, 2 mg (N - 42) . Multi-Component Normally-Paced Aversive Smoking (N = 36) 0 I I 3 6 Months After Treatment 100 80 60 40 20 J 0 12 At cach of the three follow-up points after initiation of treatment, three, six and 12 months, the treatment package utilizing both nicotine polacri- lex, 2 mg + an aversive conditioning procedure (normally paced aver- sive smoking) (was more effective than either component alone. (N = 120; p < 0.04 at time 0, 3, and 6 months.) [Adapted from Hall et al, 1985 (15).] normally paced, aversive smoking + nicotine polacrilex, 2 mg (15). Her hypothesis, appropriately, was that since the two major interventional components, the aversive conditioning vs. nicotine substitution, acted at different points in the Cessation Phase, there should be at least additive improvement in abstinence in the combined intervention. This was in- deed the case (see Figure 7). Based on these latter two studies, Sachs and colleagues did a pilot study on four of 54 patients with moderate-severe cardiac and pulmonary dis- case. This group of 54 patients, following multi-component, rapid smok- ing therapy as originally described by Lichtenstein (16), showed sustained abstinence from the end of treatment to the two-year follow-up point ofexactly 50% (17). This result was statistically significant from the 0%, two-year sustained abstinence in the medically matched waiting list con- trol group (p < 0.0001). There were a number of points about these data that surprised us. First, the subject population was older than has usual- ly been studied in the smoking cessation literature: mean age nearly 52 years. Also they were high relapsers, having tried seriously an average of 8.8 times to stop smoking in the past, using a mean of 3.3 different interventional techniques (17 and unpublished data). Moreover, many of the patients had already sustained a myocardial infarction, had been firmly and unequivocally advised by their cardiologist to stop smoking, yet had resumed smoking upon discharge from the coronary care unit. This subset of people is particularly difficult to treat. Thus, the two-year sustained abstinence of 50% with multi-component, rapid smoking treatnient, without any pharmacologic intervention, was much higher than we had anticipated. (We are currently completing five-year follow-up data col- lection on this group of cardiopulmonary patients.) In the pilot study we called back four of the subjects who had failed treatment initially and had not stopped smoking at the end of the multi- component rapid smoking treatment period. Extending Hall's approach with normally paced aversive smoking (15), we provided these four with multi-component rapid smoking therapy as they had originally received; however, we instructed them in the use of nicotine polacrilex to help con- tain smoking urges in between treatment sessions. Moreover, although these four subjects completed their rapid smoking treatment phase with- in one month, nicotine gum was available for three months. Two of these four, who had not been able to stop at all in previous treatment, did suc- ceed in stopping smoking by the end of treatment and maintained ab- stinence for six months from pilot study start (self-report confirmed by carboxyhemoglobin, serum nicotine, cotinine, and thiocyanate). 136 1 137

Conclusion The advent of nicotine polacrilex has helped to catalyze the medicali- zation of smoking. It has opened up new avenues of interdisciplinary research and clinical activity between physicians and behavioral scien- tists. Whereas in 1975 there were only one or two physicians outside of the field of psychiatry involved in smoking cessation research, the number is now growing rapidly. This, alone, should provide increasingly excit- ing and important research opportunities in behavioral medicine. From a practical standpoint, the appearance of nicotine polacrilex- the first of what I hope will be a number of pharmacologic agents to as- sist in the management of tobacco dependency-has made physicians more comfortable in treating tobacco dependency and is beginning to shift the bclief commonly held by physicians that the treatment of tobacco dependency is without a scientific base. While it is generally true that phy- sicians, as a group, feel more comfortable with pharmacologic manage- merit than behavioral modification approaches, the appearance of nico- tine polacrilex has clearly opened the door to enable many physicians to begin to appreciate the value, importance, and scientific validity of be- havioral approaches that also must be brought to bear in the treatment of tobacco dependency. All of this in turn places us on the threshold of a totally new era, analo- gous to the appearance of the first effective antibiotics. In the 1930s a new era dawned in the treatment of infectious diseases with the appearance of sulfa drugs. We now, of course, have antibiotics that provide treatments we could never have dreamt of in the'30s. I think that by the turn of the century, the picture in the smoking cessation treatment field will like- wise include treatments of which we haven't yet dreamt. References 1. Blum A. Nicotine chewing gum and the medicalization of smoking. Ann. Int. Med. 1984; 101:121-123. 2. Russell MAH, Wilson C, Taylor C, Baker, CD. Effect of general prac- titioners' advice against smoking. Brit. Med. J. 1979; 2:231-235. 3. Jamrozik K, ae~rial of three differentaantismoking inter ent ons ln H. Controll general practice. Brit. Med. J. 1984; 288:1499-1503. 4. Rose CT'tlti)smoking eadviSe1r10ycar results: J. Epide niol. trollcd trial of ant Comm. Health 1982; 36:102-108. 5. Marlatt GA, Gordon JR. Determinance of relapse: Implications for the maintenance of behavior change. In Davidson PO, Davidson SM' (Eds) Behavioral Medicine: Changing Lifc~ Styles. New York: Brun- ner/Mazel, 1980. 6. Prochaska JO, DiClemente CC. Stages and processes of self-change of smoking: Toward an integrative model of change. J. Consult. Clin. Psychol. 1983; 51:390-395. 7. Lichtenstein E. The smoking problem: A behavioral perspective. J. Consult. Clin. Psychol. 1982; 50:804-819. 8. Russell MAH, Merriman R, Stapleton J, Taylor W. Effect of nico- tine chewing gum as an adjunct to general practitioners' advice against smoking. Brit. Med. J. 1983; 287:1782-1785. 9. Fagerstrom K-O. Effects of nicotine chewing gum and follow-up ap- pointments in physician-based smoking cessation. Prev. Med. 1984; 13:517-527. 10. Glassman AH, Jackson WK, Walsh BT, Roose SP, Rosenfeld B. Cigarette craving, smoking withdrawal, and clonidine. Science 1984; 226:864-866. 11. Sachs DPL. Pharmacologic, neuroendocrine, and biobehavioral ba- sis for tobacco dependence. Current Pulmonology 1987; 8:In Press. 12. Schneider NG, Jarvik ME, Forsythe AB, Read LL, Elliott ML, Schweiger A. Nicotine gum in smoking cessation: A placebo- controlled, double-blind trial. Add. Beh. 1983; 8:253-261. '' 13. Jarvis MJ, Raw M, Russell MAH, Feyerabend C. Randomised con- trolled trial of nicotine chewing-gum. Brit. Med. J. 1982; 282:537-540. 14. Killen JD, Maccoby N, Taylor CB. Nicotine gum and self-regulation training in smoking relapse prevention. Behav. Ther 1984; 15:234-248. 15. Hall SM, Tunstall C, Rugg D, Jones RT, Benowitz N. Nicotine gum and behavioral treatment in smoking cessation.J. Consult. Clin. Psy- chol. 1985; 53:256-258. 16. Lichtenstein E, Harris DE, Birchler GR, Wahl JM, Schmahl DP. Comparison of rapid smoking, warm, smoky air, and attention place- bo in the modification of smoking behavior. J. Consult. Clin. Psychol. 1973; 40:92-98. 17. Hall RG, Sachs DPL, Hall SM, Bcnowitz NL. Two-year efficacy and safety of rapid smoking therapy in patients with cardiac and pulmo- nary discase. J. Consult. Clin. Psychol. 1984; 52:574-581. 18. Pomerleau OF, Pomerleau CS. Neuroregulators and the reinforce- ment of smoking: Towards a biobchavioral explanation. Neurosci. Bio- behau Rev. 1984; 8:503-513. 19. Petraglia F, Facchinetti F, Martignoni E, Nappi G, Volpe A, Genaz- zani AR. Serotoninergic agonists increase plasma levels of beta- endorphin and bcta-lipotropin in humans. J. Clin. Endocrinol. Metab. 1984; 59:1138-1184. 138 1 139

20. Rosen MA, Logsdon DN, Demak MM. Prevention and health pro- motion in primary care. Preu Med. 1984; 13:534-548. Problems of Nicotine Gum John R. Hughes, M.D. University of Vermont College of Medicine Burlington, Vermont Abstract Side effects from nicotine gum are not uncommon; however, they are rarely serious and are not a major deterrent to its use. Both behavioral and physical dependence on the gum can occur among those who stop smoking while using the gum. Low quit rates with thegum are probably due to physician noncompliance (with giving appropriate instructions, rationale and expectancies for the gum) and to patient non- compliance (with medications and adjunctive therapy). In addition, how best to select patients, adjunctive treatments, doses and duration of treatment is unclear. Introduction Drug treatment is not simply knowing which drug to prescribe for which illness. The effective clinician knows that several factors control whether the drug will or will not be effective. The factors include ad- junctive non-pharmacologic treatments, compliance, dose, duration of treatment, expectancy, instructions, patient selection, rationale and side effects. These factors are especially influential when behavioral disorders are being treated. The thesis of this paper is that problems in the above mentioned areas presently limit the efficacy of nicotine gum and, in fact, may limit the efficacy of any drug treatment for smoking. Until solutions to these problems are empirically tested, the optimal efficacy of nicotine gum will not be known. Adverse effects Side effects and contraindications. The most common side effects from nicotine gum are jaw ache, belching, hiccups, nausea, burning or sore throat and upset stomach (11). The first two symptoms arc from the stiffness of the gum and swollowing air; the remainder arc from tuco- Acknowledgments: Preparation of this article was supported by a grant (DA-04066) "'ld a Research Scientist Development Award (DA-00109) from the National In- stitnte on Drug Abuse. Thanks to Merrell Dow for perrnission to use their tele- P"()tle survey data. 141 140

tine itself. These side effects abate over time and can be diminished by chewing more slowly. Serious side effects of the gum are quite rare. Side effects have been reported to occur in 40% of users; however, they pre- vent adequate use of the gum in only 25% of users (15). Contraindications to the gum seldom prevent its use; however, some people who might benefit most from the gum (e.g., post-MI patients) should not receive it according to Federal Drug Administration (FDA) guidelines (17). Dependence. Fear of developing dependence on nicotine gum is one of the major reasons smokers do not use nicotine gum and physicians do not prescribe it. $ehavioral dependence (i.e., chronic use) has been reported to occur in 10% of smokers (11). Although this rate seems low, it is about the same as that for bcnzodiazepines (14) and is much greater than that for nar- cotic analgesics (19). The quoted 10% rate is deceptively low for two rea- sons: 1) the criteria used has been continuous use for one year and 2) the rate is based on all smokers prescribed the gum. If instead, the criteria is use beyond the recommended period (i.e., > 6 months) and only those who quit successfully with the gum are examined, then the rate of be- havioral dependence is much higher; i.e., 35-54% (6,8,15). However, in one study the incidence of prolonged use of gum (> 6 months) among ex-smokers was identical for nicotine and placebo gums (8). Thus, prior reports that among all smokers, nicotine gum is used longer than placebo gum appears to be because nicotine gum produces more cessations rather than nicotine gum is more addictive. Thus, psychological factors (e.g., attributions of success to gum) may determine dependence more than pharmacological factors. Several factors could influence the prevalence of behavioral dependence on nicotine gum: e.g., dose, cost, schedule (use as needed versus fixed- time), duration of treatment, and prior history of drug abuse. Prelimi- nary results of our study indicate dependence does not differ between the 2 and 4 mg doses (55% and 60% used the gum > 4 months) (3). The effect of other factors on behavioral dependence has not been tested. Physical dependence (i.e., withdrawal upon cessation) on the gum has been documented in two double-blind, placebo-substitution trials (9,24). In one of these, withdrawal occurred in almost all subjects, was qualita- tively and quantitatively similar to tobacco withdrawal, and appeared to prompt relapse to smoking (9). Abuse of nicotine by nonsmokers has not been reported. In addok ~ nicotine is not a reinforcer, but rather is a punisher among never-sm and long-abstinent ex-smokers (23). Lack of efficacy Patient noncompliance. A recent U.S. telephone survey found that among all those who obtained the gum, the average number of boxes of gum used was 3.0 and the average length of use was 36 days (15). Among those who quit with the gum, the average number of boxes was 4.2 and the average duration was 54 days. Many investigators believe more frequent and longer use of the gum increases cessation rates (2). Motivation appears to influence the amount and duration ofuse of nico- tine gum. When nicotine gum is given to motivated smokers (i.e., those enrolled in a study on the gum), almost all fill their prescription and many (e.g. > 50%) use at least two boxes of gum (8). However, when the gum is given to all smokers regardless of motivation, fewer (54%) fill the prescription and only a minority (35%) use at least two boxes of gum (18). Cost is probably a factor also. In the only study in which subjects paid for nicotine gum, even fewer (35%) filled their prescription (16). To verify this cross-study comparison, we are now directly comparing the use and efficacy of the gum when patients pay 0%, 25%, or 90% of its cost. Dose appears to have effect on both the initial and long-term use of gum. The prevalence of use of 2 and 4 mg doses after three weeks is simi- lar (95% and 100%) and greater than that of 0 and I mg (unbu~fered) gums (40% each). After nine months, only the 2 mg dose users main- tained use of gum (44% vs. 0-15 % for other doses) (3). Finally, factors known to influence compliance with other prescription medications (e.g., perceived need for medicine and duration of medication regimen) might also influence use of nicotine gum but have not been tested. The FDA-approved indications for nicotine gum state it is to be used "while participating in a behavioral modification program" (17). However, less than 3% of those using the gum are participating in any type ofpsy- chological treatment (bibliotherapy excluded) (8,15). Whether this represents patient noncompliance or physician noncompliance in instruct- ing patients is unknown. Physician noncompliance. The efficacy of most psychoactive medi- cations is highly dependent on the patient receiving information about the appropriate expectancy, rationale, and instructions about the medi- cation. According to patients, 66 % of physicians spend < 10 minutes talking about the gum; 90% did not tell them to stop smoking abrupt- lY; 90% did not tell them about side effects; 78% did not give them any type of cessation materials and 84% did not schedule a follow-up visit within a month (15). Training does improve physician compliance (8) and, as Dr. Wilson and colleagues' study (this volume) shows, such training can improve the efficacy of the treatment. 142 1, 143

The unknowns. One of the major problems with the use of nicotine gum is the lack of empirical data about how best to use it. For most drugs, patient selection, adjunctive treatment, dose of the drug and duration of treatment can have a profound impact on the efficacy of the drug. Whether this is true for nicotine gum has only begun to be tested. At present, several common ideas about how to best use the gum have not been verified by experimental tests: 1) only motivated smokers benefit from thpy 3) the the gum works only when accompanied byanbfcha~ avy therapy; and 4 mg dose is more effective than the 2 mg g 4) the longer patients chew the gum the better. In this section we review the evidence to support these ideas. Patient selection. One could make a case to prescribe nicotine gum to 1) smokers who have not decided to quit, 2) only motivated smokers, or 3) only smokers who are dependent on nicotine. Prescribing the gum to undecided smokers might be best as one of the main effects of the gum is to prompt smokers to try to quit (18,20). Six-month quit rates among all smokers, which includes mostly undecided smokers, range from 12%- 28% (11). Prescribing the gum to only motivated smokers might be best as many believe that in order for the gum to work smokers must be motivated to overcome the habit component of smoking. In addition, this is the FDA approved use of the gum (17). Six-month quit rates in withdrawal clin- ics, ics, which include only motivated smokers, range from 23 /o-63 "/0 (11). Prescribing the gum only to smokers who arc dependent on nicotine might be best because they will likely benefit the most. In addition, high rates of cessation with this hard-core group might encourage physicians to prescribe nicotine gum. Six-month quit rates among dependent smok- ers prescribed the gum range from 27%-71 %(11). In summary, it appears that the more selective the criteria for patient selection, the higher the quit rate. However, the price paid for this strategy is the failure to offer treatmfits harms and smokers. of t c.at ngws. not reatl considers thce relative bene ing each type of patient (e.g., see Oster, this volume) can determine to whom it is best to prescribe the gum. Adjunctive therapy. Smoking cessation requires overcotning not only nicotine dependence but alsothe l h b intc is ty and conte t aOnenst dy the habit of smoking vary widey found an additive effect when behavior therapy was added to nicotn1e gum (13). Quit rates at 10.5 months were 23% for gum alo 30`)I°thor behavior therapy alone and 50% for gum plus behavior thcra pAn study found that the addition of behavior therapy to a gum plus educa- tion and group therapy protocol increased one year quit rates only slightly (from 37% to 44%) (4). Some believe adjunctive therapies are essential for nicotine gum to be effective. A nonrandomized study compared the effects of nicotine and placebo gums when given via a dispensary and when given with behavior therapy (22). In this study nicotine gum was superior to placebo gum only when subjects also received behavior therapy. Contrary to this find- ing, preliminary results from a randomized study indicate nicotine gum is superior to placebo gum even when an intensive behavior therapy is not given (8). In addition, several non-placebo studies have found the gum effective even when there is little extra therapy; i.e. less than 10 minutes of physician advice (1,18,20). Thus, the important question that remains to be studied is what is the cost-effectiveness of adding adjunctive therapy? The cost-effectiveness will likely be determined by the type of adjunctive therapy and type of smoker (e.g. level of nicotine dependence). Dose. Low doses of nicotine gum (1 mg) have been reported to be in- eflective (16). High doses (4 mg) have been reported to be helpful to heavy or dependent smokers yet more aversive to most smokers (21). However, none of these conclusions are based on experimental trials. Preliminary results from a small dose-response study found little effect of dose on six-month cessation rates (1 mg = 40 %, 2 mg = 53 %, 4 mg = 47%) (3). This study also found little difference in compliance between the 2 mg and 4 mg doses. Thus, these experimental results contradict the eGrlier clinical reports (21). At present, a multicenter study is testing whether 4 mg gum is effective for smokers who failed on 2 mg gum. Duration. Smokers who use the gum longer have higher quit rates (2). Some have assumed that this relationship is causal and have encouraged ex-smokcrs to chew the gum for longer periods of time. However, this relationship may be due to a third variable; i.e., the more compliant smokers both quit smoking and chew gum for longer. The single prospec- tive, experimental test of the effect of duration found that those given the gum for six months did no better than those given the gum for only one month (2). Conclusion At the conference from which these papers were compiled, several in- vestigators were excited about going "beyond" nicotine gum; i.e., the prospects for new pharmacological treatments for smoking. My thesis 144 1 145

in this review has been that many of the problems of nicotine gum are not due to the gum itself, but rather, to problems with patient and phy- sician compliance and with knowing how to best use the drug. These same problems will occur with any pharmacological treatment of smok- ing and until we begin to solve them, any drug for smoking is unlikely to be highly successful. References 1. Fagerstrom K-O. Effect of nicotine chewing gum and follow-up ap- pointments in physician-based smoking cessation. Preu Med. 1984; 13:517-527. 2. Fagerstrom K-O, Malin B. Nicotine chewing gum in smoking ces- sation: efficiency, nicotine dependence, therapy duration, and clini- cal recorrunendations. Pp. 102-109 in J Grabowski and SM Hall (Eds.) Pharmacological Adjuncts in Smoking Cessation, National Institute on Drug Abuse Monograph 53, Washington, DC: DHHS Pub No. (ADM)85- 1553, 1985. 3. Gust SW, Hughes JR, Keenan R. A randomized, placebo-controlled study of the effects of dose on nicotine gum self-administration. Presented at the Annual Meeting of the International Study Group Investigating Drugs As Reinforcers, Baltimore. June, 1985. 4. Hall SM, Tunstall C, Rugg D, Jones RT, Benowitz, N. Nicotine gum and behavioral treatment in smoking cessation. J. Consult. & Clin. Psych., 1985; 53:256-258. 5. Haynes RB, Taylor DW, Sackett DL (Eds.) Compliance in Health Care, Baltimore: Johns Hopkins Press, 1979. 6. Hjalmarson Al. Effect of nicotine chewing gum in smoking cessa- tion. J. Am. Med. Assoc. 1984; 252:2835-2838. 7. Hughes JR. Identification of the dependent smoker. Validity and clin- ical utility. Beh. Med. Abstracts 1984; 5:202-204. 8. Hughes, JR, Gust SW, Keenan S, Skoog KP, Pickens RW, Ramlet D, Healey M. Efficacy of nicotine gum in general practice. Annual meet- ing of the American Psychological Association, Washington, August, 1986. 9. Hughes JR, Hatsukami D. Physical dependence on nicotine gum: A placebo-substitution trial. J. Am. Med. Assoc., in press. 10. Hughes JR, Kottke T. Doctors helping smokers: Real world tactics. Minnesota Medicine, in press. Med, 11. Hughes JR, Miller S. Nicotine guwn to help stop smoking.J• Am. Assoc. 1984; 252:2855-2858. 12. Hughes JR, Pickens RW, Keenan W, Spring W. Instructions control whether nicotine will serve as a reinforcer. J. Pharm. & Exper. Ther. 1985; 235:106-112. 13. Killen JC, Maccoby N, Taylor CB. Nicotine gum and self-regulation training in smoking relapse prevention. Beh. Ther. 1984; 15:234-248. 14. Marks J. The benzodiazepines: Use, over-use, mis-use, abuse. Lan- caster, PA: MTP Press, 1978. 15. Merrell Dow Pharmaceuticals. Nicorette Long Term Users Survey, Merrell Dow Pharmaceuticals, 1984. 16. Ohlin P, Westling H. Nicotine containing chewing gum as a substi- tute for smoking. Pp. 171-174 in RG Richardson (Ed.) The Second World Conference on Smoking and Health, London: Pittman, 1972. 17. Package insert, Nicorette, 1984. 18. Page A,, Walters DS, Schlegel RP, Best JA. Smoking cessation in fam- ily practice: the effects of advice and nicotine chewing gum prescrip- tion, unpublished manuscript. 19. Porter J, Jick H. Addiction rare in patients treated with narcotics. N. Eng. J. Med. 1980; 302:123. 20. Russell MAH, Merriman R, Stapleton J, Taylor W. Effect of nico- tine chewing gum as an adjunct to general practitioner's advice,against smoking. Brit. Med. J. 287:1782-1785. 21. Russell MAH, Sutton SR, Feyerabend C, Cole FV. Effect of nico- tine chewing gum on smoking behavior and as an aid to cigarette withdrawal. Brit. Med. J. 1976; 2:391-393. 22. Schneider NG, Jarvik MR, Forsythe AB, Read LL, Elliot ML, Schweiger A. Nicotine gum in smoking cessation: A placebo- controlled, double-blind trial. Add. Beh. 1983; 10:253-261. 23. Strickler G, Hughes JR, King D. Nicotine as a reinforcer among never-smokers and ex-smokers. Presented at the Annual Meeting of the American Psychological Association, Washington, DC, August, 1985. 24. West RW, Russell MAH. Effects of withdrawal from long-term nico- tine gum use. Psych. Med. 1985; 15:891-893. 146 1 147

Discussion: Development of Nicorette: Its Uses and Limitations Chair: Ellen Gritz, Ph.D. Jonsson Comprehensive Cancer Center, University of California, Los Angeles Presenters: M.A.H. Russell, M.R.C.P. Institute of Psychiatry, London Neal L. Benowitz, M.D. University of California, San Francisco David P. L. Sachs, M.D. Stanford University School of Medicine, Palo Alto John R. Hughes, M.D. University of Vermont College of Medicine, Burlington DR. GRITZ: I think we're ready to take questions and discussion from the audience. Yes, why don't you begin there in the back. NORMAN WORTHER: I'm a family practitioner in the area of Philadelphia. We have been interested in smoking cessation for a num- ber of years, and I found Dr. Hughes' point concerning patient selectivity to be extremely important. Dr. Hughes commented that family practice patients are not as motivated as clinic patients. But in our family prac- tice, we felt that the problem related not to the fact that the patients were not as motivated, but that patient selectivity was probably the key fac- tor. We attempted to achieve a partial selectivity, less than Dr. Fagerstrom's, but morc than most physicians in general will attempt to do. I think part of the problem has been that a physician will say to a patient, "You have lung disease. You will develop heart disease. You must stop smoking. Here is your gurn" A patient goes through a spectrum of changes, an evolu- tion from the point, as noted by Dr. Sachs, of initiation of the cessation process all the way to the finishing of the process, when the patient is abstinent. It is inappropriate, I think, for a physician to attempt toY ur a patient stop smoking when the patient is in the initiation phase. success rate is going to be low. We felt that the best place for a physto nt to start would be to introduce that personalized target message at the p where the patient is most susceptible to thc message. And yet, "`ie did not know how to define that patient or pull that patient out of the general context, so patients signed up for a program in response to a sign in the office, thereby achieving our selectivity without going through Fager- strom's study. We had 140 patients in our last study, and we have already achieved the guaranteed 17 percent success rate primarily based on the selectivity process. In order for a physician to be relatively successful, he has to have an interest in the problem of smoking cessation. He has to be able to be trained to some degree in this problem, and he has to have the time to spend on it. In addition to the doctors' attitude and the patients' attitude, the rapport aspect is also extremely important, but rapport can be de- fined in many different ways and has different faces to present. We found the financial burdens of treatment were extremely impor- tant. In our practice of greater than 10,000 patients, at least 7000 of whom are HMO, we achieved a population in our study that was probably 80 to 90 percent HMO. We attribute this to the fact that the program for them cost five to ten dollars and the gum was free. Among our private patients, we had a very poor success rate in attracting them. DR. GRITZ: Thank you very much. Dr. Henningfield. Dependence liability of nicotine gum DR. HENNINGFIELD: A bit of an observation about a comment mainly by Dr. Hughes. I agree completely with you that the main problem wthatixhbthire gumto is not inherent to the gum itself, but how it's used. What ngs rnind is the issue of dependence liability versus therapeu- tic compliance. You showed that the nicotine gum can function as a rein- forcer relative to placebo gum. That's important for a number of reasons and has some of its own dependence liability. However, relative to tobacco and other substances of abuse, the gutn has a very low dependence lia- hility. In fact, we found that as we increased the dose of the gum, we had decreased liking scale scores. And, in fact, it was at higher dose levels, Which also produced significant decreases in liking scale scores, that we found efficacy of the drug in our acute paradigm. This is also consistent With Dr. Benowitz' pharmacokinetic data which show lower blood lev- els With the gum than when people are smoking their regular cigarettes, the same kind of relationship he found when people were switched to cigarettes that were not their preferred brands. bR• HUGHES: I think the question of the relative reinforcing efficacy ')ftohacco and nicotine gum is a very interesting question. I submit that °ne of the studies that you would probably be best at doing,jack, would bc to get smokers to stop and give them the choice between the two 148 1 149

substances. See how hard they'll work for nicotine gum and how hard they'll work for cigarettes. Again, if I show that nicotine is self- administered more than placebo in a therapeutic setting, does that mean that it's efficacious or does that mean it has dependence liability? I'm not going to try to answer that question. I' think that it's the same issue as this: if someone's hooked on the gum, it's not as bad as being hooked on cigarettes. On the other hand, as Dr. Shiffman showed, 39 percent of the people want to quit smoking to be free from an addiction, and here we are putting them on another one. Combined treatments DR. PUENTE: I would like to raise a point about the combined use of behavioral treatments with chewing gum. Our experience in Mexico during the last four years is that it doesn't matter what kind of treatment you combine with the chewing gum. All those combinations produce the same results. We have been using health education, and psychologi- cal techniques about stress, depression, and relaxation. And we have also used cognitive rehearsal. And our results are very much on the same line. Seventy percent of the people who finish a group program of four weeks, meeting twice a week in a group of ten people, stop smoking. The problem then is relapse. We have started a follow-up with two aims; one of evaluation of theru aphese follow ups attone week after, ~ wo weeeks, abstinence. We cond a month, two months, four months, and six months. The other point I would like to make is that we find psychologists sometimes are better therapists than doctors for many reasons. One rea- son is that sometimes doctors are very reluctant to do health education. And doctors are relucta supervise followeupaTlus is something the finished, to do a supervision during psychologists do very well. Doctor-patient relationships DR. SACHS: I'd like to make a comment about that. fl nd th~is~ a~~ her will illustrate how a doctor-patient relationship or, pp rg way, how a physician who is sensitive can benefit from what the wife chological sciences have to teach. At another meeting, a husbandoand- team-he's a pulmonary physician, she's a nurse in Billings, M na were planning to set up a smoking ce~ h ap ~ gs psy ~o ~~offi be~~ were asking me about teaming up s that are he wasn't sure he could really master all the very subtle thing important in treatment. He then told me a very interesting story about an elderly lady in her 60s whom he had been advising repeatedly to stop smoking. She told him several years ago, "Look, Doctor, I enjoy smok- ing, and there's just no way I'm going to stop smoking, period." He would sound her out on an annual basis, and got basically the same response. About one or two months after his previous attempt to get her to stop, she was in the hospital for a simple medical problem, not smoking-related. During rounds, she said to him, "You know, one of the physical ther- apists is able to get me my vitamins at discount price." And she had plotted out how much she was saving in dollars and cents because she was get- ting cut-rate vitamins. And he said, "I'm amazed that you're so impressed at saving this much money from vitamins. Have you ever stopped to think how much money you would save if you never smoked?" Now, normally that's the kind of intervention that wouldn't do any good. But he knew what she would be sensitive to. And she has now not smoked in some- what over a year. That illustrates one of my major theses, that it's worth- while considering combining the two relationships. DR. GRITZ: I'm going to take only one more question. That gives us two minutes total. Physician noncompliance CINDY RAND: I'm from Johns Hopkins University. First, I'd like to thank Dr. Hughes for introducing the term physician noncompliance. It's a refreshing contrast to the focus that we all have on those difficult, obstinate noncompliers in our treatment programs. In keeping with that, I have a question for Dr. Sachs. Those of us who have committed our- selves to research or treatment regarding smoking cessation have to live with the fact that we have relatively poor success. If we manage to have 40 percent abstinence among our people at the end of a year, we'd con- sider ourselves lucky. And yet, when we look at Dr. Russell's work, Dr. Fagerstrom's work with the use of gum, we're talking about failure rates at the end of the year of 80 to 90 percent. My question to you is: in your training of residents, in your training ofphysicians, how do you address the issue that those who try hard are still going to have the vast majority Of their patients failing? I find in my own work with pulmonary physi- cians, one of the reasons they no longer make a concerted effort to prescribe the gum is such repeated failure, not just with gum, but with general advice. How can you overcome that? DR• SACHS: My own personal experience in house staff training and in dealing with physicians has been somewhat different than yours. I found 151 150

that when physicians become aware of Dr. Russell's data, for example, their expectancy is different. They aren't expecting to cure 90 percent of the people who smoke. Then they feel very comfortable, even though they might only have a five percent one-year sustained abstinence rate from simple advice. In terms of 30 and 40 percent cure rates from much more laborious interventions, there are groups of physicians who do that day in and day out. I refer particularly to oncologists, physicians who specialize in treating cancer disorders. Many cancer chemotherapy and radiation therapy cure rates are no more than 15 to 30 or 40 percent, and moreover these treatments provoke intense iatrogenic responses that land the patient in the intensive care unit for weeks because their blood counts are so low. Skin sloughs off. The side effects can be horrendous. I think it's a matter of expectaticy. There arc some people who simply are not going to be happy if they're making their living doing something that is only 40 percent effective, and those people obviously aren't going to be interested in using nicotine gum or anything else. There are other phy- sicians who, knowing that this intervention produces a higher sustained abstinence rate than the control condition and knowing the tremendous costs involved with continued smoking - health costs - are provided with the motivation. DR. GRITZ: John Hughes claims one last ten-second comment, and then we move. DR. HUGHES: That is testimony for a great diagnostic tool - optimism versus pessimism. You know, I've shown the four percent and the nine percent, and the optimist says: boy, I've increased it 225 percent. And the pessimist says: I've only bumped it up five percent. DR. GRITZ: I want to thank you all very much. And now, we're go- ing to move on frotn the present to the future, to methodologies and de- velopment and fantasy, as well as actual forms of new treatment for cigarette or other forms of tobacco dependence. 152

00 N Overview: Alternative Forms of Pharmacologic Treatment Murray E. Jarvik, M.D. Veterans Administration Medical Center University of California School of Medicine Los Angeles Introduction Some years ago in a paper entitled "Can Drug Treatment Help Smok- ers?" (2), 1 concluded that no drug to that date (1977) had been proven successful in the treatment of smoking. It appeared then that rucotine itself was probably the best agent to be used as a substitute. However, I must emphasize that the fact that no drug had been successfully used to treat nicotine dependence did not preclude the future development of a drug for the treatment of nicotine dependence. I further concluded that psy- chological techniques are absolutely essential to help smokers to initiate cessation and prevent relapse. I still believe this. The three papers in this section approach this subject from slightly different angles. Jarvis and Rose each discuss the use of an alternate route of nicotine adrninistration as a possible substitute for the pulmonary route-the route by which subjects self-admiruster tobacco smoke. Rose also deals with the very important problem of the inadequacy of pure nicotine substitution, and suggests substitutes for other components of cigarette smoking, particularly those involving the respiratory tract. Glass- man describes a treatment with a drug other than nicotine. I will discuss each of these papers and consider their rationale and possible usefulness. Nasal nicotine solution Jarvis and his colleagues have found that a nasal applicator devised by AB Lco Laboratories is an efficient means of administering nicotine. The blood levels obtained compare favorably with those seen following in- gestion of nasal snuff, and are intermediate between those derived from cigarette smoking and nicotine chewing gum. Since nicotine chewmK gum is commercially available, one might question the need for nasal nico- tine. Evidently absorption of nicotine from the nasal mucosa is more ef- ficicnt than from the buccal mucosa. Nasal administration of drugs is not 154 common, though it has been used in nasal inhalers such as those delivering phenylephrine and ephedrine, as well as pitocin and certain other pep- tides. One of the fitnctions of the blood supply in the nasal mucosa is to warm the air on its way to the lungs. The improved absorption of nico- tine from the nose as compared with the mouth may provide advantages over nicotine chewing gum. Higher levels are reached more quickly, though not the peaks which the smoker obtains from inhaling cigarette smoke. The present study was not intended to establish the clinical su- periority of nasal nicotine solution over nicotine chewing gum; it de- tertnined its clinical acceptability and its potential as a treatment aid. Jarvis reports that one-third of his subjects remained abstinent through- out a year of follow-up, a figure comparable to the results of clinic treat- ment with nicotine chewing gum. A problem with this study is that it did not use placebos. All of the subjects were highly motivated to stop smoking, and in the context of a very effective smoking treatment clinic there may indeed have been a strong placebo effect, as Jarvis himself admits. Nasal nicotine is potentially a very useful tool in the study of the phar- macodynamics of nicotine. It is one of the few methods of nicotine ad- ministration from which almost peak levels of nicotine can be achieved without invasive (e.g., intravenous) procedures. This is especially useful in studying the effects of nicotine in nonsmokers, and Jarvis has indi- cated that with a simple motor task nasally administered nicotine indeed produces a reliable increase in tapping speed which could be blocked with mecarnylamine. Transdermal patch Rose describes the transdermal patch, an alternate route of administering rucotine which he and I developed together. Nicotine has the chemical characteristics which make it particularly suitable for transdermal adrninistration-it is potent, lipid soluble and has a relatively short du- ration of action. Rose, himself a nonsmoker, was the first subject in our transdermal experiments. The patch produced significant effects which could be attributed only to nicotine, making it another substitute method for delivering nicotine. In the experiment, Rose found that transdcrmal nicotine diminishes the desire for nicotine when subjects are allowed to regulate nicotine concentration using a smoke mixer. We were interest- ed to find that subjects could not discriminate the effects of nicotine from the patch alone, and are conducting other studies to elucidate the phar- macokinetics of transdermal nicotine. Some of the advantages of the trans- derinal patch are that it requires little or no compliance, and a patch can 1S5

deliver nicotine at a fairly constant rate. The constancy can be enhanced by using a rate-limiting membrane, which is one of our aims in the future. Rose gives an excellent summary of the shortcomings of nicotine sub- stitution. He points out that experiments involving intravenous adminis- tration of nicotine do not duplicate the effects of cigarette smoking nor do they produce a similar satisfaction in smokers. Furthermore, they do not always produce a diminution in cigarette smoking. The stimulation produced by cigarette smoke on the respiratory tract appears to be a salient cue in smoking. Rose proposes that the mood-altering effects of nico- tine are too subtle to be perceived by themselves and need pairing with a highly discritninable peripheral cue for identification. His studies, which indicate that tracheal sensation is a very important arrd perhaps essential component of reinforcement by cigarette smoking, should be taken into account by all investigators of smoking. Although cigarette craving decreased after smoking, surprisingly, anesthetizing the trachea with lido- caine blunted the decrease significantly. The fact that tracheal irritation from citric acid mimicked that produced by cigarette smoke and actually produced satisfaction again highlights the importance of peripheral sensations in cigarette smoking. Rose de- vised a very clever technique whereby the bolus of cigarette smoke could be localized in various parts of the respiratory tract by preceding or fol- lowing it with air. When this was done it was evident that stimulation by smoke in the tracheal region was the most satisfying. Furthermore, diluting the smoke removed its sensory qualities but not its pharrnaco- logical qualitics, and this dilute smoke, which presumably delivered the same amount of nicotine to the blood as a real cigarette, was not rein- forcing. It would be useful to follow these studies with pharmacokinct- ic measurements of blood nicotine. Clonidine Dr. Alexander Glassman describes his experiences with clonidine as a means of inhibiting smoking. His results are such that one comes away with the feeling that clonidine might yet be a useful treatment but it is not yet definitive. In essence, Glassman showed that clonidine signifi- cantly reduced the distress of smoking cessation, and that the control drug, alprazolam, also re<luced the distress of smoking cessation but to a less- er extent. Most importantly, clonidine reduced subjective craving to smoke, while alprazolam did not. Glassman discusses the rationale for possible effectiveness of clonidine. Its major action is to decrease t~~cus tivity of the locus ceruleus. He points out that activation of the t~ld ceruleus causes anxiety in humans, and that reduction of activity sh° 156 rclieve anxiety. Presumably smoking withdrawal causes anxiety, arrd this is relieved more effectively by clonidine than it is by a benzodiazepine. His study is very important in the context of the nicotine-oriented studies, because it addresses the effects of nicotine at a different point in the nico- tine cascade than at the more commonly studied site, the cholinergic syn- apse. Ifclonidine can really reduce the stress as well in the cigarette with- drawal syndrome as it does in the alcohol or opioid withdrawal syndrome, then indeed it might be a type of panacea, a general anxiolytic specifi- cally designed for drug withdrawal. Obviously, well-controlled, double- blind replications will be necessary. Hopefully Glassman and others will conduct these studies. Conclusion It appears that the major pharmacological treatment of smoking is sub- stitution therapy with iucotine. Clonidine appears to show promise in alleviating the withdrawal syndrome acutely, and raises the possibility that other drugs may also be useful in the treatment of nicotine with- drawal. Of course, the nature of the withdrawal syndrome must be more intensively studied (1) and the sources of variance producing it better un- dcrstood, so that a rational approach with pharmacological ag`ents may be madc. References 1. Hatsukami DK, Hughes JR, Pickens RW, Svikis D. Tobacco with- drawal symptoms: An experimental analysis. Psychopharrn. 1984; 84:231-236. 2. Jarvik ME. Can drug treatment help smokers? Pp. 509-513 in J Stein- fcld, W Griffiths, K Bell, R Taylor (Eds.) I'roceedinWs qf the Third World Conference on Srnokinq and Health. Washington, DC: US Department of Health, Education, and Welfare, Public Health Service, DHEW Publication No. (NIH) 77-1413, 1977. 157

Transdermal Nicotine as a Strategy for Nicotine Replacement Jed Rose, Ph.D. Veterans Administration Medical Center Los Angeles, California Introduction Transdermal nicotine administration is a strategy for nicotine substi- tution which may realize the promises of nicotine chewing gum while avoiding sorne of its problems. It is widely known that nicotine can be absorbed through the intact skin, but this fact has usually been cited in the context of nicotines toxicity. Indeed, "green tobacco sickness" results when tobacco harvesters handle wet tobacco leaves with their bare hands (4). Nonetheless, the successful application of nicotine chewing gum with cigarette smokers suggested that the controlled application of nicotine via a transdermal patch might also be effective in aiding smoking cessa- tion. In principle, transdermal nicotine administration might have two significant advantages over nicotine chewing gum. First, transdermal nicotine would avoid the bad taste and gastrointestinal complaints some- tinus associated with nicotine gum (18). This, in turn, could maximize therapeutic effects by allowing a higher nicotine dose to be tolerated. Sec- ond, a long-acting patch which releases nicotine into the skin for 24 hours or longer would minimize the effort required on the part of the patient, thereby facilitating compliance with treatment. Results of Transdermal Nicotine Studies In the first pilot study of transdermal nicotine, I used myself as a sub- ject in order to establish roughly the dose required to produce signifi- cant systemic nicotine levels (12). Nicotine concentrations were measured in saliva after applying a 9 mg nicotine base to the volar surface of the left forearm. The nicotine was applied in a 30% aqueous solution under a polyethylene patch which was taped in place. A significant amount hf nicotine (50 ng/ml) appeared in saliva within 30 minutes of patch app Sali- cation, and levels remained cl ~~a~ed ~° in hca0rt rateland blood pressure~ va levels werce paralleled by c g ~,,,ith Encouraged by these preliminary findings, we conducted a study ten cigarette smokers (15). Subjects received either 8 mg nicotine or placebo, double-blind, on two different days. On each day, subjects en- tered the laboratory nonabstinent from smoking. At the beginning of the session subjects smoked using a smoke mixing device (6, 11, 13) to select their preferred nicotine deliveries. By turning a knob which blended the snioke from a high nicotine cigarette and a low nicotine cigarette, sub- jects controlled the nicotine delivery in each puff The two cigarettes were identical in all respects aside from nicotine delivery; the nicotine deliv- ery of one of the cigarettes was selectively enhanced by injecting nico- tine into the filter. Dial settings corresponding to the smoke mixture selected for each puff were recorded; additional measures of nicotine preference were collected, based on diverting and trapping a portion of the smoke particulates obtained from each cigarette. The ratio of trapped particulates correlated highly with the mean nicotine preference calcu- lated from dial settings. After the first smoking period, a polyethylene patch was taped to the volar surface of the nondorninant forearm. Under the patch was applied either 8 mg nicotine in a 30% aqueous solution or a placebo solution colored to mimic the nicotine solution. An opaque piece of plastic was placed over the patch to prevent subjects from noticing any reddening of the skin that would be more likely to occur with the nicotine applica- tion. Subjects were deprived of cigarettes for 90 minutes after patch ap- plication. Every 30 minutes subjects reported their craving for cigarettes, and saliva samples were collected in order to assess nicotine absorption. Saliva pH was measured in an attempt to correct for the variable ratio of proportion of nicotine in the uncharged form, which in turn depends on pH. At the end of the 90 minutes smoking deprivation period, sub- jects were allowed to smoke, using the smoke mixer again to select their desired nicotine delivery. We observed a slight, but significant elevation in saliva nicotine with- in 30 minutes after application of the nicotine patch. Using the theoret- ical ratio of blood/saliva nicotine concentrations, we estimated blood levels to be 16 ng/ml in the nicotine condition versus 10 ng/ml in the placebo condition. Transdermal nicotine prevented the rise in the craving for cigarettes seen in the placebo condition (Figure 1). As shown in Figure 2, transdermal nicotine also reduced subjects' nicotine preference dur- 'rig the initial puffs of the second smoking period. Despite the fact that transdermal nicotine affected subjective desire for cigarettes and initial 1ucotine preference when smoking, subjects were generally unable to relia- bly discriminate whether they received nicotine or placebo, based on end- of session interviews. ()verall, these results support the hypothesis that transdermal nico- 158 1 159

Figure 1 6-7 0 Figure 2 v .40~ z w M wd ~_ 30 w+ wU z ~ z c~ 20 zz z Q ~ 10 Placebo ~ .____ -----------0------ _ . ,,~ - -------- - - 30 TIME (MIN) 8 mg Nicotine 60 8 mg NICOTINE ~ 4b. N 0 00 SMOKE SMOKE ON, W MIXER TEST MIXER TEST VA L90 2 -MIN-_ -I 90 tine may help smokers resist the rise in cigarette craving accompanying deprivation. Therefore, transdermal nicotine administration merits fur- ther investigation as a smoking cessation strategy. Shortcomings of Nicotine Substitution It may be important to consider the shortcomings of the nicotine sub- stitution strategy for smoking cessation which may limit its effective- ness. Although reductions in smoking withdrawal symptoms, includ- ing craving, are often reported after nicotine administration by buccal, transdermal or intravenous routes, the effects are usually substantially less than those produced by smoking (8). The desire for a cigarette is not quenched by simply increasing plasma nicotine concentrations, even when nicotine is injected rapidly to produce dramatic rises in plasma nicotine levels (5). The simple nicotine regulation model of cigarette smoking is inadequate to account for this. Recent evidence, summarized below, sug- gests that this shortcoming of nicotine substitution is probably due to the lack of the familiar sensory cues of cigarette smoking, which are po- tent reinforcers. The most salient of these cues seems to be the distinct tracheal sensations accompanying each puff (3), as well as the aroma of the smoke. Moreover, the tracheal sensations are triggered in part by nico- tine (1) and are blocked by mecamylamine (14). Taste, per se, dbes not seem to be an important component of smoking reinforcement, although the word "taste" is often used in a general sense to refer to arorna and tracheal cues. Three lines of evidence suggest that sensory components ofsmoke rival nicotine's pharmacologic etl^ects as reinforcers maintaining smoking. These data were derived from three very different methods for dissociating the pharmacologic and sensory components of smoking satisfaction. The first technique was to locally anesthetize the respiratory tract to blunt the per- ception of smoke while preserving nicotine's pharmacologic effects (16). Subjects rinsed their mouths, gargled and inhaled a mist of a solution con- taining either lidocaine or saline, and subsequently received inhalations ofcigarette smoke. Nicotine intake was regulated by controlling the num- ber of puffs, puff volume and inhalation depth. Subjects reported their craving for cigarettes before and after a block of puffs. Usually, smok- ers' subjective satisfaction is manifested as a drop in reported craving for cigarettes, which was observed in the saline condition (Figure 3). In con- trast, blockade of most of the sensory qualities of smoke with lidocaine removed much of the satisfaction associated with smoking. This find- ing underscores the similarity between cigarette smoking and other con- summatory behaviors such as eating and drinking, in which pcrinheral 161 160

Figure 3 5 PRE SMOKING SALINE POST SMOKING chemoreception plays an important role in satiety. In contrast, self- administration of other substances often displays a less critical role for sensory cues associated with the route of ingestion. For example, there is little reason to believe that a habitual heroin user would not enjoy the drug effect if the arm were anesthetized at the injection site, or that a co- caine user would not enjoy that drug if the nasal passages were numb prior to ingesting the cocaine. The relatively subtle mood-altering ef- fects of nicotine may necessitate its being paired with a highly discrimi.na- ble peripheral cue for the smoker to clearly identify its reinforcing ef- fects, especially the reduction in craving for cigarettes. The second line of evidence implicating the importance to smokers of the tracheal sensations produced by each puff was derived from a study in which we sought to mimic these cues with aerosols containing no nico- tine (10). A fine aerosol of a 15`%, aqueous citric acid solution was used to produce a tracheal sensation similar to that associated with cigarette smoke. To focus specifically on tracheal cues, we blocked oral sensations with lidocai.ne and olfactory cues with noseplugs. In blind ratings, sub- jects reported substantial enjoyment of puffs of citric acid and rated these puffs as significantly more similar to their own brand of cigarette than puffs from a very low tar and nicotine cigarette. A third procedure for dissociating the sensory and pharmacologic ef- fects of smoking is based on the fact that the sensory and pharmacolog- ic actions of smoke occur in different places. Sensory effects are perceived mainly in the upper respiratory passages, the trachea and bronchi. In con- trast, most of the nicotine in smoke is absorbed in the alveoli of the lungs, where few sensory effects are noticed by smokers. Little smoke is deposit- ed in the upper airways due to the size of smoke particles, which does not facilitate deposition in the large airways by diffusion, inertial impaction or gravitational settling (9). Therefore, in one procedure we can deliver the relevant respiratory sensations by restricting smoke to the upper air- ways. This was accomplished in one study by having subjects inhale two liters of air prior to inhaling each measured puff of smoke, and allowing only 60 cc of air to follow each pufE In the complimentary procedure, nicotine is delivered with a minimum of sensory effects. This was ac- complished by diluting each puff of smoke in two liters of air prior to inhalation. The entire two liters was then inhaled, but the extreme dilu- tion drastically reduced the intensity of the sensory effects. Nonetheless, the full nicotine content was delivered after a breath-holding period of tWo seconds. Absorption of smoke particulates was measured by first determining how much particulate matter was delivered from the apparatus and then 163 162

subtracting the amount of particulate matter recovered in Cambridge filter pads that filtered the air exhaled after each puff. llcep inhalations of di- lute smoke were shown to produce significant increases in expired air car- bon monoxide concentrations and heart rate, an index of nicotine absorp- tion. The shallow smoke inhalations produced no significant rise in either parameter. However, shallow smoke inhalations were rated by subjects as much more desirable and satisfying than deep inhalations, which, despite their nicotine content, were rated as little different from control inhalations of air. Thus, the sensory characteristics of smoke, without nicotine, were preferable to nicotine without the sensory cues. Conclusions The results summarized above suggest that transdermal nicotine ad- ministration may partially substitute for the nicotine which smokers ob- tain from cigarettes. The possibility of minimal side effects and ease with which it could be used by patients make it an especially promising tech- nique of nicotine substitution. The other evidence cited above argues that the inability of pure nicotine to fully substitute for cigarettes is due to the lack of sensory qualities inherent in cigarette smoke, and is not due to the lack of an adequate rate of nicotine absorption. Although it is widely believed that the 7-10 second rate of absorption of nicotine from each puff of smoke is important in mediating smoking satisfaction (17), there is little evidence to support this view (7). Indeed, the results of the studies described argue against that hypothesis. A more plausible view of the importance of the rate of nicotine absorption is that rapid absorption, as occurs with inhalation, tends to produce high plasma nicotine levels due to the short distributional half life of nicotine (2). Even so, it may require very high nicotine levels, or a state of prolonged smoking depri- vation, to clearly demonstrate the pleasurable effects of pure nicotine when administered in the absence of associated sensory cues. These effects may nonetheless be insufficient to completely satisfy smokers. Of course, fi-om the standpoint of smoking cessation, satisfying smokers may not he as important as reducing by alternate means their motiva- tion to smoke. For example, transdermal nicotine offers the possibility of producing sutficiently high levels of nicotine to discourage smoking via the aversive effects of excessive nicotine that would occur if an in- dividual smoked when wearing a nicotine patch. The possibility ofpre- cise regulation ofnicotine levels with transderrnal administration tnight allow for the tailoring of dose to individual smokers in a manner that would provide contingent punishment for smoking while also substitut- ing for the pharmacologic components of smoking rcinforccment. In addition to the nicotine replacernent/aversion treatment approaches which lend themselves to the transdermal method of nicotine adminis- tration, an extinction strategy could be envisioned. In order to diminish the craving for familiar sensory aspects of smoking, similar cues could be presented in the absence of nicotine reinforcement, alternating with periods of transdermal nicotine delivery. It nught be expected that this procedure would lead to the extinction of the sensory cues, by breaking their association with nicotine in two ways; the stimuli would be presented during periods of nicotine withdrawal, and nicotine would be present- ed in the absence of the usual sensory cues. Alternatively, it might be effective to give smokers the benefit of sub- stitutes for both the sensory and pharmacologic effects of smoke to mitumize their discomfort, and then fade each out in turn. Some smok- ers may find it easier to break their dependence on nicotine's pharmaco- logic effects first, and later discontinue the sensory components. Other smokers might benefit from the reverse sequence. Which of the several potential approaches just mentioned will be more effective is an empirical issue that will be decided in future studies. However, it is likely that combined strategies addressing both the phar- macologic dependence on nicotine and the potent sensory factors main- taining cigarette smoking will provide a far greater treatment effe+ot than therapies directed to either component alone. References 1. Ashton H, Stepney R. Smoking Psychology and Pharmacology. London: Tavistock Publications, 1982. 2. Benowitz NL, Jacob P, Jones RT, Rosenberg J. Interindividual varia- bility in the metabolism and cardiovascular effects of nicotine in man. /. Pharm. and Exper. Ther. 1982; 221:368-372. 3. Cain WS. Sensory attributes of cigarette smoking. Pp. 239-249 in GB Gori and FG Bock (Eds.) Banbury Report 3: A Safe Cigarette? New York: Cold Spring Harbor Laboratory, 1980. 4. Gehlbach SH, Williams WA, Freeman JI. Protective clothing as a means of reducing nicotine absorption in tobacco harvesters. Arch. qf .Envir. Health. 1979; March/April:111-114. 5. Henningfield JE, Katsumasa M, Jasinski llR. Cigarette smokers self- adtninister intravenous nicotine. Pharm. Biochem. and Beh. 1983; 19:887-890. 6. Herskovic JE, Rose JE, Jarvik ME. Cigarette desirability and nico- tine preference in smokers. Pharm. Biochem. and Beh. 1986; 24:171-175. 164 1 165

7. Kozlowski LT. The determinants of tobacco use: Cigarette smoking in the context of other forms of tobacco use. Canadian J. Pub. Health. 1982; 73:236-241. 8. Kumar R, Cooke EC, Lader MH, Russell MAH. Is nicotine impor- tant in tobacco smoking? Clin. Pharm. and Ther. 1977; 21:520-529. 9. Raabe OG. Physical properties of aerosols affecting inhalation toxi- cology. Pp. 1-28 in CL Sanders, FT Cross, GE Dagle, JA Mahaffey (Eds.) Pulmonary Toxicology of Respirable Particles. Oak Ridge, TN: Tech- nical Information Center, Department of Energy, 1980. 10. Rose JE, Hickman C. Mimicking cigarette smoke with aerosols. Paper presented at the 93rd annual convention of the American Psycho- logical Association, Los Angeles, 1985. 11. Rose JE, Jarvik ME, Ananda S. Nicotine preference increases after cigarette deprivation. Pharm. Biochem. and Beh. 1984; 20:55-58. 12. Rose JE, Jarvik ME, Rose KD. Transdermal administration of nico- tine. Drug and Alcohol Depend. 1984; 13:209-213. 13. Rose JE, Lafer R, Jarvik ME. A smoke-mixing device for measuring nicotine preference. Beh. Res. Meth. and Instrumentation. 1982; 14:501-503. 14. Rose JE, Sampson A, Henningfield JE. Blockade of smoking satis- faction with rnecamylamine. Paper presented at the 93rd annual con- vention of the American Psychological Association, Los Angeles, 1985. 15. Rose JE, Herskovic JE, Trilling Y, Jarvik ME. Transdermal nicotine reduces cigarette craving and nicotine preference. Clin. Pharm. and Ther. 1985; 38:450-456. 16. Rose JE, Tashkin DP, Ertle A, Zinser MC, Lafer R. Sensory block- ade of smoking satisfaction. Pharm. Biochem. and Beh. 1985; 23:289-293. 17. Russell MAH. Nicotine intake and its regulation. J. of Psychosomatic Res. 1980; 24:253-264. 18. Russell MAH, Raw M, Jarvis MJ. Clinical use of nicotine chewing gum. Brit. Med. J. 1980; 280:1599-1602. Nasal Nicotine Solution: Its Potential in Smoking Cessation and as a Research Tool Martin Jarvis, MA, BSc., M.Phil. Addiction Research Unit Institute of Psychiatry London Introduction Nicotine chewing gum has made a great impact on smoking cessa- tion, as is evident from the papers in this volttme and the many reports in the literature. We use it routinely in our clinic and it greatly enhances our success rate. But the gum is only one means of nicotine replacement, with certain disadvantages. First, absorption of nicotine from the gum is slow, and up to 30 minutes chewing is needed to release it all. People who use it rarely maintain blood nicotine concentrations greater thAn one- third their normal smoking level. There are also dependent smokers who cannot use the gum, either because of problems with dentures or pre- existing conditions such as peptic ulcer. There is, therefore, good reason to explore alternative and possibly more efficient ways to provide smokers with a temporary substitute source of nicotine. A few years ago we were impressed by the absorption of nicotine from dry tobacco snuff taken through the nose (2). The boost in plasma nico- tine concentration from a single pinch in a snuff user was similar to that from a cigarette, and subsequent study of a group of regular snuff users confirmed that the level and pattern of nicotine intake was similar to that found in cigarette smokers (3). We were therefore very interested when Ove Ferno developed a device for the application of a droplet of nico- tine in the nose. We call it nasal nicotine solution (NNS). The dose, which is usually I mg or 2 mg, is packaged in a small plastic container from which it is released by breaking off one end and squeezing. Preliminary studies showed that the absorption rate of nicotine from NNS was intermediate between cigarettes and nicotine chewing gum (4). A peak concentration of 14 ng/rnl in plasma was reached about seven and a half minutes after taking a 2 mg dose (see Figure 1). This was similar to the time required to reach a peak blood nicotine level from smoking 166 1 167

a cigarette and was considerably quicker than with the gum. Neverthe- lcss, the nicotine peak was less than that obtainable from a cigarette. When NNS was taken on an hourly schedule over the course of a day, the average peak plasma nicotine concentration reached in five subjects was 16.3 nghnl which is higher than is normally obtained from nicotine chewing gum (5). There was considerable variation between subjects in the nicotine lev- els obtained from the NNS, one subject obtaining concentrations simi- lar to those found in heavy cigarette smokers. These results suggested that NNS may have considerable potential as a form of nicotine replacement in smoking cessation, although the vari- ation in nicotine levels between individuals on the same dosing sched- ule may indicate that attention needs to be focused on the technique of self-administration. We were sufficiently encouraged to mount a prelimi- nary trial of clinical acceptability to see what potential NNS might have as a treatment aid and to find out what problems it presented. This paper gives an outline of the results of that trial; a full account will be published elsewhere. In addition, we found that NNS is a satisfactory way to give nonsmokers nicotine, and brief mention will be made of work we have been doing on the effects of nicotine on simple motor performance in nonsmokers using NNS as a research tool. Short-term trial of clinical acceptability Our intention in this trial was to gather exploratory data on the ac- ceptability of NNS as a treatment aid for dependent smokers. We invit- ed consecutive new attenders at the Maudsley Smokers Clinic to use NNS as a form of nicotine replacement for at least the first two weeks of their quit attempt. Our clinic attendcrs, most of whom are self-referred, are heavier smokers than the population average, and many have very low confidence of ever quitting as a result of a long history of unsuccessful quit attempts. None refused the invitation, and the 26 volunteers in the trial comprised a typical sample of attenders at our clinic. Since we were concerned with establishing information on clinical use of NNS which might form the basis for a full-scale trial later, we focused on the short-term efficacy and on process variables rather than on long- term efficacy. In addition to the fundamental issues of acceptability in terms of local irritancy, adaptation; relief of withdrawal symptoms, and side cffects, we also needed to establish what dosage and pattern of self- administration subjects preferred when they took NNS ad libitum dur- ing cigarette withdrawal. We g:t, r subjects supplies of both 1 tng and 2 mg strengths and suggested that they start with the 1 mg in order to facilitate adaptation. Once the subjects had adapted to NNS, we 185-, 165-~ 145-I 125- 105-I Plasma nicotine 85-I (nmol/I) 65-I 45-I 25-I 5-I I 0 NNS Cigarette Nicotine gum T 10 T------- 1- 20 30 Time (min) t L_1 1 40 1- 50 1 60 Figure 1. Average plasma nicotine concentrations of three subjects after smoking a cigarette, taking nasal nicotine solution (NNS), and chewing nicotine gum. Doses of nicotine were 2 mg for nasal nicotine solution and nicotine gum and averaged 1.97 mg for the cigarette. Conversion: SI to traditional units - Nicotine: 1 nmol/1 ::= 0.16 ng/ml. IReproduced with permission from Russell et al, 1981 (3).] 168 1 169

suggested they continue with I mg dose or combine it with or transfer to 2 mg according to their preference. Daily records of consumption were kept. Four therapists were involved and each subject was seen four times over the course of the first two weeks of treatment. These visits provided an opportunity to complete rating scales, to take blood samples in order to determine nicotine levels, and to dispense supplies of NNS as needed. The rationale given to subjects for using NNS was similar to that for nico- tine chewing gum-viz. that since some of the problems of withdrawal are mediated by nicotine depletion, nicotine replacement by NNS should make it easier to cope with withdrawal. As with gum, we stressed that NNS could not quit smoking for them, but at best could supplement their own efforts. At the end of the two-week period, subjects were given the option of either continuing with NNS or switching to our routine clin- ic treatment with nicotine chewing gum if they preferred. No subjects elected to do this. Results Though the trial provided a wealth of information, I will only sum- marize the main points that emerged. About two-thirds of our subjects stopped smoking successfully in the short term, and the great majority of these used NNS and found it helpful. In the long term, one-third re- mained abstinent throughout one year of follow-up, a figure compara- ble to the results of clinic treatment with nicotine chewing gum, although our sample size was too small to establish the likely outcome of treat- ment with NNS with any accuracy. Patterns of NNS use displayed a num- ber of features of interest. As with nicotine chewing gum, not all sub- jects found it necessary to use it. Among those who did, however, a clear preference for the stronger 2 mg dosage soon emerged. Numbers of doses taken daily showed wide variation between subjects, ranging from over 20 per day in some to less than five in others. The overall average was about 8 doses of 2 mg per day among those who used it at all. Several of our subjects used NNS longer than six months. Ratings of NNS. Overall ratings were positive. Subjects reported that it substantially reduced their craving for cigarettes and helped them to refrain from smoking. They found it somewhat embarrassing to use in the company of others and rated it slightly on the unpleasant side of neu- tral_ These ratings presumably referred to mainly sensory aspects of NNS since, paradoxically, they reported that it was only slightly less satisfy- ing than their usual cigarettes. Withdrawal symptoms. We found no change from baseline in a num- ber of affective symptoms which are commonly exacerbated by smok- ing withdrawal. These included irritability, inability to concentrate, and feeling miserable. There was, however, a significant increase in hunger. Comparison with other studies in which people have abstained from cigarettes without nicotine replacement suggests that NNS may have sub- stantially mitigated the experience of withdrawal. Side effects. Minor problems of adaptation to NNS were very com- nnon initially, and our subjects typically reported irritation in the nose, watering eyes, and sneezing after use. These difficulties rapidly receded, although local irritation in the nose remained common. Although there were few reports of more serious unwanted effects, one subject vomited shortly after taking a dose and subsequently abandoned NNS. Overall, three-quarters of our subjects rated side effects as no problem or not scri- ous, and a quarter as unpleasant but bearable. Drug ef£eets. A calming effect from taking NNS was experienced by two-thirds of our subjects. Other drug effects reported included light- headedness and a euphoriant effect. The frequency of reports of positive drug effects from NNS contrasts with nicotine chewing gurn, and pos- sibly reflects the more rapid absorption of nicotine from NNS. Blood nicotine levels. Blood samples for assessment of nicotine from NNS were not taken in any particular time relationship to self-dosing. They were therefore "trough" levels, and where NNS consumption was low, the measure might have been taken some hours after dosing. The most marked feature of the data, not surprisingly, was the wide range in concentrations, going from nonsmoker values to plateau cigarette smoking levels. The overall mean was about 7 nghnl and this showed little change from Day I to Day 11. Samples taken later in treatment (at five months to one year) from four subjects who became long-term NNS users averaged over 20 ng/ml, and were similar to trough cigarette smoking concentrations. Comment We view the results of this trial as encouraging. Subjects found NNS helpful in abstaining from smoking, and side effects were not a serious problem. It was striking that several of our subjects began the trial with low expectations of success because of previous failed attempts, but received a great boost to their confidence when they found that NNS had a real effect in allaying withdrawal symptoms. Of course, our study 170 1 171

did not include controls, so it is riot possible to make any formal assess- ment of placebo factors. No doubt the positive expectations and en- thusiasm conveyed by the therapists may have played some part in the results achieved. It is nevertheless our view that nonspecific factors are unlikely to account for the larger part of the therapeutic effect. As could be expected with a new and untested product, our trial raised as many questions as it answered. Average blood nicotinc levels were low, although the data from our heavier users showed that concentrations sitni- lar to those from cigarette smoking can be maintained from NNS. This suggests that, as with nicotine chewing gum, many patients may tend to underdose. As well as evidencing a disinclination to take what is clearly seen as a drug, this also may indicate a need to develop a formulation of NNS that is less irritant and hence more acceptable. Using the inforrna- tion gathered on dosage preference, side effects, etc., we intend to con- duct a formal placebo-controlled trial of NNS with outcome assessed over a full 12-month period. We can then see more clearly the potential role of NNS alongside nicotine chewing gum in smoking cessation. Effects of NNS on nonsmokers In addition to its potential as a treatment aid, NNS is of interest as a research tool for studying nicotine's effects in nonsmokers. Previous work in this area has been hampered by the difficulty of finding an acceptable and effective means of delivering nicotine. Quite apart from ethical con- siderations, cigarettes arc of little value because nonsmokers find the smoke too irritating to inhale. Nicotine chewing gum may induce nausea. Nico- tine tablets have been used but these are a poor means of delivery be- cause swallowed nicotine is not well absorbed, and most of what is in- gcsted is metabolised in the first pass in the liver. Crushing the tablets and holding thetn in the mouth may allow some buccal absorption, but blood nicotine levels are likely to be low. We have tound that NNS provides a convenient method of administer- ing moderate cioses of nicotine to nonsmokers, and we have examined its effects on performance of a simple motor tapping task. These experi- ments are described more fully elsewhere (6), so for this volume, I will only summarize the main features of the results. Finger tapping is one of the simplest motor activities. It has been report- ed that tapping rate can be increased by drugs with stimulant actions and irnpaired by a range of conventional depressant drugs (1). In a double- blind placebo-controlled experiment, we found that a single 2 mg dose of NNS improved performance speed by about five percent. Adnunis- tration of the central cholinergic blocking agent mecamylamine prevented this effect. When subjects took one 2 mg NNS on an hourly schedule over a period of six hours there was a reliable increase in tapping speed after each dose with no evidence of acute tolerance. The results of these experiments indicate that when nicotine is ad- ministered to nonsmokers via nasal nicotine solution, it has appreciable and reliable effects on performance of a simple motor task. The fact that these substantial effects have not previously been well documented in the literature suggests that NNS may, for the first time, offer an effective means of nicotine delivery to nonsmokers. The way is now open to in- vestigate nicotine's effects on a wide range of cognitive and performance tasks. Conclusion Since studies of nicotine chewing gum established the value of tern- porary nicotine replacement as a treatment aid for cigarette smokers, there is now a growing interest in alternative routes of nicotine delivery. The buccal route is relatively inefficient, and nasal administration gives bet- ter absorption. Our preliminary studies of a nasal nicotine solution lead us to think that it has considerable potential both for experimental study of nicotine's effects and as a treatment aid complementary to nicotin~ gum. References 1. Hindmarch I. Psychomotor Function and Psychoactive Drugs. Brit. j Clin. Pharm. 1980; 10:189-209. 2. Russell MAH, Jarvis MJ, Feyerabend C. A New Age for Snufl? Lancet 1980; 1:474-475. 3. Russell MAH, Jarvis MJ, Devitt G, Feyerabend C. Nicotine Intake by Snuff Users. Brit. Med. J. 1981; 283:814-817. 4. Russell MAH, Jarvis MJ, Feyerabend C, Ferno O. Nasal Nicotine So- lution: A Potential Aid to Giving Up Smoking? Brit. Med..J. 1983; 286:683-684. West R, Jarvis MJ, Russell MAH, Feyerabend C. Plasma Nicotine Concentrations From Repeated Doses of Nasal Nicotine Solution. BYit. J. Addict. 1984; 79:443-445. West RJ, Jarvis MJ. Effects of Nicotine on Finger Tapping Rate in Nonsmokers. Pharm. Biochein. and Beh. In press. 5. 6. 172 1 173

Clonidine and Cigarette Smoking Withdrawal Alexander H. Glassman, M.D. The Department of Clinical Psychopharmacology, New York State Psychiatric Institute and the Department of Psychiatry, College of Physicians and Surgeons, Columbia University New York, New York Fay Stetner, M.S. Pamela Raizman, M.S.W. Introduction There are three pieces of good news: first, I am not going to talk about nicotine chewing gum; second, I am the last speaker for today; third, this stuff may even work! The "stuff' I refer to is clonidine. Clonidine is an antihypertensive drug which has been in wide use for almost two decades. In 1978 investigators at Yale showed that clonidine was useful in suppressing symptoms associated with opiate withdrawal (4). They suggested that clonidine tnight be useful in the relief of cigarette withdrawal syrnptoms, as well as the withdrawal symptoms of almost every other dependence-prodtrcing drug. But perhaps because there are very marked differences between opiate and cigarette stnoking withdrawal symptoms, it was some time before the use of clonidine in smoking ces- sation was invcstigated.Wit became work werwer adoingcw th bulimic wc In~ smoking cessation as a re en, women who were biig c Satstemai bulinlianWcVth o ght there ni ght thee role of the noradre.nc g y stem bc similarities in the way bulimia and smoking influenced thisWs e rea~ such that smoks ogcTchd ty existed, itd should be diminished by clonidine soncd that if th Methods tcks We recruited 21 hcav smo y smokers with da t and averaggCd almost t~~ o p t eY kcd a minimum of one pack y the~~t a day; they had withdrawal symptoms in the morning that forcc attctnptcd cd out' to smoke before they had breakfast; they had all previous y to quit and had failcd~ O b~i ~csix f thisecxperimentua ad "`'o pd go ~to four started on the p g t further because they could not get through the experimental placebo day. One person dropped out because she became sleepy on clonidine, and one person on clonidine did not return because of events in his own life. Because clonidine, in addition to affecting blood pressure, is a very sedating substance, our design included comparison of three treatments: clonidine, a sedative control, and an inactive placebo. We used the seda- tive alprazolam, a benzodiazepine-like sttbstance very similar to valium, as placebo. Each subject was randomly assigned to one of the three treat- ments on the first experimental day and returned to the lab on two sub- sequent experimental days for the remaining two treatments. The three treatments were separated by at least three days of normal smoking. Usually it was easier for most subjects to return to the lab on the same day each week, thus there were usually six days of normal smoking be- tween each treatment. The subjects were instructed not to smoke after they went to bed and to report, without smoking, to the lab at 8:30 the next morning. A se- ries of baseline measures were made at 8:30, then at 9:00 subjects were given their first dose of medication. Vital signs and a series of self-ratings were monitored every hour; data are available for the next seven hours of each experimental day. A second dose of the test medication was given an hour and a half after the first if the vital signs were stable. At the end of each experimental day, the patients were asked to make a 0obal rat- ing of how much each medication (A, B, or C) helped them not to smoke by marking a 100 mm line. To avoid crossover effects, subjects were asked to resume their normal smoking pattern after each experimental day. Results The global ratings of the difficulty of not smoking under each of the three experimental conditions were compared by analysis of variance using a randomized block design. Fifteen subjects, two men and 13 women, completed all three treatments. Thirteen subjects noted that they clearly preferred one of the drug conditions over the inactive placebo. Of the two subjects who did not prefer active medication, one showed only modcst withdrawal symptoms in the placebo condition and could not distinguish between the drug and placebo conditions, in spite of a his- torY of smoking more than two packages of cigarettes per day. The other subject who did not prefer drug treatment manifested marked withdrawal sY»lptoms during the placebo period and experienced no relief from either d,ug. This subject also experienced very little pulse or blood pressure effect ftom clonidine. Overall, clonidine (p < 0.001) and alprazolam (p < 0.05) "gnifrcantly reduced the severity of craving when not smoking (Figure 1). 175 174

COMPARISON OF THE EFFECT OF THREE DIFFERENT TREATMENTS ON "CRAVING" FOR CIGARETTES 8 7 4 ** 3 1 0 Clonidine 0 1 2 3 4 5 Time (hours) 1 6 1 7 Figure 1. Average hourly ratings of craving. Conidinc is significantly better than placebo (p <.05)** (p < 0.1)*. Alprazolatn is never signifi- cantly different from placebo. There is a trend for clonidinc to be better than alprazolain (p < .1) 0. 176 Of the 13 subjects who found drug treatment effective, ten preferred cloni- dine (binomial calculation, p = 0.046). For anxiety, irritability, concentration, and tension, both drugs were clearly better than placebo and essentially identical to each other. Even the two most common side effects, drowsiness and dizziness, were report- ed to the same degree. Only the scale measuring craving (thoughts about or the wish to smoke) reflected the strong tendency in the global ratings for subjects to prefer clonidine. Here clonidine was rated as significantly more effective than both placebo (p < 0.01) and alprazolam (p < 0.05). Alprazolam was not significantly more effective than placebo. Previous studies of the smoking withdrawal syndrome have shown that craving is the most consistently observed withdrawal symptom and that it tends to be least intense in the morning but increases as the day progress- es (10). We plotted the hourly rating for craving during the first seven hours of treatment (approximately 08:30 to 15:30). Craving among sub- jects given placebo increased during the course of the day while those on clonidine experienced a decrease. As a result, the contrast between clonidine and placebo increased as the day progressed (Figure 1). Discussion The observed effect of clonidine on cigarette smoking withdrawal'raises the issue of how this effect occurs. This implication of clonidine's activity is particularly interesting because clonidine has a relatively restricted area of activity in the central nervous system. The nerve fibers in the body resemble wires in that they transmit messages down their length elec- trically and one nerve fiber, in this sense, is very similar to any other. However, at the nerve ending, this resemblance to a wire disappears. At their terminals, nerves transmit their message not electrically but chem- ically. The electrical activity of the nerve fiber, when it reaches the nerve terrninal, results in the release of a small packet of chernicals, and it is these chemicals that stimulate the next structure, be it a muscle, gland, or another nerve. This quality of nerve conduction allows a drug to affect certain nerves in the brain and not others. If nerves conducted impulses only elec- trically, a drug that affected one nerve would affect all. But because the brain uses several dozen different transmitter substances, it is possible for drugs to affect one system or groups of systems, and not others. Certain transmitter substances such as gama amino-buteric acid (GABA) are very widely distributed and can be found in almost half of the cells in the brain (3). Acetylcholine is another widely distributed transmitter substance. The cells that contain biogenic amines (serotonin, norepinephrine, and dopa- n1i11e), in spite of being well known, represent a very small percentage 177

4~! of the cells in the brain (1). Norepinephrine involves only one quarter of one percent of all the cells in the brain. Clonidine affects only fibers that use norepinephrine as a transrnittcr substance. A drug can attack this chemical transmission system in a number of different ways. It could act on the process of synthesis of these transmit- ter substances; it could act on the storage granules that hold these trans- mitters in the nerve ending; or it could affect the release process of these transmitter substances. These functions all occur in the presynaptic cell that produces the transmitter substance. Drugs could also act on the post- synaptic cell where the transmitter interacts with the receptor by block- ing, or in some way altering, the characteristics of that receptor. Under normal conditions, the transmitter substance is removed from that post- synaptic receptor and reabsorbed by the presynaptic fiber in a process called re-uptake. The tricyclic antidepressants in particular affect trans- mission by altering this process. An additional part of this chemical trans- mission system is a group of receptors for the transmitter substance lo- cated not on the postsynaptic cell, but on the same presynaptic cell that produces the transmitter. These are called auto-receptors. These auto- receptors, in general, act as a brake on the production of the presynaptic fiber transmitter substance. A cell that is actively firing and releases large amounts of its own transmitter substance, will, in essence, tend to decrease its own firing rate as the transmitter substance diffuses back onto its own auto-receptors. Clonidine has particular selectivity for these auto-receptor sites (6). At these sites, clonidine acts as an agonist and decreases the ac- tivity of the presynaptic cell, shutting down the noradrenergic system. Not only are the number of noradrenergic nerve fibers limited, but also the geographical distribution of their cell bodies (5). All the norepincphrinc cell bodies in the brain are located in the brain stem, either in the medulla or the pons. There is a ventral and dorsal string. The most anterior dorsal nucleus is called the locus ceruleus, and contains about half of all the norepinephrine in the brain. It is visible to the naked eye in a cross section of hutnan brain and is a likely site of action for clonidrtre- What does this cluster of noradrenergic cells normally do? Ncuroscieu- tists have been actively working on this issue. There are several points about which there already is agreement, and one of them is that this lo- - cus plays a major role in attention or vigilance (2). Another point o agr« ' ment is that it is very stress-scnsitive (7). Locus cells arborizc widely into other areas of the brain. At these "c~~~` endings the released norepincphrine can activate either beta or alpha I synaptic receptors. Some of the postsynaptic alpha receptors ~~~~~hy~ 2 types, but these are a very small percentage of the total. Man 178 postsynaptic alpha-2 receptors are located in the prefrontal cortex. It is possible that clonidine would be active at this site, rather than on au- toreceptors back on the cell body. Another intriguing characteristic of this system is that the locus is thought to be, by a number of investigators, an alarm system (9). Some evidence to support this is that every drug that sedates human beings, whether morphine, alcohol, benzodiazepines, phenobarbital or clonidine, quiets the locus (8). In addition, those drugs that activate the locus cause anxiety in humans. This relatively small noradrenergic system is intimately involved with attention or vigilance and is stress-sensitive, and it may be involved in states of tension or anxiety. It can also be manipulated in such a way as to decrease smoking withdrawal symptoms. It is an in- triguing neurophysiological convergence. Summary We found that clonidine does decrease, in a rather dramatic way, report- ed short-term withdrawal symptoms from cigarette smoking and does this with an entirely different pharmacology than the nicotine-containing treatments. Though it is natural to ask which treatment is better, I think that is the wrong question. We are seeing the development of a.series of pharmacological agents that modify the underlying neurophysiology of addiction. We are witnessing the development of a variety of pharmaco- logical compounds that aid patients in withdrawing from cigarettes, and new agents of this type will be developed in the future. References 1. Amaral DG, Sinnamon HM. The locus ceruleus: Neurobiology of a central noradrenergic nucleus. Prog. Neurobiol. 1977; 9:147-196. 2. Foote SL, Bloom FE, Aston Jones G. Nucleus locus ceruleus: New evidence of anatomical and physiological specificity. Physiol. Rev. 1983; 63:844-914. 3. Glassman AH, Jackson WK, Walsh BT, Roose SP, Rosenfeld B. Cigarette craving, smoking withdrawal, and clonidine. Science. 1984; 226:864-866. 4. Gold MS, Redmond Jr. DE, Kleber HD. Clonidine blocks acute opiate-withdrawal symptoms. Lancet. 1978; 11:599-601. 5. Grant Sj, Redmond Jr. DE. The neuroanatomy and pharmacology of the nucleus locus ceruleus. Pp. 5-27 in H Lal, S Fielding (Eds.) Psy- ,hopharmacoloQy of Clonidine. New York: Alan R. Liss, Inc., 1981. ~. [saac L. Clonidine in the central nervous system: Site and mechanism 179

of hypotensive action. J. Cardiovasc. Pharrn. 1980; 2:S5-S19. 7. Korf J, Aghajanian GK, Roth RH. Increased turnover of tress: Role of the i d ng s ur norepinephrine in the rat cerebral cortex locus ceruleus. Neuropharm. 1973; 12:933-938. Majcwska MD, Harrison NL, Schwartz RD, Barker JL, Paul SM. Steroid hormone metabolites are barbiturate-like modulators of the t r Science 1986• 232:1004-1007. o 8. . GABA recep 9. Redmond, Jr. DE. New and old evidence for the involvement of a brain norepincphrine system in anxiety. Pp. 153-203 in WE Fann (Ed.) The Phenornenology and Treatment of Anxiety. New York: Spectrum, 1979. 10. Shiffman S. The tobacco withdrawal syndrome. Natl. Inst. Drug Abuse Res. Monograph Series. 1979; January 23:158-184. Discussion: Alternative Forms of Pharmacologic Treatments Chair: Ellen Gritz, Ph.D. jonsson Comprehensive Cancer Center, Los Anqeles Presenters: Jed Rose, Ph.D. Veterans Administration Medical Center, West Los Anqeles 1Vlartin Jarvis, M.Phil. Institute of Psychiatry, London Alexander Glassman, M.D. New York State Psychiatric Institute, New York City DR. GRITZ: Any questions? Pleasc come up. Saul, you're first. DR. SHIFFMAN: We heard in Dr. Hughes' talk in particular about the dependence potential of nicotine substitutes. And I think, Dr. Jarvis, your talk speaks to the issue of better minucking the natural dynamics ofsmok- ing. I wonder if theree isn't a conflict between, on the one hand, irnprov- ing the delivery so that it's a more effective substitute, a more effective treatment, and creating an increasingly greater problem of dependence. DR. JARVIS: I think you're right. There may well be a trade-off. I did briefly show nicotine levels in long-term users, and obviously it follows from that that we did have some people who were long-term users of the NNS. We've actually got one person who is still using it now. The other three ofour long-term users have successfully weaned themselves off. But I think you may well be right. It's something that we'll have to see as more data is gathered on this, it may be part of the trade-off in at- taining a more effective dosage that you have a greater dependence potcntial. Craving LYNN KOSLOWSKI: I'm from Toronto. I'd just like to make a point Ibout the use ofthe term, craving, as it's appeared a number oftimes to- (lay. it scetns to mc that by definition English requires that craving is an extreme desire for a cigarette, not thinking about a smoke, not wanting aligarette a little bit. I think that very often craving appears on the slides iso 1 181

and causes some mischief when you're dealing with fairly modest changes in desire to smoke or, indeed, thinking about cigarettes. I'd like to dis- courage use of the word craving unless one is really talking about crav- ing as it exists in the dictionary. It refers to an extreme desire for that thing, rather than thinking about it or othcr descriptions of readiness to smoke. FLOOR: I'd like to pursue that question a bit more. Dr Glassman, I've admired your work for some time. I hope you'll forgive a somewhat picky question. It seems to me there's a literature that suggests that the effect of clonidine on the locus ceruleus is to reduce the distractiveness of ex- traneous stimuli. It's sort of the opposite of increasing central noradrener- gic activity in this particular set of pathways, which sharpens the focus- ing of consciousness. The particular question you phrase as your dependent measure is: how preoccupied are you with smoking? And I'm wondering if you can put those two things together. by taking into ac- count what Lynn Koslowski has been saying about the definition of the term "craving" I'm wondering if perhaps your effect is, indeed, a cloni- dine effect, but not necessarily an effect on craving as it is typically used in the smoking literature. DR. GLASSMAN: It's an interesting question. I must say it's an im- pressive question. In other words, does the pharmacological activity of clonidine on the locus, which has a vigilant function, in fact, change the nature of craving, and that's all we're observing? I don't have a single ex- periment to address the question that you're asking. What I've done is to use clonidine to withdraw people from cigarettes, and the only thing I can say in defense of the point you're raising is that people still quit smok- ing when they take clonidine. I don't think this is a good forum for us to argue because it's a very complex issue about what clonidine does to attention. I think in terms of the earlier comment about our use of the word "craving," I find the whole thing problematic. Basically we're dealing with words, and the real question is: what's the underlying physiology and without any handle on that, it's difficult to decide which is the right word. I would be bothered by a system that used the word craving only to refer to a prcoccupying obsession with smoking. On the scales that I showed you, when people got an eight, they were so preoccupied with thoughts of cigarettes that they really couldn't work. They couldn't func- tion. They couldn't sit down and do anything anymore. And at 10, they got up out of the experinicnt and they smoked. And I have no question that I think everybody in this room would have agreed that 4, 5, 6, 7, and 8 on that scale were all real craving. What are we measuring when one or two people say they only have transient or a little bit of thought about smoking or someone on that scale was a zero, the thought about stnoking never crossed his or her mind? I won't argue with you about what's the best definition of craving. I have thoughts about the under- lying neurophysiology, which may be simplistic, too, and prove not to be true, but they would require that there be very modest levels of some sort of thinking about smoking, about an object that reduces tension. I could see reasons for defining it in other ways. I wouldn't quibble with you about words, but I think there are merits to recognizing mild degrees of craving. Nasal nicotine solution (NNS) trials DR. SACHS: I have what I think are two very quick questions. One to Dr. Jarvis. How long overall were people able to use the NNS? Was there an average length of time that they used it or was it ad lib, and, if so, what percentages roughly were still using it at six months and a year later? DR. JARVIS: They were able to use it ad lib. I don't have the exact per- centages, but the durations of use were pretty similar to what we've seen with nicotine gums. So, most of them who did use it stopped between three and six months. As you saw in the data, there were some people who had already stopped by the end of the trial period, but the qiajority used it for longer than that. DR. SACHS: And what was the behavioral treatment component? DR. JARVIS: They were seen individually by four of us: Mike Russell, Peter Hajck, Robert West, and myself; and they were given supportive treatment of the kind we offer with nicotine gutn. So, it was really much the same as the kind of treatment we offer with gum, except offered in- dividually. Clonidine and smoking cessation DR. SACHS: Right. Very good. Dr. Glassman, do you have any out- come data on the use of clonidine in terms of producing smoking cessa- tion or does anybody that you know of? DR. GLASSMAN: If you're talking about double-blind placebo- controlled data, I don't have any and I don't know anyone who does. DR. SCHNEIDER: Dr. Glassman, how do you account for the role of nicotine and theti nicotine replacement in this low-dose fashion that We've been discussing with quieting of the locus ceruleus activity? DR. GLASSMAN: No drug in pharmacology is entirely clean. They're all to some extent dirty. Nobody's found a perfect drug. But clonidine is relatively clean as drugs go. It has a very primary effect on this 182 1 183

nonadrenergic system. I didn't present the data, but we have done dose- response curves, and the response of this craving effect occurs at very low doses of clonidine. It is suggestive that this is a pure alpha-2 effect of the drug. Now, that alpha-2 effect could be in that locus, it could be in A-2 or A-4, those clusters of cell bodies that precede the locus in the brain-ose stem. It could be ~~` a np~~e pof ~cations wdhen you of any of he num- but that's a rather ber of cells in the brain. You're then asking the question: if decreasing the firing of these cells by using clonidine helps in cigarette withdrawal, what relationship does that have to what nicotine does to these cells? I think that awaits real ex- periments with nicotine on this noradrenergic system. Torgis Svenson has done some IV infusion of nicotine and done recordings from locus, but I believe that those are all acute experiments. They're not like a smoker who chronically takes nicotine. DR. GRITZ: Neal. Skin absorption of nicotine DR. BENOWITZ: I have a quick comment for Jed Rose and then a cou- ple of quick questions for the other speakers. With respect to skin ab- sorption, I wanted to relate an experience of an interesting patient that I saw, who was a young woman who had scabies and in self-treatment she doused herself with Black Leaf 40, which is a solution of 40 percent nicotine. And, of course, she became extremely sick and vomited and was hly somewhat lrypotcnsiv She to b~shk forpeighteen hours~Shegwas dccontaminated, but c rams maintaining a nicotine le ~oursaboeut a 9er shcgwas d eont rrr~mated~So, for all twelve to eighteen leach my intcrprctation of it is °iatnf o wouldrrepeat this kind of dosing out very slowly. It may be t you on a daily basis or a long- das•and daysoand days1all hink that's somel and is slowly released over y begin thing to keep in nrindwhh~you' kind c fep olong d effecottof a h b at t t s vc n ~ DR.ROSE:Sofa v single application at these lo allo of thesc dose patient ootherthmg have exceedcd ~ t lf she may se to keep in mind, as you've shown ~ boyour l sn°c uld work, alecount for some dividual variability in nicotine a of the differences in the persistence of nicotine. Do you have any idea w at her metabolic rate was? DR. BENOWITZ: No. But just the fact that she maintained a constan 1R4 level for more than 12 hours suggested to me that she has a continued absorption phenomenon. Martin, you showed a tremendous variability with the nasal solution, and there you're delivering presumably a known constant dose each time. I was wondering what was the source of the twcnty-fold variability? DR. JARVIS: Well, I think it's partly the number of doses people are taking. Obviously, that's one factor. There was a correlation between the nicotine level that they got and the number they used. But I think more importantly probably, we've got a lot to learn about the technique of self- dosing with this. You need to sniff it in the right way and get it not too far back in the throat and not too peripheral. If you get it too far back, it tastes horrible and it's not very pleasant. i think the people who used it successfully and went on long-term with it clearly acquired a satisfac- tory technique. Others maybe didn't persist and didn't acquire that tech- nique. That's really the only explanation that I can offer for this very tnarked variability. Certainly in the repeated dosing study to which I brief- ly alluded, we did find that people who had had previous experience in using the NNS and had used it irregularly over a period got more nico- tine from it than people who were naive users, which certainly does sug- gest that there's a skill to be learned. Just as, indeed, there's a skill to be learned in chewing nicotine gum in order to use that efN6ctively. DR. GRITZ: And in learning to smoke, as well. DR. JARVIS: Yes, indeed. Clonidine and nonabstinent smokers DR. BENOWITZ: One last brief question for Dr. Glassman, and that is: for people who don't stop smoking, does clonidine have any effect on spontaneous smoking behavior? Do people smoke less? DR. GLASSMAN: I said that people were intrigued with the thought that nicotine would work in anything that could be described as a with- drawal syndrome, when Mark Gold originally published that material. Bochringer Ingelheim, owner of clonidine, looked at their hypertension studies where they had recorded whether people smoked or didn't smoke. They looked to sec if the rate of smoking decreased in the people who were on clonidinc as compared to placebo, and found no change. Actu- ally, that's our experience as well. Ifyou don't get people to stop smok- tt1g, the clonidine doesn't do any good. Clonidine is only useful when people really stop. It interrupts the withdrawal symptoms, but if they're uot withdrawing, it doesn't do any good. DR• GRITZ: I thank you all for this second half of the afternoon. 185

v: An International Update on Nicorette

An International Update on Nicorette A panel discussion entitled `An International Update on Nicorette" was convened at the World ConXress on the Pharmacologic Weatment of Tobacco Dependence on Monday afternoon, November 4. Four researcher:c from around the world present.ed data from studies using nicotine gum as an aid in smoking cessation. Moderated by Karl-Olov Faqerstrom of Sweden, the panel included 'I'eresa Salvador- Llivina of Spain; Renee Bittoun of Australia; Marcel Kornitzer of Belgium; and Philip 7i)ennesen of Denmark. Each panel member gave a short presentation and then the audience was invited to ask questions or make comments. What follows is a sum- mary of the panel members' presentations and the subsequent discussion. Teresa Salvador-Llivina, Ph.D. Health Promotion Service, Public Health Service of Catalonia and the Smoking Cessation Programme, Hospital Clinic i Provincial de Barcelona Barcelona, Spain Teresa Salvador-Llivina discussed smoking cessation in Spain and presented findings from a 1983 comparison study of a support-only treat- tuent and a support/gum/feedback treatment. She noted that smoking ces- sation in Spain lags behind the United States, Canada, Central Europe, and Northern Europe, but Spanish health professionals now accept cigarette smoking as the major cause of many cardiovascular diseases. Not only must health professionals counteract the tobacco industry's adver- tising campaigns, they also must deal with a tobacco industry which is a powerful state monopoly enjoying econoinic and technical support from the Ministry of Agriculture. And although isolated attempts to carry out smoking prevention/cessation campaigns have occurred at both the na- tional and autonomous (state) levels, there is as yet no comprehensive smoking control policy or agency at the federal level which implcmeuts price increases, regulates indirect taxation, enacts legislation to promote long-tenu health education programs and provides for smoking cessa- tion services to smokers who wish to quit smoking. ~~tlish I~r. Salvador-Llivina reported that recent data show 41 /o of the SI ~ population stliokc regularly. She C~aihtiia,c52`%~tof the doct ~rs s T k crs is exceeded among doctors: in And among Spanish chest physicians, the medical group most awa~~~~t the }1ealth hazards of sma'k~' int~rn~tionalt health )agcncicsoklv° atc for cd that many nationa smoking cessation services, including the World Health Organization, the International Union Against Cancer, the Royal College of Physicians of London, and the United States Surgeon General. Dr. Salvador-Llivina presented data from a comparison study between two treatments for tobacco dependence: the first used support only, and the second used support, nicotine chewing gum and feedback. In a clin- ic setting, 339 subjects began the non-randomized study, both men and women with mean consumption of 25-26 cigarettes per day, and mean years smoking 23. Seventy-fivc percent of the subjects had been smok- ing for more than ten years and had tried to quit several times without success. No smoking related diseases were evident in any subjects. Sub- jects were divided into the two treatment groups with a mean of ten sub- jects assigned to each therapy support group. A total of 187 subjects received the support-only treatment; 152 subjects received support/ gum/feedback treatment. Treatment A consists of an initial individual session of approximately 45 tninutes during which the subject's history as a smoker is taken, the treat- tnent is described, and the subject is given a card on which to record smoking patterns in the days prior to quit day. After quit day, Treatment A subjects attend eight group support sessions in one month's time. Treatment B con- sists ofan initial individual session ofapproximately 45 minutes during which the subject's history as a smoker is taken, the treatment is described, and the subject is given a card on which to record smoking patterns in the days prior to quit day. After quit day, the Treatment B subjects attend five group support sessions in one month's time, use nicotine chewing gum, and receive feedback from the support group leaders and participants. Monthly follow-up was conducted for one year and abstinence was moni- tored at one, three, six, nine and twelve months. Reported abstinence was verified with CO levels using an ecolyzer. Of the 187 subjects who began Treatment A (support only), 100 sub- jects dropped out of the study leaving 87 to complete treatment. Of the 152 subjects who began Treatment B(support/gum/fecdback), only 20 dropped out and 132 subjects completed treatment. Of the support-only group, 55 subjccts (63%) were abstinent at the end of the one-tnonth treatment; of these, only 16 subjects (18`X, of subjects beginning treat- 'nent) were abstinent at the end of onc year. Of the supportlgttm/feedhack group, 112 subjects (85%>) were abstinent at the end of the one-month trtaqncnt; of these, 62 subjects (47'%, of subjects beginning treatment) wcre abstinent at the end of one year. Sttbjects receiving support/gum/fecdback achieved a substantially better °utcome as regards abstinence both at the end of treatment and at one- 188 1 189

year follow-up than did subjects receiving support only. At one-year follow-up, the success rate for support/gum/feedback subjects is more than twice that for support only. Fewer subjects dropped out of the sup- port/gum/feedback group, and attendance at follow-up monitoring ses- sions was notably higher for these subjects. Dr. Salvador-Llivina reported that of the 132 subjects receiving support/gum/feedback, nine subjects did not use nicotine gum: one subject who did not want it quit smoking without gum, and eight subjects were excluded because of reported previ- ous gastric problems. Four percent of subjects completing treatment with the gum chewed gum for more than one year, and all of these subjects weaned off gum before 14 months of use. A total of 57 subjects used the gum for more than three months; of these, 85% were still abstinent at one-year follow-up. Eleven subjects stopped chewing the gum during the first week of treatment; of these, only one subject remained absti- nent at one-year follow-up. During the first month, mean consumption of the gum was eight pieces per day. Dr. Salvador-Llivina concluded her talk by saying she and her col- leagues are currently conducting a double-blind, placebo-controlled trial using nicotine gum, the results of which will be available in the future. Renee Bittoun, Ph.D. Director of the Smokers Clinic St. Vincent's Hospital, Sydney, Australia Dr. Bittoun outlined for the audience the state of smoking cessation in her country and presented data from a cessation trial using rapid in- halation and rapid puffing techniques. Dr. Bittoun noted that more Nicorette is sold per capita in Australia than any other country in the world. She described an anti-smoking campaign conducted by the government which is carefully evaluated for effectiveness, and is carried out both in major newspapers and on prime time television. During the past three years, anti-smoking advertisements have been directed at cer- tain target groups. The first year, ads were directed toward the general smoker; the second year towards women smokers; and the third year towards adolescent smokers. Dr. Bittoun reported that, despite a 400 per- cent increase in tobacco advertising, people are "coming in droves" to ces- sation clinics to get help in quitting. She noted that the impact of the anti- smoking ad campaign is very good and "the tobacco industry is not ap- preciativc." Dr. Bittoun reported that she and Drs. D. C. Clarke and D. H. Bryant use nicotine gum in conjunction with other techniques, specifically rapid inhalation and rapid puffing. The technique of rapid puffing was 190 developed for patients with chronic obstructive airways disease for whom rapid inhalation would present health risks. Rapid puffing is the same proce- dure as rapid inhalation except that the smoke is not inhaled into the lungs, but rather puffed in and out of the mouth. Dr. Bittoun presented results of a study to determine the influence of patient information about nico- tine gum on the outcome of gum therapy. She and her colleagues conducted a trial in which smokers were randomly divided into four groups: a group of rapid inhalers prescribed 2 mg nicotine gum and given proper instruc- tions as to the use and effects ofthe gum; a group ofrapid inhalers prescribed 2Ing rucotine gum and given instructions about the use of the gum and °counterdemand" information about the effects ofthe gum; a group ofrapid puffers prescribed 2 mg nicotine gum and given proper instructions as to the use and effects ofthe gum; and finally a group ofrapid puffers prescribed 2 mg gum and given proper instructions about the use and counterdemand information about the effects ofthe gum. No placebo gum was used. The counterdemand information given to smokers was that the nicotine gum would have no effect for two weeks, but that they should continue to use it. Smokers in the trial had no previous information about nicotine gum because the product had not been available to the public in Australia. No evidence of hypertension or cardiovascular disease was present. All patients were under 55 years old and had blood pressures ofunder 160/100 rnmHg. There were 140 subjects in this study. After three months, Dr. Bittoun reported that among smokers using the rapid inhalation technique, 41.8% of those given proper instructions about the use and effects of nicotine gum were abstinent while 22.2% of those given the "counterdemand" instruction were abstinent. Among smokers using the rapid puffing technique, 62.5% of those given proper gum instructions and information were abstinent after three months while 35.3% of those given the counterdemand information were abstinent. Overall, 51.8% of the smokers given proper instructions and informa- tion about the gum were abstinent after three months; 28.8% of smok- ers given the counterdemand information were abstinent after three months. Abstinence was biochemically validated. Dr. Bittoun conclud- ed that information given to smokers was critical to the outcome. Marcel Kornitzer, M.D. Universite Libre de Bruxelles, Faculte de Medecine et de Pharmacie, Laboratoire d'Epidemiologie et de Medecine Sociale Brussels, Belgium Dr. Kornitzer presented results of a study to compare the abstinence rate at one year between males using 2 mg or 4 mg nicotine gum in a 191

double-blind randomized trial among hcavy smokers. One hundred and ninety-nine subjects were recruited at a factory by means of a poster which read: "You smoke at least 15 cigarettes a day and you want to quit. Con- tact your factory physician today. He can help you." There were no sig- nificant differences in baseline characteristics for the two groups regarding age, number of cigarettes smoked per day, quit attempts prior to trial, Fagerstrom addiction scores, weight and serum thiocyanate. Subjects were randomly assigned 4 mg or 2 mg nicotine gum. Dr. Kornitzer conduct- ed baseline examinations of each subject in which weight was measured and blood samples for thiocyanate were taken. Each subject also received instruction about how to use the gum. At 14 days after quit day, sub- jects received reinforcement from the factory physician. During the en- tire study, subjects had unlitnited access to nicotine gum, chewing as much as suited them. At three months, subjects could elect to chew either 2 mg gum or 4 mg gum, regardless of which dose they had been assigned at the outset. Dr. Kornitzer reported abstinence rates at three months of 36.2% for subjects using 2 mg nicotine gum and 44.8% for subjects using 4 mg gum. At one year, he and his colleagues found abstinence rates of 22.3% for 2 mg gum users and 32.2% for 4 mg gum users. The abstinence rates of smokers with a baseline Fagerstrom addiction score greater than 5 were analyzed separately. At one year, 18.5% of these subjects using 2 mg gum were abstinent and 32.9% of these subjects using 4 mg gum were ab- stinent. Philip Toennesen, M.D. Department of LunA Medicine, AMTS Hospital of Copenhagen Lyngby, Denmark Dr. Toenncscn reported the findings of a double-blind, placebo- controlled dose-response study. Dr. Toennesen and his colleagues exam- F..~ ined the possible dose=response effect from 2 and 4 mg gum used in com- ~ bination with group meetings. One hundred and seventy-seven patients to ~ ~ F-+ p in a chest clinic attended group discussion meetings. All were told to s smoking at the first meeting and to use nicotine gum for three to four months. The series ofineetings, held at weeks 1, 2, 6, 12, and 16, included videotaped instruction with information about cigarette withdrawal symptoms, effects of nicotine chewing gum, and detailed chewing »n- structions. At week 12, subjects were shown another videotape which focused on the long-term benefits of quitting smoking. Abstinence was confirmed by CO levcls in expired air. Subjects were asked to keep a di- ary of their daily withdrawal symptoms and daily usc of nicotine gt'm- 192 The subjects were divided into high dependent and low dependent smoker groups according to a modified Hall and Russell questionnaire. Smokers with a total score of 19 or more were classified as high depen- dent, the others as low dependent. High dependent smokers were ran- domly assigned 4 tng or 2 mg gum; 33 used 4 mg gum and 27 used 2 mg gum. Low dependent smokers were assigned to 2 mg gum or place- bo gum; 61 used 2 mg gum, 56 placebo. After six weeks, Dr. Toennesen reported finding success rates with high dependent smokers of 85 % for those using 4 mg gum. For high dependent smokers using 2 mg gum, Dr. Toennesen reported abstinence rates at six weeks of 58%. Among these same high dependent smokers, at 24 weeks 4 mg users were 59% abstinent while 2 mg users were 33% abstinent. In the low dependent smokers group, Dr. Toennesen said quit rates with the 2 mg gum were significantly higher than with placebo gum: at six weeks, 74% of smokers using 2 mg gum were abstinent, while 41% were abstinent with place- bo gum. After 24 weeks, Dr. Toennesen reported a 49% abstinent rate with 2 mg gum and a 21 % abstinent rate with placebo gum. Dr. Toennesen noted that in a matched non-intervention, 56 persons received advice only from their physicians to stop smoking. Their suc- cess rates were considerably lower in relation to the treatment groups: after six weeks 12 % of these smokers were abstinent and after 24 weeks only six percent were abstinent. He concluded that there was a signifi- cant dose response effect among high dependent smokers using 2 mg gum and 4 ing gum and a significant effect among low dependent smokers using 2 mg gum and placebo gum. And, finally, he noted that treatment was considerably more successful than advice only. Discussion Nina Schneider, University of California at Los Angeles, posed a ques- tion to Dr. Bittoun concerning the expectancy effect of counter infor- niation to patients using nicotine gum. "Did the people who received counter information reduce their intake of gum as a consequence and was that responsible for the difference in efficacy of the gum, or is just ex- Pecting the gum not to work causing it to be ineffective and not allevi- ate withdrawal symptoms?" 1:)r. Bittoun replied that patients given the cOunter information had used the same amount of gum as those given proper information. "Our feeling is that it's the expectancy, absolutely:' she said. An unidentified speaker asked L)r. Kornitzcr to describe the specific interventions he and colleagues made in his factory trial. He replied, "I used a flipovcr with about 12 diagrams and figures and I explained the 193

toxicity of cigarettes and the psychosocial component of smoking and nicotine dependence. I also explained how to use the gum, how to chew the gutn, and side effects of the gum. This took about 30 minutes total. Then, the factory doctor had only to make reinforcements at two weeks and three months:' Neal Benowitz, Chief of the Division of Clinical Pharmacology and Experimental Therapeutics at the University of California, San Francis- co, School of Medicine, observed that biochemical assessment techniques had changed such that measurements of thiocyanate and carbon monoxide were no longer preferred measures for nicotine intake. Both thiocyanate and carbon monoxide had been mentioned by international panel par- ticipants. Dr. Bcnowitz noted that recent studies have shown that patients with higher cotinine levels pretreatment as well as higher scores on Fager- strom's scale and dependency scales, benefitted most from nicotine gum therapy. Patients with low cotinine levels benefitted no more from gum therapy than they did from ordinary behavioral therapy. Dr. Bcnowitz also said that using cotinine levels allows for mid-treatment assessment of nicotine levels from gum chewing so that dose can be adjusted, if neces- sary. Both Dr. Fagerstrom and Dr. Kornitzer, of Sweden and Belgium respectively, responded by saying that although measuring cotinine levels made a great deal of sense, finding laboratories which routinely conduct cotinine level testing was extremely difficult for the general practitioner in their countries. VI: Integrating Pharmacologic and Behavioral Approaches 194

Overview: Integrating Pharmacological and Behavioral Approaches to Smoking Cessation Saul Shiffman, Ph.D. University of Pittsburgh Pittsburgh, Pennsylvania Science fiction scenario: 1978 - Nicotine gum is introduced as a treatment for smoking cessation. After years of making do with marginally effective behavior modification approaches to smoking cessation, the world has been waiting for a "magic bullet." These hopes are finally met by the gum as it eliminates smoking painlessly and effortlessly. Stung by the com- petition, behavior modification specialists denounce the use of nicotine sup- plementation and refuse to incorporate it into smoking clinics. Physicians prescribe the gum for nearly all their smoking patients and most quit suc- cessfully and permanently. Soon, smoking disappears as a public health problem. Obviously, this scenario has not come to pass. Nicotine gum has not proven itself to be the "magic bullet" hoped for by some. Many physi- cians do not prescribe it, while others do not prescribe it properly; pa- tients often do not use it or do not use it properly. When the gum is prescribed without additional intervention, even among motivated volun- teers, failure rates of over 85 % have been reported (1,3) [results with un- selected patients are often only half as good *(9)] In contrast, programs combining nicotine gum with behavioral interventions in motivated smokers report six-month quit rates as high as 50%-63% (2,4,11). The effects of pharmacotherapy are most impressive when it is combined with behavioral intervention and with those elusive patient qualities- commitment and effort. *This may at fir.ct seem a contradiction to some of the published trials, which report- ed much higher success rates. In many of these trials, however, nicotine was com- pared to placebo against the background of a multi-component behavioral intervention. The success rates implicitly attributed to nicotine were actually due to the combina- tion of nicotine pharmacotherapy and behavioral intervention. Shall we conclude that nicotine gum has failed to live up to its promise? Shouldii t behavioral treatment be irrelevant i f the drug itself works? Only if we expected it to be a self-contained, complete cure -a "penicillin" for smoking-should we regard nicotine gum as a failure. The introduction of any new pharmacologic agent arouses extravagant expectations and from this naive "magic bullet" perspective, nicotine gum has achieved only qualified success. If instead of viewing nicotine gum as a sufficient cure for smoking, we view it as a new and powerful addition to the clinician's armamen- tarium, then pharmacotherapy of smoking has been quite successful. Combined with behavioral interventions, the gum can nearly double the success rates achieved without it (4,11). In one of the most important studies to date on the clinical use of nicotine gum, Schneider and her col- leagues (11) showed that although the gum had little effect in a "dispen- sary" condition that provided minimal support or treatment, it was dra- matically effective when combined with behavioral treatment. The question, then, is not whether nicotine therapy should be combined with psychosocial interventions, but how. The papers in this chapter present examples of how pharmacotherapy can be creatively combined with behavioral interventions to maximize long-term success. In this brief overview, I will concentrate on a few issues that arise in attenipting in- tegrations of pharmacological and behavioral treatment. Historical background for an integrative approach As Edward Lichtenstein points out, nicotine gum came on the scene at a propitious moment in the development of behavioral treatments for smoking and has found a receptive audience among practitioners of be- havior modification. Several factors contributed to this receptivity. Behavior therapy has recently become less ideological and more prag- matic. One hears few worries about whether the use of nicotine sup- plementation is consistent with behavioral theory. The philosophy, con- gruent with behavior therapists' long tradition of empiricism, is: "If it works, use it." Despite some impressive advances, behavior therapy has also not produced a "magic bullet" for smoking cessation. Relapse con- tinues to be a major problem and success rates seem to have reached a plateau at about 40%, even for the best programs (8). These trends have themselves prompted a shift towards multi- Component treatment. While initial attempts to modify smoking behavior were often based on a single theoretical principle, it is now common for treatments to incorporate multiple principles and approaches, typically without rnuch concern for their theoretical compatibility. In this context, 196 1 197

nicotine gum has been readily accepted as yet another ingredient in an evolving recipe for effective smoking cessation. Another trend contributing to nicotine gum's positive reception is the increasing recognition that biological factors play a role in behavioral problems. The advent of behavioral medicine has popularized the "bio- psychosocial model" of behavior, which strives for an integration of bio- logical and behavioral factors in understanding and facilitating a change in behavior. A final factor promoting acceptance of pharmacotherapy has been the development of a public health perspective on smoking control, with an attendant recognition of the magnitude of the smoking problem and the limitations of traditional behavioral approaches. We are increasingly aware that even if behavior therapy clinics were highly effective, they would make only a small dent in the national public health problem of smoking. One limi- tation arises out of the sheer number of smokers: one-third of all adults smoke in America and 38%-42% of adults smoke in Western Europe (5,13). Behavior therapy clinics simply don't have the capacity to treat the millions of smokers who need treatment (10,13). Secondly, we are discovering that relatively few smokers want to be treated in clinics. It has been estimated that 95% of those who quit smok- ing do so without clinic treatment, and only 36% of smokers express a willingness to attend a smoking cessation clinic (6,13). Thus, there is an evident need for simpler, cheaper, more easily distributed treatments for smoking, and nicotine gum seems to fit this criterion. However, the trend towards combining nicotine gum with behavioral treatment pro- grams can be at odds with the need for simple, cheap, easily disseminat- ed treatment, as I will discuss later. The atmosphere in the medical community has also been conducive to collaboration with behavioral approaches. Besides a new awareness of the role of behavioral factors in disease, the growing literature on non- compliance emphasizes the importance of behavioral factors even in purely pharmacologic treatments. Thus, the notion of collaboration and integra- tion of behavioral and pharmacologic treatment found fertile and receptive ground in which to grow. Modes of mutual facilitation between pharmacological and psychosocial interventions How do pharmacological and behavioral treatmcnt work together to enhance the effectiveness of both? On one level, the enhanced effective- ness of combined programs is not surprising. The effect can be viewed simply as evidence that "more is better." In fact, howcvcr, "more" is not "better." Lando (6) has demonstrated that program effectiveness can ac- tually decline as more components are added, so the "kitchen sink" ap- proach is not a sound model of treatment design. A more refined approach is to consider the potential for mutual en- hancement between the two elements in a combined program: what can pharmacological and behavioral treatments contribute to each other? Complementarity. Pharmacologic and behavioral approaches to smoking cessation are complementary. Because cigarette smoking is maintained by both pharmacological and behavioral factors, each approach addresses only part of the problem. For example, behavioral treatment programs have not been very effective in addressing withdrawal symp- toms. Other than recommending relaxation techniques for the emotional distress that accompanies withdrawal, behavior therapists can only suggest that smokers simply tolerate the initial period of discomfort. The use of nicotine gum may provide specific relief for symptoms of early with- drawal, thus filling this gap in the behavior therapy approach. Conversely, not all the problems of cessation are pharmacologic. Es- pecially later in maintenance, some temptations to smoke are driven not by withdrawal or by a need for nicotine but by such factors as social pres- sure or habit. In these cases, the gum's pharmacological action• may be of little consequence. A combined approach to treatment thus addresses the complex nature of smoking behavior. Mutual enhancement. Although the complementarity of behavioral and pharmacologic treatments goes a long way towards explaining their combined effectiveness, it does not go far enough. This view still im- plies that the effects are simply additive - each treatment adds something to the other, but does not actually change or enhance the other. In statistical terms, it implies independent main effects but no interaction. Yet the find- ings on combined treatment suggest that there is an interaction (see es- pecially Schneider and colleagues, 11). To understand the basis for this interaction, we must examine how each approach enhances the efficacy of the other. One way each treatment enhances the other is by attracting participants, as is evident in Wilson and colleagues' data. Patients are more likely to accept pharmacological treatment when the physician delivers an ap- propriate intervention. Call this a"persuasion" effect. This seems trivial until one notes Wilson and his colleagues' finding that 40% of patients refused the gum even when it was prescribed by their physician and that 88% of those who accepted a prescription used it for less than two weeks. Consider, too, Jamrozik and colleagues' (3) report that only half of 198 1 199

motivated volunteer patients use the gum regularly. To state the obvious: the gum is not effective when it is not used. Conversely, an offer of nicotine gum can entice reluctant smokers into behavioral treatment. Many smokers have a long history of cessation failures (even those who eventually quit successfully have typically failed three times, 12) and as a result are reluc- tant to go through yet another effort. To them, nicotine gum represents new hope -a credible reason to try again despite previous failures. Psychosocial interventions also enhance the effectiveness of nicotirre gum by facilitating proper use of the gum. The evidence suggests that many treatment drop-outs and failures can be prevented with proper use of the gum. Proper use includes learning to chew each piece correctly so as to titrate the dose appropriately, using enough pieces of gum to be effec- tive, and persisting in using the gum long enough to protect against relapse. As Schneider points out, there are many attitudinal and educa- tional barriers that inhibit proper use of the gum. By addressing these through education, instruction, persuasion, and support, psychosocial interventions enhance the effectiveness of pharmacotherapy. Conversely, pharmacotherapy may also improve the outcomes ofpsy- chosocial interventions by enhancing compliance. Providing nicotine sup- plements increases smokers' sense of their own efficacy, making them more likely to persist in coping efforts. This may express itself in lower drop-out rates and increased participation in program activities, which enhance the effectiveness of treatment. Potential incompatibilities. Pharmacological treatment also has the potential to undermine psychosocial interventions. Writing about other be- havioral techniques, Lichtenstein observed that some procedures "may lead participants to adopt a passive, 'do it to me set that is incompatible with efforts to teach self-control and problem-solving strategies" (8, p. 812). If smokers are allowed or encouraged to view nicotine gum as a magical cure for a purely physiological problem, they will reduce their in- vcstment in psychosocial interventions. Smokers who rely mechanical- ly on the pharmacological actions of the gum without investing effort in coping activitics arc likely to fail. Pharmacological and psychosocial interventions thus enhance each other only when they are properly t"- tegrated and balanced in the treatment implcmentation and in the smoker's own mind. Directions for further enhancements of treatment outcome A major lesson from the papers in this chapter is that continucd inr provement in treatment success rates is possible. Such irnprovements w'll be realized by attending to basic and universal principles of therapy and behavior change, improving patient recruitment, and perfecting the use of the gum itself. Hajek's findings support the need for attention to ba- sic principles of therapy. While all groups received nicotine gum and ap- propriate education and instruction, smokers were most successful when the format maximized their participation, efficacy, and social support. We must learn from the broader literature on therapy and behavior change. We should attend to what works in other contexts and be prepared to apply this information to smoking cessation. "Non-specific" factors such as group composition and cohesion, therapist relationship factors, and client expectancies may be important sources of improvement in smok- ing cessation treatments. Whilc not usually considered part of the treatment regimen, recruit- ment procedures have substantial impact on program success rates. Smokers who don't avail themselves of treatment may not degrade a program's outcome statistics, but they do lirnit its true effectiveness. Sutton's study of workplace programs neatly illustrates this. On one level, the intervention failed: workers who were invited to join a cessation pro- gram were no more likely to quit than those who were not recruited. Yet, the treatment itself was effective: smokers who entered treatment were much more likely to quit. Clearly, the way to enhance the program's impact is to improve recruitment. Wilson and his colleagues' results also illustrate this point. Training physicians to intervene with smokers results in improved recruitment to nicotine gum therapy. This will obviously improve outcomes without actually modifying the "treatment" itself. Outcomes will also be improved by attending to the process of gum use. Parameters of nicotine gum use-e.g., length of use, dose, and timing-control the gum's effectiveness. Improper use of the gum is ttn- doubtedly a major cause of treatment failure. Wilson and his colleagues show a dramatic increase in short-term gutn use when the gum is in- troduced by a trained physician. Schneider's work highlights many of the issues that need to be addressed clinically for the gum to be maximally effectivc. Simply providing factual information and guidance is not al- 11'ays enough; programs need to address barricrs to proper use. A fttr- ther step may be to focus on gum use itself as a target for behavioral in- tervention: monitoring, prompts, and even incentives for proper and Continued gum use should be considered. As suggested above, one of the activc ingredients in behavioral programs using Nicorette trray be the abil- ~~ato promote proper gum use. Systernatically enhancing this effect may xtnlize treatment effectiveness. 200 1 201

Combined vs. integrated treatment Careful consideration of the interactions between behavioral and phar- rnacologic treatment implies that integrating these components requires more than simply combining them. Integration implies a package in which the components are organized so as to maximize their mutual enhance- ment. In an integrated program, each component is allowed to perform the functions it does best: the gum is used to cope with withdrawal, while behavioral techniques are used to instill long-term changes in lifestyle. Each component serves the needs of the other: behavioral interventions specifically promote proper gum use; gum use is incorporated as an ac- tive coping strategy in the ex-smoker's larger coping repertoire, rather than as mechanical medical therapy, and so on. Melding the two ap- proaches into a tightly-integrated package will result in a treatment whose effect is greater than the sum of its parts. The future of integrated programs The trend towards integrated programs presents a dilemma. On the one hand, integrated programs seem to be more effective and arc there- fore preferable. Yet, a driving force behind the acceptance of Nicorette is the recognized need to treat smokers who are not reached by clinic pro- grams. If we promote the use of Nicorette in these enriched contexts, arc we not again ignoring the great majority of self-helpers? And if we pro- motc the use of the gum on its own, are we not doing less than we could to promote cessation? The solutions to this dilemma are many, and all lie somewhere between the extremes. It makes little sense to expect all smokers to attend inten- sive clinics or to claim that clinics are obsolete. One possible compromise is the use of abbreviated versions of well-validated intensive programs. Another option is the use of electronic and print media to convey the pro- gram material where direct intervention is not an option. Non-traditional interventionists and intervention sites should also be considered. Not only physicians, but medical receptionists and dental hygenists may be able to effectivcly dcliver abbreviated interventions in novel sites. Just the sim- ple device of bringing the program to the workplace, as Sutton has done, may provide an avenue for broadening the treatment population. Whatever directions the smoking cessation field takes, it seems clear that we will need to maintain multiple options for treatment. There wlll always be some smokers who require the intensive intervention of a clinic or even individual treatment, and ther ~ will always be others who want nothing morc intensive than a prescri~tion. We need to accomn'~od` 202 them all. Between these extremes will be the great majority of smokers, many of whorn would do better with behavioral treatment if they could be recruited. It may be helpful to develop a better understanding of smok- ers' resistance to treatment and compliance, along with more effective strategies for overcoming these barriers. It also seems likely that only some smokers profit from behavioral intervention. To date, we have only ag- gregate data suggesting that the average smoker does better in a com- bined program. An understanding ofwho most profits from behavioral intervention may enable development of screening and triage strategies. Identifying those who can succeed with gum alone and those who re- quire additional intervention may be the solution to the dilemma. Conclusions The best current treatments for smoking combine pharrnacotherapy and behavioral interventions. We have only begun to explore the possi- bilities and forms of this combination. Some improvements in outcome will come frorn attending not to the treatment itself, but to "peripheral" issues such as patient recruitment, retention, and compliance. Improve- ments in the treatment itself will come both from improving its separate components and from improving their integration. Programs that com- bine thee best possible pharmacological product and the best possible be- havior change strategies into a truly integrated package will set new stan- dards of success in smoking cessation treatment. References 1. Christen A, McDonald JL, Olson BL, Drook CA, Stookey GK. Ef- ficacy of nicotine chewing gum in facilitating smoking cessation. J. Am. Dent. Assoc. 1984; 108:594-597. 2. Fagerstrom KO. A comparison ofpsychological and pharmacolog- ical treatment in smoking cessation. J. Beh. Med. 1982; 5:343-351. 3. Jamrozik K, Fowler G, Vessey M, Wald N. Placebo controlled trial of nicotine chewing gum in general practice. Brit. Med. J. 1984; 289:794-797. 4. Killen JD, Maccoby N, Taylor CB. Nicotine gum and self-regulation training in smoking relapse prevention. Beh. Ther. 1984; 15:234-248. 5. Kunze M. Smoking Trends in Europe. EuropeanJ. of Respir. Dis. 1983; 64 (Supplement 26):103-106. 6. Lando HA. Effects ofpreparation, experimenter contact, and a main- tained reduction alternative on a broad-spectrum program for clitninating smoking. Add. Beh. 1981; 6:123-133. 203

7. Lee JW, Hart R. Techniques used by individuals who quit smoking on their own. Paper presented at the annual meeting of the American Psycho- logical Association, Los Angeles, August, 1985. 8. Lichtenstein E. The smoking problem: A behavioral perspective. J. Consult. and Clin. Psych. 1982; 50:804-819. 9. Russell MAH, Merriman R, Stapelton J, Taylor W. Effect of nico- tine chewing gum as an adjunct to general practitioners' advice against smoking. Brit. Med. J. 1983; 287:1782-1785. 10. Russell MAH, Wilson C, Taylor C, Baker CD. Effect of general prac- titioners' advice against smoking. Brit. Med. J. 1979; 6184:231-235. 11. Schneider NG, Jarvik ME, Forsythe AB, Read LL, Elliott ML, Schweiger A. Nicotine gum in smoking cessation: A placebo- controlled, double-blind trial. Add. Beh. 1983; 8:253-261. 12. United States Surgeon General. The Health Consequences vf Smoking. Rockville, MD: U.S. Department of Health and Human Services, 1979. 13. United States Surgeon General. The Health Consequences of Smoking. Rockville, MD: U.S. Illepartment of Health and Human Services, 1979. Clinic Based Cessation Strategies Edward Lichtenstein, Ph.D. University of Oregon and Oregon Research Institute Eugene, Oregon Introduction This volume, and the conference which gave rise to it, reflects a resur- gent interest in pharmacological approaches to smoking cessation. Evalu- ation of this approach may be sharpened by considering the context in which it occurs. This paper attempts to provide such a context. First, some key historical trends arc briefly noted. Second, the current "state-of-the- art" of clinic-based smoking cessation is described with broad strokes. Lastly, pharmacological strategies are considered with respect to the most current and powerful trend in smoking cessation-the adoption of the public health perspective. Historical trends Serious smoking cessation research and service programs have a rather short history-about 25 years. Even early on there were proponents of both pharmacological approaches and educational/psychological ap- proaches. The early pharmacological strategies were rather primitive by current standards. While the role of nicotine in the maintenance and ces- sation of smoking was poorly understood, the notion of replacing or nunucking nicotine's action was plausible. Much of this early pharmaco- logical research focused on lobeline presented in over-the-counter products like Bantron and occasionally by injection. Placebo designs yielded fairly convincing demonstrations oflobcline's lack of efficacy (24). Cigarette smoking quickly attracted the attention of workers in be- havior therapy. Cigarette smoking lends itself nicely to the target behavior research strategy that has proved fruitfid in several areas of behavior ther- apY. For the investigator of behavioral change, smoking provides an i're Paration of this paper was partly supported by Grant # CA38243 from the .Vationa( Cancer Institute. Portions (f the paper are based on a review by Kamarck `'"d Lichtenstein (21). 204 1 205

opportunity to study behavior changes in a meaningful, naturalistic con- text that still permits adequate measurement and controls. Behavioral approaches to smoking cessation tended to reflect current practices or the zeitgeist in behavior therapy rather than to derive from an analysis or understanding of smoking behavior per se. Thus, initial treatment strategies were often borrowed from other addictions such as alcoholism or obesity. Smoking was considered to be a learned habit; pharmacological and biobehavioral processes were neglected. Earlier be- havioral approaches to smoking featured conditioning methods andlor self-control strategies and tactics that represented behavioral thinking in the 1960s and early 1970s. Current behavioral approaches have major cog- nitive components reflecting recent interest in cognitive behavioral strate- gies. Within this general trend, however, behavioral workers have gener- ally maintained an empirical attitude and have become increasingly sensitive to the developing body of knowledge concerning both psy- chosocial and pharmacological processes in smoking behavior. For ex- ample, behavioral workers have evolved nicotine fading (11) or brand switching strategies to deal with pharmacological processes and have tend- ed to be sympathetic to nicotine chewing gum as an adjunct or even major component of cessation programs. Another noteworthy trend has been the shift from an emphasis on ces- sation of smoking to an emphasis on maintenance and relapse preven- tion. This shift was fueled by repeated observations that most participants in cessation programs either quit or greatly reduced their smoking but the majority of those who even initially succeeded subsequently relapsed, most of them soon after program termination. While early programs tend- ed to focus on simply getting people to quit smoking by the end of the program, now considerable program time is devoted to relapse preven- tion. Marlatt and Gordon's book on relapse prevention (32) epitomizes this trend. Psychological, especially cognitive, and pharmacological strate- gies are being employed in the service of maintenance and relapse preven- tion. Relatedly, longer follow-up (e.g., one year or longer) for interven- tion programs is now demanded before a seemingly successful program is taken seriously. An important historical trend which has some interesting implications for both behavioral and pharmacological investigators is the shift frotn a clinical to a public health perspective. Most of the early research and service programs in smoking cessation had a clinical focus in which in- terventions were delivered largely in a one-to-one or small group set- ting to smokers who had self-selected themselves to seek out and attend such programs. The aim of behavioral and pharmacological research was to develop effective strategies that could be incorporated into the treat- ment armamentarium of service providers and which could be dissemi- nated in group programs sponsored by voluntary organizations or even the private sector. Many behavioral programs were intensive in the sense of both number of sessions and the use of aversion or other strategies that required clinical monitoring and skill. There has now been a shift toward a public health perspective on smoking cessation in which it is necessary to reach large numbers of smokers, in- cluding those who do not wish to attend scheduled, formal programs. Improved self-help materials, more effective use of the primary care phy- sician, contacting smokers at worksites, media campaigns, primary and secondary prevention programs aimed at youth, and community-wide smoking interventions are the approaches that now draw attention and funding. A major challenge is how to apply the behavioral and pharmaco- logical knowledge gained from clinically focused cessation research to these broader contexts where there is much less opportunity for monitor- ing and control. A final historical trend to be noted here has been the important methodological improvements made in smoking research-including an increasing recognition of the importance of long-term follow-up, bi- ochemical verification of self-reports, and information about sample characteristics and attrition (27). Some promising treatment approaches have emerged during this time as well; there now exists a group of in- novative studies which utilize various combinations of behavioral, cog- nitive, pharmacological, and aversive approaches plus relapse prevention strategies that report 40%-50% abstinence six months to one year after treatment is over (1,5,6,7,16,18,20,22,25). Behavioral Cessation Strategies Behavioral cessation strategies have been developed from treatment techniques used to intervene with a wide range of behavioral and addic- tive disorders. These strategies include self-management techniques, aver- sive procedures and some combination of the two. Self-Management Strategies. Most behaviorally-based treatment programs have included "self-management" procedures in which smokers are taught to identify the cues associated with smoking, to modify or avoid these cues, and to alter the social and psychological reinforcers contribut- ing to their smoking behavior. These treatments are based on the rationale that smoking is analogous to any other overlearned behavior which is maintained by environmental cues and consequences which accompany it (27). 207 206

The research on the effcctivencss ofsingle sclf-management procedures in smoking treatment is not encouraging. Sclf-monitoring-recording the times, places, and situations associated with smoking - has been shown to produce only temporary reductions in cigarette consumption (29). Contingency contracting - arranging monetary consequences for smok- ing reduction-has resulted in short-term reductions, which last only as long as contingencies are in effect (39). "Stimulus control" procedures, which involve either restricting the situations in which smoking occurs or altering the cues for smoking, have been shown to have only moder- ate effect when used alone (29). The specific treatment effects attributa- ble to response substitution techniques (e.g., taking a walk after a meal instead of smoking, 27) have never been evaluated. Stress-related smoking has been a frequent intervention target for self- management procedures. Again, single procedures targetting this con- cern, such as relaxation training or systematic desensitization (an anxie- ty reduction procedure) have not been shown to be effective (29). Aversive Procedures. A second approach to smoking cessation aims to reduce the reinforcing value of smoking by pairing it with aversivc stimuli. Three major kinds of avcrsive stimuli have been paired with smoking: electric shock, negative images, and cigarette smoke itself. Cigarette smoke seems to be the most effective aversive stimulus, and "rapid smoking" is the most widely researched aversive smoking method. In this procedure, participants in the laboratory arc asked to smoke con- tinually, inhaling every 6 to 8 seconds, until subjective tolerance is reached. A variant of rapid smoking is called "satiation." This is generally a take- home procedure in which subjects are asked to double or triple their base- line smoking rate in the natural environment. Satiation is a convenient procedure, but does not lend itself to careful monitoring. The rapid- smoking literature is a large one and results are often inconsistent, but positive results (15,1(i) continue to outnumber negative findings (35). Possible negative side effects of rapid smoking have been investigat- ed; cardiovascular irregularities have bcen documented in several studies (15), but these have not led to any significant clinical symptoms. Rap~`1 smoking seenu safe, especially for nonsymptoniatic, relatively yoUI'f' adults; the monitoring procedures needed to carry out this technique With certain high-risk patients may limit its usc in nonmedical scttings. There have been several attempts to develop effective but less risky ternatives to rapid smoking. "Smoke holding;" in which subjccts arc askc~ to hold smoke in their niouths for a specified period of time witlloOt tn haling, was found to bc the most promising procedure in a coi'T`'r~~ of avcrsivc alternatives. This procedure was rated as quite avcrsivc. 1 it produced nonsignificant physiological changes, i.e., heart rate, blood pressure, CO (33). Initial work with smoke holding showed that it led to 33%> abstinence rates at six-month follow-up (23). A very recent study found good results when smoke holding and nicotine fading were conlbined-44%> abstinence at one year (26). It is surprising that so lit- tle work has been done with this procedure. "Focused smoking" (13), in which subjects are asked to smoke at a normal rate but to concentrate on their negative sensations, and rapid puffing-smoking without inhaling- are other variations. In sum, although rapid smoking is an effective smoking cessation procedure, the search for less risky alternatives continues, motivated by a desire to increase the disseminability of the technique. Combined Self-Management/Rapid Smoking Programs. In an at- tempt to improve long-term success rates, most of the recent studies mak- ing use of rapid smoking or satiation have combined this procedure with self-management methods. Some of these rnulticomponent programs have reported six-month abstinence outcomes of 50% or better (25). Lan- do's work is noteworthy here since he has demonstrated confirmed one- year abstinence rates of 40% in several studies using a combination of satiation, self-management, and relapse prevention strategies. In the few studies that have explicitly examined the incremental effect of aversive and self-management procedures in combination, some find an incremen- tal effect with the use ofself-tnanagemcnt (5,6), and others do not (4,12). Nicotine Based Strategies Recently, a growing consensus has emerged implicating nicotine in the maintenance of smoking. Two recent treatment strategies are based upon the rationale that smoking is maintained by the reinforcing effi~cts of nico- tine intake. The first, "nicotine fading" or brand switching (11), is a proce- dure in which subjects are asked to switch cigarette brands in graduated increments until they are smoking cigarettes containing about 90% less nicotine. The second, the use of a nicotine substitute, such as nicotine gum, is recommcnded to be used only after the smoker has stopped the use "f cigarettes; the gum is intended to reduce withdrawal reactions. Nicotine Fading. Subjects using this procedure may be asked to plot their daily tar and nicotine intake as well. By presumably decreasing nicotine I ntake, srnokers reduce their reliance on the drug, making it easier to quit c°j»pletely. Initial studies showed positive results when self-monitoring Was combined with nicotine fading in a fivc-weck or eight-week treat- entProcedure achieving a 40%, six-month abstinence (10,11). Unfor- 208 1 209

00 01 ® tunately, other reports have not replicated these results, finding lower long-term abstinence rates (3,26). The procedure is simple and is very well accepted by program participants; it may have good placebo value. Lando and McGovern (25) recently found that a substitution of nico- tine fading for satiation in their multicomponent program did not dirnin- ish abstinence rates. Nicotine Gum. Two types of treatment studies have investigated the use of nicotine gum as a smoking cessation aid. The first type investigates the effectiveness of gum in the context of minimal supportive treatment. Schneider et al (38) compared a nicotine and a placebo gum condition in a "dispensary treatment" study and found no differences in abstinence rates between the conditions at six months (15% for nicotine group, 18% for placebo) or at one year follow-up (8% vs. 13% respectively). Three other studies (2,8,37) each compared groups of patients randomly assigned to various levels of physician intervention, some of which included nico- tine gum. While two (8,37) reported significant effects of gum availabil- ity on cessation and long-term maintenance, the third study did not (2). A second group of treatment studies uses nicotine gum as a supple- ment to ongoing psychological treatments, ranging from supportive counseling to multicomponent behavioral/aversive programs. Although these studies produce mixed long-term results, all but one of them (9) show at least directional support for nicotine gum treatment, and they present much better abstinence rates than the minimal treatment studies described above. The most encouraging studies in this second group of reports found one-year differences between nicotine and control treatment when used in conjunction with clinic treatments. Hjalmarson (19) reported absti- ncncc rates of 29% (nicotine gum) and 16% (placebo) at one year. Jarvis et al (20) reported a 47% one-year abstinence rate in his nicotine group and a 21 °% rate in placebo controls. Raw et al (36) also reported signifi- cant one-year treatment differences (38% vs. 14%)), but their control groups did not include placebo gum and were drawn from an earlier treat- ment cohort. Fagerstr6m (7) compared nicotine and placebo gum in the context of an individualized behavioral treatment, including self-control and aver- sive methods, and found significant treatment effects through six months (63% vs. 45% abstinence rates) but not at one year (49% vs. 37%). Hall et al (18) compared three conditions: behaviorally-oriented treatment, gum treatment, and a combined condition (behavioral plus gum), and, although results were promising in favor of the combined treatment, group differ- ences were not significant at one-year follow-trp (behavioral-28%, gu'n 210 only-37%, combined-44%). Killen et al (22) compared three programs for enhancing maintenance in an intensive behavioral/aversive treatment program, and found directional support for a combined gum and skills training program, with a 50% abstinence rate for the combined group 10.5 months after treatment. These studies must be considered stringent tests of gum effects in light of the comprehensive treatment received by the comparison groups. They demonstrate that when proper interven- tion is given with the gum it can have a significant effect on short-term success and a demonstrable, although generally not statistically signifi- cant effect, on long-term success. Relapse Prevention The high rates of relapse which typically characterize smoking treat- ment have led many investigators to experiment with interventions designed to enhance maintenance of treatment gains. Nicotine gum can be construed as a way of avoiding short-term relapses associated with nicotine withdrawal. Outside of the nicotine gum approach reviewed above, four major types of relapse prevention strategies have been described in the literature: booster treatment, social support, coping skills training, and multiple factor approaches. Each will be reviewed below. Booster treatments. By and large, booster treatments, in the form of ongoing group meetings (5), telephone contacts (4), and aversive booster sessions (5), have not been shown to have long-term effects on nonsmok- ing maintenance. Social support. The social support approach to relapse prevention em- phasizes the social reinforcers in a smoker's environment as potential fac- tors in maintenance. Although some correlational studies suggest that certain measures of partner helpfulness and perceived social support seem to predict maintenance of treatment gains (28), intervention studies of attempts to maxirnize social support have not shown positive results. Coping skills approach. The coping skills approach to maintenance emphasizes preparation for situations that arise after treatment which may place ex-smokers at risk for relapse. Many investigators have developed and evaluated multicomponent programs incorporating coping skills maintenance procedures. Several of these programs have yielded good results (e.g., 34), although others have not (e.g., 12). Multiple factor models. Two multiple factor models of relapse prevention have been developed and programs based upon these models have been tested, with mixed results; Marlatt and Gordon's (31) influen- 211

tial cognitive-behavioral model of relapse emphasizes successful coping with "high risk" situations and the motivational and cognitive response to lapses in abstinence as determinants of long-term maintenance. The model de-emphasizes biobehavioral processes and seems implicitly con- cerned with preventing late relapses. Two studies comparing a relapse prevention approach with appropriate control groups found no signifi- cant differences (3,30). Hall (14) developed a model of relapse emphasizing the interaction be- tween coping skills and nonsmoking commitment and tested this ap- proach by comparing a skills training/commitment enhancement proce- dure to a discussion control following aversive smoking cessation treatment (17). Their commitment enhancement procedure included phys- iological feedback and information about the negative aspects of smok- ing. They found modestly significant results for their relapse prevention package at one-year follow-up, 46% vs. 300% abstinence. In summary, multicomponent aversive/self-management packages have produced the best results among non-pharmacological approaches. Nico- tine gum appears to produce consistent positive results and seems espe- cially promising in the context of clinical treatments. While the absolute differences between placebo and gum conditions are not large, the con- sistency of findings in the recent gum literature is unparalleled in the rest of the smoking cessation literature. Public Health Perspective and Cessation Strategies A clinical perspective implies intensive therapeutic efforts aimed at those who need and seek out help. Given the very large numbers of smokers in most western societies, a clinical approach to smoking cessation can only have very limited impact. It could be argued that a clinical approach should be reserved for those who are particularly at risk, such as post- MI patients who resume smoking. The public health perspective implies delivering education and intervention programs that will reach large populations. Smoking intervention research is clearly shifting toward broader ancl different contexts, including the physician's office, worksites, schools, community-wide interventions using the media, and unaided quitting or self-help. While the effectiveness or success rates of such pro- grams may be considerably lower than the intensive clinical programs, the overall efficacy and cost-effectiveness may be much greater. This shift in direction described here can be framed in another way. Clinically focused cessation research basically deals with smokers who are sufficiently motivated and concerned enough about their smoking to attend treatment sessions and often pay modest or sizeable fees and deposits. Public health-focused programs are aimed at a broader range of the smoking population, including those who are not yet motivated to quit. C3ne aim of these programs is to induce more people to attempt to quit. Even if success rates for quitting remain modest, if more smok- ers are induced to try and to try again, prevalence rates will go down. Implication for Behavioral Strategies. Behavioral cessation pro- grams that include risky or even nonrisky aversion have serious limita- tions from the public health perspective. Because both rapid smoking and satiation intensify the natural effects of smoking on the cardiovascular system, they require screening and are inappropriate for certain popula- tions. Even beyond risk and safeguards, many citizens have ethical or moral reservations about aversion procedures. These factors make it unlikely that any of the major voluntary health organizations such as the Ameri- can Cancer Society or American Lung Association-the major dissenii- nators of smoking cessation technologies in America - would incorporate aversion strategies into their programs. Nonaversive alternatives need to be found. Lando's program using nicotine fading instead ofsatiation (26) overcomes these objections, but is still costly in terms of requiring 16 sessionS. Cessation programs delivered in small groups are likely to be more cost- effective than individual treatment. But they will still reach'relatively small numbers of self-selected, motivated smokers. Implications for Pharmacological Strategies. The data consistently support the usefulness of nicotine chewing gum as an adjunct in sn-tok- ing cessation. The availability of gum, which is a relatively new strate- gy, is also likely to encourage more smokers, and perhaps more physi- cians, to undertake cessation attempts. Physician intervention could reach large numbers since, in America at least, it is estimated that three-quarters of the population consult a physician within a two-year window of tirne. The effectiveness of nicotine gum seems directly related to the provision of both supportive counseling and behavioral strategies. Major barriers to the utilization of nicotine chewing gum include the need to screen out certain patients, the attitudes and behavior of physi- cians, and the economics of medical care delivery. For physicians to ob- tain good results, they need to counsel their patients and to schedule follow-up visits. The medical care delivery system, at least in the Unit- ed States, however, does not provide incentives for this kind of physi- cian/patient interaction. A concerned, nonjudgmental attitude, which in- cludes replacing a "willpower theory of smoking" with an appreciation of the psychological and biobehavioral processes in srnoking, is also likely 212 1 213

to be important. Education of physicians about the nature of cigarette smoking and the proper use of nicotine gum is sorely needed. Developing ways of combining gum with both personal support and behavioral self-management strategies, and doing so in a way that is cost effective, is a challenge for the future. Several of the papers in this volume demonstrate the feasibility of employing nicotine gum in the public health approach, for example, in primary care settings and at the workplace. Such efforts will also require greater interdisciplinary cooperation and collabo- ration between medical and psychologically trained professionals. Propo- nents of behavioral treatments and proponents of pharmacological treat- ments have often been in competition. The need now is for cooperation and integration. References 1. Best JA, Suedfeld P. Restricted environmental stimulation therapy and behavioral self-management in smoking cessation. J. App. Soc. Psych. 1982; 12:408-419. 2. British Thoracic Society. Comparison of four methods of smoking withdrawal in patients with smoking related diseases. Brit. Med. J. 1983; 286:595-597. 3. Brown RA, Lichtenstein E, McIntyre KO, Harrington-Kostur J. Ef- fects of nicotine fading and relapse prevention on smoking cessation. J. Consult. and Clin. Psych. 1984; 52:3007-308. 4. Danaher BG. Research on rapid smoking: Interim summary and recommendations. Add. Beh. 1977; 2:151-166. 5. Elliot CH, Denney DR. A multi-component treatment approach to smoking reduction. J. Consult. and Clin. Psych. 1978; 46:1330-1339. 6. Erickson LM, Tiffany ST, Martin E, Baker TB. Aversive smoking therapies: A conditioning analysis of therapeutic effectiveness. Beh. Res. and Ther. 1983; 21:595-611. 7. Fagerstrom K-O. A comparison of psychological and pharmacolog- ical treatment in smoking cessation. J. Beh. Med. 1982; 5:343-351. 8. Fagerstrom K-O. Effects of nicotine chewing gum and follow-up ap- pointments in physician-based smoking cessation. Prev. Med. 1984; 5:517-527. 9. Fee WM, Stewart MJ. A controlled trial of nicotine chewing gum in a smoking withdrawal clinic. The Practitioner. 1982; 226:148-151. 10. Foxx RM, Axelroth E. Nicotine fading, self-monitoring and cigarette fading to produce cigarette abstinence or controlled smoking. Beh. Res. and Ther. 1983; 21:14-27. 11. Foxx RM, Axelroth E. Nicotine fading and self-tnonitoring for cigarette abstinence or controlled smoking. J. App. Beh. Anal. 1979; 12:111-125. 12. Glasgow RE. Effects of a self-control manual, rapid smoking, and amount of therapist contact on smoking reduction.J. Consult. and Clin. Psych. 1978; 46:1439-1447. 13. Hackett G, Horan JJ. Partial component analysis of a cotnprehen- sive smoking program. Add. Beh. 1979; 4:259-262. 14. Hall SM. Self-management and therapeutic maintenance: Theory and research. Pp. 263-300 in P Karoly and J Steffan (Eds.) Improving the Long-term Effects of Psychotherapy. New York: Gardner Press,. 1980. 15. Hall RG, Sachs DPL, Hall SM. Medical risk and therapeutic effec- tiveness of rapid smoking. Beh. Ther. 1979; 10:249-259. 16. Hall RG, Sachs DPL, Hall SM, Benowitz NL. Two-year efficacy and safety of rapid smoking therapy in patients with cardiac and pulmo- nary disease. J. Consult. and Clin. Psych. 1984; 52:574-581. 17. Hall SM, Rugg DL, Tunstall CL, Jones RT. Preventing relapse by be- havioral skill training. J. Consult. and Clin. Psych. 1984; 52:372-382. 18. Hall SM, Tunstall C, Rugg D, Jones RT, Benowitz NL. Nicotine gum and behavioral treatment in smoking cessation. J. Consult. and Clin. Psych. 1985; 53:256-258. 19. Hjalmarson AIM. Effect of nicotine chewing gum in smoking ces- sation: A randomized, placebo-controlled, double-blind study.J. Am. Med. Assoc. 1984; 252:20:2835-2838. 20. Jarvis MJ, Raw M, Russell MAH, Fcyerabend C. Randomised con- trolled trial of nicotine chewing-gum. Brit. Med. J. 1982; 285:537-540. 21. Kamarck TW, Lichtenstein E. Current trends in clinic-based smok- ing control. Annals of Beh. Med. 1985; 7:19-23. 22. Killen JD, Maccoby N, Taylor CB. Nicotine gum and self-regulation training in smoking relapse prevention. Beh. T her. 1984; 15:234-248. 23. Kopel SA, Suckerman KR, Baksht A. Smoke holding: An evaluation of physiological effects and treatment efficacy of a new nonhazardous aversive srnok- ing procedure. Paper presented at the annual meeting of the Associa- tion for the Advancement of Behavior Therapy, San Francisco, De- cember, 1979. 24. Kozlowski LT. Pharmacological approaches to smoking modification. Pp. 713-728 in JD Matarazzo, SM Weiss, JA Herd, NE Miller, SM Weiss (Eds.) Behavioral Health: A Handbook of Health Enhancement and Disease Prevention. New York: John Wiley & Sons, 1984. 25. Lando HA, McCullough, JA. Clinical application of a broad- spectrum behavioral approach to chronic smokers. J. Consult. and Clin. Psych., 1978; 46:1583-1585. 214 1 215

26. Lando HA, McGovern PG. Nicotine fading as a nonaversive alter- native in a broad-spectrum treatment for eliminating smoking. Add. Beh. 1985; 10:153-161. 27. Lichtenstein E, Brown RA. Current trends in the modification of cigarette dependence. Pp. 575-611 in AS Bellak, M Hersen, AE Kaz- din (Eds.) International Handbook of Behavior Modif:cation and Therapy. New York: Plenum Press, 1982. 28. Lichtenstein E, Glasgow RE, Abrams DB. Social support in smok- ing cessation: In search of effective interventions. Beh. Ther. 1986; in press. 29. Lichtenstein E, Mermelstein RJ. Review of approaches to smoking treatment strategies: Behavior modification strategies. Pp. 695-712 in JD Matarazzo, SM Weiss, JA Herd, NE Miller, SM Weiss (Eds.) Behavioral Health: A Handbook of Health Enhancement and Disease Preven- tion. New York: John Wiley & Sons, 1984. 30. Marlatt GA, Curry SG, Gordon JR. Treatment for smoking cessa- tion: The relative effectiveness of different programs and formats. Manuscript in preparation, Seattle: University of Washington. 31. Marlatt GA, Gordon JR (Eds.). Determinants of relapse: Implications for the maintenance of behavior change. Pp. 410-452 in PO David- son, SM Davidson (Eds.) Behavioral Medicine: Changing Health Lifestyles. New York: BrunnerlMazel, 1980. 32. Marlatt GA, Gordon JR. Relapse Prevention: Maintenance Strategies in the Treatment of Addictive Behaviors. New York: The Guildford Press, 1985. 33. Orleans CT, White ML, Nagey DA. Comparative physiological ef- fects and aversiveness of four non-hazardous aversive smoking proce- dures: Alternatives for the pregnant smoker. Paper presented at the annual mecting of the Association for Advancement of Behavior Therapy, Toronto, November, 1981. 34. Pomerleau OF, Adkins D, Pertschuk M. Predictors of outcome and recidivism in smoking cessation treatment. Add. Beh. 1978; 3:65-70. 35. Poole AD, Sanson-Fishcr RW, German GA. The rapid-smoking tech- nique: Therapeutic effectiveness. Beh. Res. and Ther 1981; 19:389-397. 36. Raw M, Jarvis MJ, Feyerabend C, Russell MAH. Comparison of nicotine chewing-gum and psychological treatments for dependent smokers. Brit. Med. f 1980; 281:481-484. 37. Russell MAH, Merrirnan R, Stapleton J, Taylor W. Effect of nico- 216 tine chewing gum as an adjunct to general practitioner's advice against smoking. Brit. Med. J. 1983; 287:1782-1785. Schneider NG, Jarvik ME, Forsythe AB, Read LL, Elliott ML, Schweiger A. Nicotine gum in smoking cessation: A placebo- controlied, double-blind trial. Add. Beh. 1983; 8:253-261. 39. Stitzer ML, Bigelow GE. Contingent reinforcement for reduced car- bon monoxide levels in cigarette smokers. Add. Beh. 1982; 7:403-412. 217

Can Training Family Physicians Improve Compliance with Nicotine Gum Use? Douglas M.C. Wilson, M.D. McMaster University Hamilton, Ontario J. Allan Best, Ph.D. University of Waterloo Waterloo, Ontario Elizabeth Lindsay-McIntyre, Ph.D. J. Raymond Gilbert, M.D. D. Wayne Taylor, M.A. Joel Singer, Ph.D. Introduction My colleague Allan Best and I are certainly honored to be invited to report on our collaborative work at the University of Waterloo and McMaster University in Canada, a project which has been very kindly funded by the United States National Cancer Institute. I will describe a study that is based on two key assumptions: 1) family physicians can help patients to stop smoking and can do so with a suffi- cient number of smokers to make the effort worthwhile; 2) the evidence for the ef6cacy of nicotine gum is strong enough in special settings to justify a study which examines whether it is effective when used in general primary care settings. Why the family physician? Some of this material was discussed briefly in a previous presentation in terms of the therapist ef/ect of the family phy- sician, something that we recognize as being powerful. We're not sure if we totally understand it. There is often an established relationship that smokers have with their family physician and so the trust and confidence is there. In both Canada and the U.S. a significant number of smokers visit the family physician within one year. In the U.S., I am told that 44 mil- lion smokers will visit the family physician in any one given year and in Canada about 4 million smokers will do the same. Family physicians arc generally knowledgeable about risk factors (i.e., factors which affect the development of disease) and can apply this knowledge to their own patients' risks. Physicians in Canada are more and more being seen as not only authorities on disease but are recognized as authorities on health. Lastly, for the purposes of the intervention that I'll describe here, family physi- cians are able to prescribe Nicorette gum. The question. Because of problems with lack of clear instructions and compliance by physicians, we phrased our key research question: does training family physicians and providing back-up materials increase the percent of patients who stop smoking over the percent who stop when the physician receives no such training? In this paper, within the larger question we will also look at the use of nicotine gum by patients of both trained and untrained physicians. This is a multifaceted randomized clinical trial and therefore we will focus on two major features of the study: what we did to train the phy- sicians, and some information about physician and patient behavior with regard to gum use. Study design and methods. First, a brief look at the study design. The 87 physicians in the study volunteered to participate after receiving a letter of invitation. We sent this letter to 600 family physicians within a 30- mile radius of the research center. We randomly allocated the volunteer physicians by practice into three groups. After withdrawals and dropouts, there were 23 practices within the usual care group; 24 practices in the untrained physician group; and 23 in the trained physician group. Patients were recruited into the study by the receptionists in the prac- tice offices, who followed a standardized recruitment procedure in all three groups. Patients were eligible if they smoked at least one cigarette a day; if they were between the ages of 16 and 65; and if they were not preg- nant or breast-fceding. The physician had no role whatsoever in the recruitment of the patients into this study. Also, patient motivation, unlike in some of the other studies that have been reported here, was not a criterion. These patients were all visiting their family physicians for usual reasons, and this study is looking at the family physician's role with all cigarette smokers in his/her practice. When the patients agreed to participate, they were not agreeing to try to quit smoking. They simply agreed to complete our questionnaires and to al- low us to follow them. Family physicians in the Usual Care group were instructed to approach their smokers in whatever way they would normally. They did not know which patients were in the study. In the group which we call Gum Only, family physicians were instructed to advise their recruited smokers to quit and to offer each of them a prescription for Nicorette. These family phy- sicians were not trained in any procedures specific to this study. 218 1 219

In thc Gum Plus group thc family physicians were trained in a specif- ic intervention to apply to the recruited smokers, and I'll be describing the components of this maneuver. One of the other unique aspects of this study relates to the method of offering the gum. There is no free gum. It is prescribed by the physician and is purchased by the patient as in real world practice. In Canada that works out to $24 to $28 per prescription. Measures. The measures reported in this presentation were essential- ly taken from the entry-to-study questionnaire, and the two-month ques- tionnaire. We will have a one-year questionnaire supplemented by coti- nine validation of those smokers who have reported quitting and are not using Nicorette. There are a number of formative measures, including the MD ques- tionnaire. We are monitoring office and recruitment procedures. We are looking at the patient's research chart which in essence is a flow chart that the physicians use. Fifteen percent of the patients in all three groups are receiving an exit phone interview which assesses compliance with protocol. Each of the physicians in all three groups have been taped in- terviewing a simulated smoking patient, and criteria for analyzing these have been established. We have also established focus groups which are open-ended debriefing sessions to discuss and receive feedback on the training, office procedures, printed materials and the basic intervention. The maneuver. I will present a very brief overview of the maneuver which is based on our understanding of efficacious treatment methods in the literature as well as on strategies which we have personally found to be effective. The basis of the maneuver goes back to Mike Russell's study of simple advice, which he demonstrated to be effective in itself. At McMaster, we built on that in a study five or six years ago, which showed that if family physicians also offered supportive follow-up visits, that there would be a doubling of quit rate or self-reported quitting at about nine months. Some of the other interventions that have been dis- cussed are those that have been studied essentially in the U.K. in general practice offices with advice/support and Nicorette. Karl Fagerstrom has looked at motivated smokers with advice/support/Nicorettc and a ver- bal contract. For the past two to three years at McMaster I have been ex- perimenting with but have not prospectively studied advice/support, Nicorette, and setting a target date. This then, is the intervention pack- age that we are teaching the family physicians to use in the Gum Plus group: 1) advice; 2) support visits; 3) the offer of Nicorette gum; 4) the negotiation of a target date; 5) some take-home tip sheets; and 6) a simplee contract. 220 physician training. The Gum Plus continuing medical education (CME) programs or training sessions had specific objectives. First, we sought to increase the knuwledAe of the family physicians of the project design, the issues of smoking cessation from the literature, gum use, and project administration. Second, we attempted to develop a better physician under- standinA of why the ingredients of the intervention were selected, what they were to do with their patients, and how to use the materials. We taught them skills in performing the intervention and in the use of the materials. We provided clear, simple delivery of facts verbally, visually and with printed materials, and used considerable repetition throughout. We used a lot of variety: of inedia, of leaders, of group size, and of learning mode. one of the slides that we p_re_sente ~ to physicians is wl at we call the Seven S's of gum use. These are included in the protocol, and we reinforced them many times by modeling and in simulations. The seven S's consist of 1. Stop Smoking-Do not Smoke and Chew 2. Substitute Gurn for Cigarettes 3. Slowly Chew 4. Several Pieces of Gum Per Day 5. Stay on the Gum for Two to Three Months 6. Staged Reduction of Gum Use over the Next Several Weeks 7. Stop Using the Gum We looked at the physicians' attitudes about the importance of family physicians working with smokers; their ability to make a difference; their use of intervention and the materials; and about following the adrninis- trative protocol carefully. We discussed with them the value of their par- ticipation in the study, in making the intervention, and of course their intcrest in the topic of smoking cessation. We provided an opportunity for thetn to practice the intervention, which led to discussion in groups when (eedbacFr was provided. The three principal behavioral skills that we taught were how to challenge patients-all family practice smoking patients-to consider quitting smoking; how to nc~qotiate with them a quit date and strategies for quit- ting; and how to offer support, at initial visits and in follow-up visits. The amount of each skill-challenQe, nc,~otiate and support-varied from visit to visit throughout the intervention. The training agenda The training sessions took three and one half to four and one half hours, during which time we provided an overview of smoking cessation is- sues, the context for the project, the study design and study measures. We also presented the key facts about grun use and an introduction to 221

the experimental protocol. We then described in detail the experimental protocol, and showed a videotaped role-modeling demonstration. Small group discussion was followed with practice with a simulated patient for each of three visits-the initial Entry Visit, the Quit Date Visit and Follow-up Visits. These were followed by a plenary discussion and ques- tions and a reception: The continuing medical education workshop is built on a three-phase cycle. For each visit in a series of three physician-patient appointments, the trainers first presented a review of objectives and intervention steps for the appointment, using slides and counseling materials. These materials were studied by participants in preparing for the workshop. Secondly, the trainer helped to demonstrate the counseling methods with one case of edited videotape of a patient-physician interaction. And thirdly, each participant rehearsed the counseling model in small groups using a simu- lated patient, and received feedback from one of the trainers as well as the simulated patient. Thus, our participating physicians were taught intervention skills in this training session. We provided physicians with patient folders, which included the flow sheets and the simple contract. We suggested the en- try visit would take five to seven minutes and we demonstrated such a visit which they practiced. At the entry visit, a decision needed to be made by the patient. The patient was given the following choices: 1) continuing to smoke; 2) try- ing to quit with the doctor's help and the use of the gum; 3) trying to quit without the use of gurn. A quit date within the next 30 days was negotiated at that visit for those who planned to quit. The tip sheets or take-home materials were offered and instructions were given in how to use them before the quit date and after the quit date. The guidelines for the quit date were again for a short visit, ten minutes. The participating physicians were shown a ten-minute quit date visit which they were able to rehearse. The flow sheet gave physicians some guidelines for what to tell and ask patients at the quit date visit and at each of the follow-up visits. The follow-up visit guidelines call for subsequent visits of five to ten minutes. This type of visit in Canada costs about $14 and is reim- bursed by the Ministry of Health (medicare). Now, my colleague, Allan Best, will present the two-month outcome results. Results Let ine begin by telling you a little bit about the sample of physicians with whom we are working. Before our study began, most of the physi- cians (81%) claimed to at least `bften" offer to help their patients with quit attempts. There were no differences between groups initially in that regard. Two-thirds of our physicians believed that the gum was at least "rrroderately effective" in helping patients quit. Almost all of the physi- cians were using the gum with patients at the time they started to take part in the study (88%) or had used the gum in the past (99%). I will now discuss questions of patient and physician compliance. John Hughes (this volume) discusses some of the compliance problems which can occur and which may mediate the effectiveness of gum use. In exit interviews with 15% of the patients in the trial, we asked what the physician had done during the initial visit. The patients in both the Gutn Only and the Gum Plus conditions reported that most oftheir phy- sicians had at least talked to them about smoking and about two-thirds of them had mentioned the gum. Significantly more of the trained phy- sicians than the untrained physicians were said to have talked about smok- ing with the patients (71 % in Gum Only vs 85% in Gum Plus, p =.013), but note that in both conditions we should be getting 100% both talk- ing about smoking and suggesting gum, since all of these physicians have been instructed to intervene in that way with their patients. The recep- tionist was trained to screen out patients with contraindications to gum use (pregnancy, breastfeeding). So there is a fairly significant level of reported physician noncompliance on which we still need to work., If a patient decides to try quitting-and 85% did if their physician offered help-the proportion deciding to use Nicorette is greater for trained physicians (71 % Gum Plus vs 61 % Gum Only, p<.001). Table I shows the proportions that actually do use the gum for at least two weeks. Significantly more of the patients working with trained physi- cians use the gum for at least two weeks, but notice that even with the trained physicians, very few of the patients really are using the gum even for as short a period as two weeks. There is still much room for improve- ment in patient compliance with gum use and with how physicians prescribe the gum. And finally, Table 2 shows the ratings of the gum's helpfulness by the patients who used it. The Gum Plus group is more likely to rate the gum as being helpful than is the Gum Only group, perhaps because they were more carefully instructed, and in fact were using the gum in a way which more closely approximated its intended use. Let me turn to sonle of the preliminary outcome data from the two- month questionnaire we gave the patients, and describe the self-reported quitting at that point. The quit rates for the three conditions are shown iu Table 3. Both Gum Only and Gum Plus conditions produce increased 222 1 223

Table 2 Table 1 Table 4 d-Duration of Gum Use. t lf R Self-Reported Cessation Rates at T o M th T b epor e - Se w on s y reatment Group. Treatment Group C ti R Pl G l essa on ates um us y Gum On 713) (N = 563) N <6% 6%-10% 11%-15% 16%-20% >20% ( = Usual Care 15 * 5 1 1 1 m Onl G 9 9 Length of Time on Gum y u 5 0 0 um at least two weeks 84 (12%) 129 (23%) On Gum Plus 2 4 2 3 11 g Off gum by second week 629 434 Chi-square = 24.59 p < .0001 (Chi-square = 27.24, p < 0.001) (Collapsed across last 3 columns) *Number of practices in this group. rates of attempted cessation. Training produces roughly a 250% increase in short-term cessation rates over the Gum Only physician condition; the Gum Plus physicians are getting a fourfold increase over the Usual Care group. Perceived Gum Helpfulness by Patients Who Tried Using Gum and Quit Smoking For At Least One Day. Rating Treatment Group Gum Only (N = 175) Gum Plus (N = 215) Not Helpful 17.7% 19.5 % Little Help 18.9% 9.3%o 29 1 0% 23.7 % Some Help Very Helpful . 24.6% 26.0% Extremely Helpful 9.7% 21.4% (C;hi-square = 15.97 p < 0.01) Table 3 Table 4 shows the variability in physician performance across these con- ditions. We're very encouraged by the fact that more than half of the phy- sicians in the Gum Plus are getting better than 20% reported cessation rates at two months. In contrast, something like three-quarters of the Usual Care practices are getting 5% or less reported cessation rates. Conclusion In summary, the McMaster-Waterloo Family Practice Smoking Ces- sation Project has accomplished the following: • A relatively comprehensive physician cessation counseling pro- tocol has been developed, building both on promising research and clinical experience. The protocol can be offered with rela- tive case to all smoking patients in a family practice. • A four-hour continuing medical education package has been tested. It uses state-of-the-art educational methods and is very well received by family physicians. • The protocol is acceptable to a representative group of patients. Gum Plus physicians see better compliance with their patients, but significant compliance problems still arc evident. • Despite the compliance problems, the Gum Plus strategy and physician trairung significantly increase gum acceptance and self- reported use, cessation attempts, and short-term success. ~, Self-Reported Cessation Rates at Two-Month Follow-up. Usual Care °/ Gum Gum Plus We look forward to the proof-of-the-pudding in our twelve-month follow-up results. (N = 578) (N=713) (N = 563) uit 24 hr. Tried to 38.1 `Yo 62.8%, 76.7 %, q No smoking in last week 4•(1% ~ 7.0 `%, 17.8%, 224 225

Nicotine Chewing Gum in Group Treatment of Smokers Peter Hajek, Ph.D. Addiction Research Unit Institute of Psychiatry London Introduction It is in group settings that nicotine chewing gum has been most thoroughly tested and shown to be effective over and above placebo (3,5,6,7,9). Randomized studies have demonstrated that groups using the gum do better than groups which do not use the gum. However, group therapy is not a unitary entity. Any group treatment involves a number of different "curative" factors that can be used with different emphases and in various ways (1,2). The aim of this paper is to show that in the coalition of nicotine chewing gwn and group treatment, purposeful utili- zation of group processes can significantly improve the results. This paper discusses two studies that investigated whether the effica- ey of the nicotine chewing gwn treatment of smokers in didactic therapist- oriented groups could be improved by incorporating methods to enhance the impact of relevant group processes,* and what influence group size has on outcome. Method Subjects. The subjects were 270 consecutive clients who started group treatment at the Maudsley Smoker's Clinic in London. One hundred and thirty-two weree treated in 14 therapist-oriented (T-O) groups and 138 in M..J 14 group-oriented (G-O) groups. Mean age was 39 years, 69%, were white ~ collar workers, 66%, were women, mean cigarette consumption was 24 per day, and niean carbon monoxide (CO) content in end-expired air (8) ~ was 30.5 ppim There were no significant differences between the two sub- k mea- s k e mta e a samples except that G-O subjects had a higher mean smo sured by the CO concentrations (F = 12.8; llF = 1, 268; p < 0.001). *'1'he study on which this part of the presentation is based has been published iit the Britisk Jonrnal of Clinical Psycholoqy (4). The study of group size added to this sample another 253 patients treat- ed in 19 more groups. This study is still in progress. Procedure. Fourteen T-O groups conducted during one calendar year were compared with 14 G-O groups run during the next year. All the referral, assessment and treatment arrangements, site of the meetings, etc., were identical throughout the study. The clinic service was provided free of charge. Smokers were either self-referred or sent by their phy'sicians. All had an individual assessment interview. The four-week course of treatment consisted of five weekly group sessions each lasting one hour. Clients were routinely offered nicotine chewing gum and were asked to stop smoking after the first meeting. All groups but two were conducted by the sarne male psychiatric charge nurse. Two male clinical psychologists participated (one at a time) as co-therapists. The principal outcome variable was smoking status at the end of treat- ment. Success was defined as not smoking during the week between the fourth and fifth meeting. People who dropped out were counted as failures. All claims of abstinence were checked by expired-air carbon monoxide. One year follow-tip data were gathered by means of postal questionnaires and telephone calls. Other variables recorded were the date of stopping smoking relative to the start of treatment, attendance, and nicotine chewing gum consumption. Therapist-oriented groups. The traditional didactic therapist-oriented approach evolved from commonsense economic considerations. Most often, the model of a teacher in the classroom is adopted. The therapist sees his/her main responsibility as being to impart advice and infortna- tion. The group's attention is thus concentrated largely on the therapist who is then compelled to keep offering more advice and information. The resources of the group itself receive little eniphasis. The T-O groups at the Maudsley Clinic followed this pattern. For much of the time the therapist talked to the group. When clients reported their experiences, they usually addressed the therapist and often phrased their contributions as questions. The therapist responded to each entry with factual information, encouragement, etc. The group proceedings were smooth and satisfactory. Group-oriented groups. The alternative approach focused on enhanc- ing group resources at the cost of the therapist's didactic activity. The aim was to heighten the development of the group's internal structure, autononry and ultimately its cohesion and attractiveness. The main practical change concerned the behavior of the therapist. Though he still provided guidance regarding use of nicotine chewing 227 226

gum, most of his othcr didactic activities were deleted. His interventions were reduced by about half and consisted primarily of various group_ oriented suggestions. Detailed sclfintroductions by group mcmbers were employed and contacts outside the group meetings were encouraged. Group pressure was intensified by initiating publicly declared weekly comtnitments of abstinence. The major part of each meeting consisted of group discussion of how members had coped without cigarettes for the past week. It was particularly during this part of the session that the therapist remained in the background. Group members quickly learned to address each other rather than the therapist and to view group inter- change as an essential component of treatment. Study of group size. This consisted of adding the results from the next 19 groups to the samplc and examining the relationship between group size and outcome in more detail. Results Looking at the number of clients abstinent at the end of treatment in each of the 28 consecutive groups, there were striking variations in suc- cess rate between individual groups of the same format. This "group ef- fect" was significant (chi-square = 49.1; DF = 26; p < 0.005). The growing experience of the therapist had no influence on the results in either T-O or G-O format (chi square = 2.29 and 0.95 respectively, DF = 1; NS). The presence of the co-therapists had no significant ef- fect on outcome (chi-square = 1.95; DF = 1; NS). Initial group size was a significant covariatc for outcome with larger groups being more suc- cessful (chi-square = 4.84; DF = 1; p < .05). None of the demograph- ic and pre-treatment characteristics recorded had a significant effect on outcomc. Table I compares the two group formats taking into account group effect and group size, where significant. The G-O format was significantly more successful. Although there was no difference between T-O and G-O groups in the proportion of people who succeeded in stopping smoking early on in the treatment course, in G-O groups there were significantly more people who suc- ceeded later after initially failing. G-O groups were also significantly better attended. There was no difference between the two samples in average daily nicotine chcwing gum consumption. At one-year follow-up there were significantly more people abstinent ] for the whole year in the G-O format than in the T-O format (22 [17% and 39 [28`%>] respectively, chi-square = 4.02; DF = 1; p < 0.05). Table 1 Comparison of Outcome, Attendance, and Gum Use in Therapist-oriented (T-O) and Group-oriented (G-O) Groups. Abstinent at the end of treatment Not smoking from the first week on Stopped smoking only later in the course Mean number of group meetings attended Dropped out (%) Mean consumption of nicotine gum per day T-O Groups G-O Groups Tbst Statistic (N = 132) (N = 138) 62 (47%) 92 (67%) X= 5.08; DF = 1; p<.025' z 44 51 X = 0.34; DF = 1; NS' 18 41 X = 7.36; DF = 1; p<.01*' 2 3.51 3.91 F = 4.58; DF = 1, 268; p<.05z 56 (42%) 38 (28%) X= 3.26; DF = 1; NS' 7.1 6.6 F= 0.80; DF = 1, 268; NS *Based on the number of clients still smoking after the first week. 'Group effect significant ZGroup size significant From Hajek, Belcher and Stapleton, Brit. J. Clin. Psych, 1985; 24:189-194. Initial group size was a significant covariate for outcome ~vith larger groups being more successful. To further explore this interesting find- ing more groups are being added to the analysis, giving a wider range of group sizes. We currently have data from 47 groups and the relation- ship between group size and outcome remains significant (chi-square = 9.42; p < 0.005-allowing for group format). However, chi-square for lack of fit in the model was close to being significant as well. More data are being collected and other sources of variance will have to be examined in more detail. The findings that have been presented here should be regarded as provisional. Figure 1 shows the relationship of group size and success rate in 47 groups ranging in size from 6 to 16 participants. As there were only a few groups at both extremes, in this histogram they were pooled together. Groups with 12 or more participants seemed to do better than smaller groups. This is a post-hoc observation but makes an interesting contrast to the number 12, which is sometimes quoted as an absolute maximum for the size of psychotherapy groups. Conclusions In a Smokers' Clinic program based on the use of nicotine chewing guni, a group format oriented towards the resources of the group itself 228 1 229

Figure 1 80% 70% 60% 50% L7 groups I 40% SUCCESS RATE IN GROUPS OF VARIOUS SIZE (N = 47) 6 groups 9 groups 7 groups 6 8 9 10 11 GROUP SIZE 9 groups 9 groups 12 13-16 was significantly more effective than a traditional didactic approach. In- dcpendent of group format, larger groups did significantly better than smaller ones. Two influences that could be stronger in G-O groups are group pres- sure and group support. Spontaneous group discussions also present more opportunity for modelling the coping responses of successfiil group mem- bers. Emphasis on the nicotine chewing gum in the two group formats did not differ and gum use was similar. Even in the G-O groups the gum remained one of the main topics of discussion. Once clear guidelines arc formulated, the G-O style of conducting groups is easy to implement. Our groups were conducted by an ex- perienced charge nurse and the presence of a psychologist did not in- flucnce the results. If we take the risk of generalizing our findings to other smoking cessation programs using groups, a practical suggestion can be made: when the therapist leaves the pulpit and invests energy in stimulat- ing the evolution of group structure and cohesion, treatment can benefit. Other indicators of the influence of group processes, i.e., "group ef- fect; (a variation between the groups) and the effect of group size merit a short comment. Groups which "gel" or groups which have a majority of early successes can have a significant advantage over groups which do not create an enthusiastic and supportive atmosphere or which are dis- couragcd by early drop-outs and failures. There may exist a kind of "bandwagon effect" dependent on the absolute number of initial successes which makes larger groups more effective. Also, the elements of group support and pressure and the example of successful group members could have been stronger in larger groups. Two groups run by the same therapist in the same style with the same type of clients can differ dramatically in outcome. Group practitioners are usually aware of this, but the phenomenon is mostly ignored in ex- perimental work. Our finding of a significant influence of group effect and other group related variables on outcome highlights the fact that studies of group treatment using only one or two groups per condition are usually of limited interest. In summary, the efficacy of nicotine chewing gum used in a group set- ting can be enhanced by selecting the most suitable psychological "paek- aging" for the intervention. In the combination of psychological and phar- macological treatments, group processes can play an important role. References 1. Block S, Crouch E. Therapet+tic Factors in Grot.ip Psychotherapy. Oxford: Oxford University Yress, 1985. 230 1 231

2. Butler T, Fuhriman A. Curative factors in group therapy: A review of the recent litcrature. Small Group Beh. 1983; 14:131-142. 3. Fee WM, Stewart MJ. A controlled trial of nicotine chewing gum in a smoking withdrawal clinic. The Practitioner. 1982; 226:148-151. 4. Hajek P, Belcher M, Stapleton J. Enhancing the impact of groups: An evaluation of two group formats for smokers. Brit. J. Clin. Psych. 1985; 24:189-194. 5. Hjalmarson AIM. Effect of nicotine chewing gum in smoking ces- sation: A randomized, placebo-controlled, double-blind study.J. Am. Med. Assoc. 1984; 252:2835-2838. 6. Hjalmarson AIM. Effects of nicotine chewing gum on smoking ces- sation in routine clinical use. Brit. J. of Add. 1985; 80:321-324. 7. Jarvis MJ, Raw M, Russell MAH, Feyerabend C. Randomised con- trolled trial of nicotine chewing-gum. Brit. Med. J. 1982; 285:537-540. 8. Jarvis MJ, Russell MAH, Saloojee Y. Expired air carbon monoxide: A simple breath test of tobacco smoke intake. Brit. Med. J. 1980; 281:484-485. 9. Puska P, Bjorkqvist S, Koskela K. Nicotine-containing chewing gum in smoking cessation: A double-blind trial with half-year follow-up. Add. Beh. 1979; 4:141-146. 232 Use of 2 mg and 4 mg Nicotine Gum in an Individual Treatment Trial Nina G. Schneider, Ph.D. University of California Medical School and Veterans Administration Medical Center Los Angeles Introduction I would like to preface my presentation with a qualifier. In the open trial I will discuss, I was not really interested in nicotine gum use in in- dividual treatment per se. I used the individual treatment setting as a means of conducting a process study. The idea was to obtain in-depth, self-report data normally inaccessible in a controlled study. The following is a preliminary report on the trial. Formal analyses of the data will be forth- coming. Clinical observations from this trial must be substantiated with systematic testing. Design One hundred and ten subjects participated. Each paid UCLA $400 for one year of treatment with nicotine gum provided free. The special na- ture of the sample should be kept in mind: highly motivated subjects, able to afford the treatment, specifically seeking the use of nicotine gum and a one-on-one approach. The trial allowed each participant up to 12 one-hour individual ses- sions of treatment. During this time it was possible to discuss withdrawal experiences (course, nature, duration) and reactions to two doses of nicotine-containing gum. There was an opportunity to identify the na- ture of slips or relapses and the development of coping skills for each in- dividual. The order of clinic visits was as follows: baseline meeting prior to quit day, quit day, at the end of the first week, weekly visits in the first month and six additional visits to one year. The timing of the latter visits was based on each subject's needs and on time periods for obtaining carbon monoxide validation of abstinence. During treatment, subjects were al- lowed access to both 2 mg and 4 ing doses of nicotine gum. The focus of this presentation is on observations regarding the nature 233

of nicotine gum use in treatment. These observations include: errors in the use of nicotine gum that afhect success; nicotine replacement and with- drawal symptoms, comparative effects of 2 mg vs 4 mg nicotine gum, and the kinds of cessation problems relieved when nicotine gum is in- corporated into treatment. I will also briefly comment on individualized vs. group treatment, viz., is individual treatment necessary? Use Errors with Nicotine Gum There has been little systematic study on the specific requirements for nicotine gum use (e.g., dose-response, frequency, length of treatment). In general, the individual is allowed to self-administer the preparation with few guidelines and inadequate training. As a consequence, compli- ance with proper chewing technique is poor and a number of conceptu- al errors in use occur which can undermine success. One purpose of this trial was to identify "errors;" why they occur, and how they can be cor- rectcd. Table 1 Errors in Use of Nicotine Gum Error Consequence 1. Chewing gum and Relapse continuing to smoke 2. Using too few pieces Inadequate relief Craving Early relapse 3. Using gwn for too short Continued craving a time Lack of alternative behaviors Relapse 4. Expecting a cure-all No coping for stress Vulnerability to craving Relapse Because of the nature of the clinic (pre-paid; gum use free and sought in treatment), cost and motivation to use gum did not contribute to non- compliance (cf Hughes article this volume). In addition, at the baseline (pre-quit) visit, approximately 1/2 hour was devoted to what nicotine gum is, what it does and how to use it. Despite the instructions at baseline, several errors in use were made. These "errors" were determined in three ways: 1) by the presence of side effects and lack of relief (reported and observed), 2) early relapse in contrast to the same individual's experience when gum was chewed as instructed, and 3) in contrast to effective use by other individuals in treatment. The consistent finding was that problems in compliance and use errors were attributable to fears and mis- conceptions about nicotine gum. The key use errors with nicotine gum are described in Table 1. Smoking and chewing simultaneously, rather than quitting cigarettes on the target date, occurred from misunderstanding replacement and be- lieving that gum could be used to cut down on cigarettes. Smoking while chewing may allow for continued reinforcement with the "bolus" nico- tine in smoke, high nicotine blood levels and concommitant psychoac- tive changes. There is also continued reinforcement of sensory/ritual phenomena. In this trial, generally, the use of any cigarettes led to in- creases in smoking and then to full-blown relapse. Underdosing. A second key error in use was chewing too few pieces of gum per day. Rarely did the smokers understand buccal administra- tion versus nicotine inhaled in smoke. Many expressed concern that chew- ing a 2 mg piece of gum would result in obtaining 10 to 100 times the nicotine they received from their cigarettes and attempted to use the fewest pieces possible. With inadequate dosing or inconsistent use, subjects reported a reappearance of withdrawal symptoms and increased craving. This often led to early relapse. To help subjects understand how nicotine gum works and to correct the problems of underdosing and fear, buccal absorption was explained again. This was followed by showing each subject 3 graphs illustrating the differences between inhaled nicotine levels and bucally absorbed nicotine. First, using data from Russell and his colleagues (6), it was possible to derive a graph contrasting plasma nicotine levels from smoking 1.2 mg cigarettes once an hour with plasma nicotine levels from chewing 2 mg nicotine gum once an hour. With the 1.2 mg cigarette, there is a spiked and treadmill-like effect with high peaks achieved within minutes of starting each cigarette and a rapid falloff rate. By contrast, a 2 mg piece of gum chewed once an hour produces no "boli" or "peaks" and there is a smooth, slow-rising level of nicotine. This visual reassures the sub- jects that buccal absorption results in lowered and slower-rising levels of nicotine. It is also stressed that gum use eliminates the damaging con- stituents of the smoke. A second graph is based on heart rate data from Krivokapich and her colleagues (5). This graph contrasted one 2 mg piece of gum with a sin- gle cigarette over 90 nunutes and served the same purpose, i.e., to demon- strate lower and slower-rising levels of nicotine with gum use. 234 1 235

A conceptual graphic was added to underscore why enough gum must be used daily. A number of clinic subjects reported that previous "cold turkey" attempts had involved severe withdrawal symptoms and a sense of "falling off a cliff:' A graph of a cliff was used to demonstrate that too few pieces per day (3-5) may not control withdrawal and the same sense of abrupt discomfort was likely to occur. Subjects were instructed that minimal use, at least 10 to 15 pieces a day, should produce a high enough replacement level to ward off symptoms and for weaning off nicotine. With heavy use of 2 mg or moderate use of 4 mg, some subjects report- ed being able to prevent the onset of symptoms. The three graphs provided a quick and effective means of comrnunicat- ing with the subjects. Understanding how gum works (buccal absorp- tion) affected both the elimination of minor errors (e.g., chewing too quickly with resultant side effects) and of major errors that lead directly to relapse. Short-term and long-term gum use. A third problem was the use of nicotine gum for too short a period of time (although gurn was provid- ed free and subjects were allowed extended access). In this trial, for those who tried to stop after one or two boxes or after a few weeks, relapse occurred. Longer use (3-9 months) appeared to ward off cravings and relapse while the subject developed coping strategies and while distance developed between the smoker and his/her last cigarette. The need for longer use is consistent with post hoc observations by Russell and his colleagues (8), Wilhelmsen and Hjalmarson (13) and from our own work (10) that several rnonths of use may be necessary. In a phy- sician study by Russell and his colleagues (7), validated success rates tri- pled to 24% at one year when at least three boxes were used compared to success with less use. However, as pointed out by Dr. Hughes, these data are retrospective and in the only prospective study (1), there was no long-term difference in success rates between one month and six months of use. Fagerstrom and Melin (1) qualify their findings by noting there was more intervention in the one-month group and thus an interactive trade-off between length of time on gum and degree of intervention. There are several issues involved in longer use, including: 1) whether there is a need for continued negative reinforcement (relief of withdrawal) in the presence of an extended withdrawal syndrome; 2) whether longer use is necessary for slow weaning -that is, a need for minimal positive reinforcement for nicotine seeking (craving) while the user develops cop- ing strategies and a nonsmoking lifestyle. Continued chewing may also allow time for confidence to develop in staying quit and time to extin- guish reactions to "triggers" to smoke. At the time of this trial, no prospective studies existed and the retrospec- tive observations pointed to better success with longer use. Therefore, subjects were allowed to use gum freely over time and were encouraged to use it for at least several months. To reduce any fears with longer use, the subjects were reminded that the dangers of smoke (carbon monox- ide, carcinogens, etc.) are eliminated with use of the gum; and, weaning off nicotine begins with the switch from cigarettes to the lower and slower rising dose of nicotine in gum. In addition, it was noted that in the 14 years since nicotine gum was introduced, and with the exception of dental problems from chewing, there have been no reports of long-term side e_f_fects. Therefore, use of nicotine-containing chewing gum, even for lengthy periods, is superior to smoking. Clearly, prospective studies on length of gum use are needed. The expectation of a "cure-all" is an obvious problem. Smoking is a complex interaction of multiple forms ofreinforcement. For most smok- ers, it is unlikely that any pharmacological agent will suffice, in itself, in producing long-term success. Relapse Crises and Re-use of Gum One potential error not described in Table 1 is the reluctance to re-use gum in crisis. Though instructed to keep gum for emergencies, the sub- jects forgot to do so or assumed nicotine could not be involved in later relapse. Nicotine has not been adequately considered in relapse occurring af- ter a period of successful abstinence. I would suggest there is a strong "niemory" for nicotine and for nicotine-induced state changes that un- derlies relapse at any stage. In particular, for relapses induced by negative affect (stress, anger, deprivation), subjects reported seeking what appeared to be relaxation induced by nicotine (vs. ritual factors). If this can be sub- stantiated, resumption of nicotine replacement using gum or an alter- native form (nicotine nasal solution, transderrnal patch) may provide a means of preventing relapse. In this trial, the resumption of gtun use did not induce a return to smoking as did attempting to smoke one or two cigarettes in a crisis. While there were successful slip-recoveries, for the most part, smoking in crisis led to a full blown relapse. Mechanism of Action The basis or rationale for nicotine replacement is to alleviate or pre- vent withdrawal symptoms and craving while the person extinguishes the smoking habit. This assumes withdrawal symptoms are nicotine- 236 1 237

specific. However, are all symptoms nicotine-specific? To what degree? What is necessary to relieve them? In our placebo controlled trial (9), we found that 2 mg nicotine gum alleviated some but not all symptoms reported by smokers. The data in Table 2 are from Schneider and Jarvik (9) and show the symptoms moni- tored and the effect of treatment with nicotine replacement vs. placebo gum. Table 2 Smoker Complaint Scale: Item Analyses for Nicotine vs. Placebo Gum in a Controlled Trial Symptom Appearance of Symptom with Cessation Effect of Treatment Annoyed p<.001 p<.008 Hostile p<.001 p<.03 Irritable p<.001 p<.05 Fluctuations in Mood p<.01 p<.05 Frustration p<.001 p<.06 Depression p <.002 p <.07 Left Out p <.02 ns Anxiety p<.001 ns Concentration p <.001 ns Disorientation p < .001 ns Restlessness p<.001 ns ~ Trouble Sleeping Concern About Weight ns Ils ns ns ~ Craving ns ns ~41 (Reprinted from Schneider and Jarvik, 1985 (9).1 N 00 Note: In contrast to the last two items on above table, three items (Urge to Smoke, Miss a Cigarette, Hunger) from the Shiffman Jarvik Scale all ~ showed significant effects of treatment, p < .05. d .p A significant result in the middle column reflects the changes in affect from baseline (while still smoking) to quitting for both groups (main efl"ect across treatments). Eleven of the 14 items tested showed an increase during the first five days of abstinence. The right hand column shows those symp- toms that were significantly relieved with active gum compared to placebo gum. This is consistent with Hughes and colleagues (3), I-latsumkami and colleagues (2), and West and colleagues (12) who also found that not all symptoms are alleviated with nicotine replacement in gum form. However, these studies and our own study involved ad lib administra- tion of 2 Ing gum. We do not know whether additional relief would oc- cur with more pieces, with fixed interval use or with a higher dose. We also do not know the degree to which nonrelief is a function of the loss of the "boli;' the high levels of nicotine or the sensory and ritual com- ponents. In the present trial, where gum use was monitored and use was adapted to needs, there were many symptoms reported and many more of them relieved with gum use (2 mg and 4 mg). Table 3 is a list of withdrawal symptoms reported by smokers in the clinic setting. Withdrawal types. While symptoms crossed over categories, there ap- peared to be a suggestion of three basic groupings of withdrawal "types": those smokers for whom anxiety and panic predominated, others for whom irritability, hostility and anger were the main problems, and a group whose complaints focused on feeling spacey, disoriented, and un- able to concentrate. These observations require rigorous, systematic test- ing. The latter group's reports of disorientation and inability to concen- trate could be important in consideration of the use of drugs like clonidine where the drug's side effects (e.g., drowsiness) might compound the problem. Dose Effects: 2 mg vs. 4 mg Nicotine Gum. Relief was related to dose of gum. By using two doses (2 mg and 4 mg) in this trial, it was possible to obtain specific feedback on the degree of relief by dose, reactions to each dose and on short-term efficacy. This section provides a summary of the clinical observations with the use of 4 mg gum. Relief of withdrawal. The 4 mg dose of nicotine gum was more ef- fective than the 2 rng gum in alleviating certain withdrawal symptoms including: disorientation, inability to concentrate, severe incidences of anxiety, tension and irritability. In the first week of treatment in partic- ular, subjects reported that most symptoms were better relieved with the 4 mg dose. The 2 mg dose was sufficient in alleviating irritability, anxi- ety and other symptoms when enough pieces were used consistently (e.g., 239 238

10 to 15 pieces daily). Characteristics of the 4 mg dose. Subjects were allowed access to 2 mg and 4 mg doses and almost all of these subjects reported that the 4 mg dose (vs. 2 mg) was characterized by the following: Quicker Onset Stronger Effect (desired action) Longer Lasting Effect Better Taste Curbed Appetite Notably, even two pieces of 2 mg gum chewed simultaneously were not perceived as equal to one 4 mg piece in terms of quicker onset, stronger Irritability Hostility Annoyance Anger Frustration Edgy Nervous Tense Restless Anxious Table 3 Withdrawal Symptoms Lighthcadedness Disoriented Unable to Concentrate Spacey Slowed Down Lethargic Panicky Out of Control Fluctuations in Mood Headaches Constipation Trouble Sleeping Esoteric symptoms: e.g., Depressed Sad Sense of Loss Tearfulness Hunger Craving for Sweets Craving a Cigarette Urge to Smoke Missing a Cigarette Sweating Bodily Achcs and Pain Tingling/numbness effect and better taste. Nemeth-Coslett and Henningfield of the Addic- tion Research Unit in Baltimore are currently in the process of systemat- ically examining some of these differences. Relapse prevention and dose. The 4 mg dose was reportedly more effective in warding off relapse particularly when a relapse crisis ema- nated from strong negative affect. This was noted both in the beginning of cessation and in relapse crises long after initial cessation. While there could be an expectation effect associated with stronger taste, subjects claimed that craving was relieved faster and more sufficiently with the higher dose. Side effects. The 4 mg dose was more likely to produce hiccups and gas than the 2 mg gum, particularly with improper chewing (chewing too fast). With either dose and proper chewing, nicotine side effects (vs. eflects of chewing a gum) could be controlled entirely. In terms of chewing per se, there was less jaw irritation reported with 4 mg use than with 2 ing. The 4 ing giun occasionally produced euphoria and lightheaded- ness not reported with 2 mg gum. Appetite suppression. Subjects had been asked to use gum when they felt an urge to eat. The 4 mg dose reduced appetite as compared to the 2 mg gum. With consistent use of 4 mg or use of a significant number of 2 mg pieces (12 or more pieces), individuals reported little weight gain. As soon as they reduced their intake, food consumption increased and/or weight gain was reported (unverified self report). Combined Method for 2 mg and 4 mg Gum Use In the beginning of treatment, subjects were given boxes of 2 mg and 4 mg gum in an effort to observe their choices in dose and use (e.g., when, why). A few subjects who chose to use only the 4 mg dose appeared to like and crave it. However, at six weeks, those subjects who chewed only the 4 mg dose reported more cravings for cigarettes than those on 2 mg gum for the same amount of time. In an effort to eliminate the craving, subjects tried to cut down from ten 4 nig pieces to five 4 mg pieces. This abrupt change in dose brought on a reported increase in withdrawal. This is consistent with I7r. Hughes' r.eport and our own pilot experiences with abrupt cessation of 15 pieces of 2 mg, that terminating gum use abrupt- ly will induce withdrawal. As an alternative, subjects were asked to decrease from ten 4 mg pieces to ten 2 mg pieces and the cravings disap- Peared without problems. 240 1 241

The best combined replacement strategy in this trial was achieved by having subjects alternate 2 mg and 4 mg pieces for the first 3 or 4 weeks and then reduce 4 tng use. Thereafter, 4 mg use was restricted (by limit- ed access) to use in crisis and 2 mg pieces were chewed for the duration of treatment. I found 4 mg nicotine gum an extremely useful, if not essential, dose in helping the smokers in this clinic quit. It was superior to 2 mg in 1) relief- allowing for a smoother transition from cigarettes in the begin- ning of cessation and 2) in warding off relapse at any point. Dependence on Gum Dependence on gum occurred occasionally, probably because subjects were allowed access to long-term use. In this instance, dependence was operationally defined as an unwillingness to give up nicotine gum after a targeted time to do so. It was usually the heavier, long time smokers or smokers under chronic stress who resisted giving up gum. Ironically, the better the gum worked in the beginning of treatment, the more fearful the individuals were of giving it up and the more success was attributed to gum rather than to personal achievement. In the 30-40 year smokers, it appeared more time was necessary for extinction of both internal and external responses to the cues to smoke. In the trial, "dependent" users were allowed continued access for six months or longer but treatment visits were focused on eliminating resistance to developing coping strate- gies and to building confidence. As noted earlier, if abrupt reduction in 4 mg gum occurred or abrupt reduction of high levels of 2 mg gum (10-15 pieces), then withdrawal symptoms appeared. This indicated the individual was still at a level of nicotine intake that maintained part of his/her original dependence on nicotine in cigarettes. In that the purpose of nicotine gum is to wean off cigarettes, this was expected. That is, withdrawal would be observed with abrupt cessation until levels are sufficiently reduced. Weaning In contrast to quitting smoking (which involves so many reinforcers), stopping gum, for most subjects in this trial, was relatively easy. Slow weaning will prevent the reappearance of withdrawal. There are a num- ber of methods which can be used for weaning from gum: 1) using few- er pieces over time, 2) chewing for 10 vs. 30 minutes to obtain less nico- tine, 3) alternating with regular gum, and 4) cutting the pieces in half The last was used in a two-step process: first fixing the schedule of use to break up associations or habits; then cutting the pieces in half to create a low steady-state dose from which further weaning is easy. Problems in "Cold Turkey" Cessation Alleviated With Gum Use In addition to relief of withdrawal and craving, a number of problems encountered in "cold turkey" cessation were eliminated or reduced for most of these subjects when gum was used. These are summarized below. Problems Controlled With Gum Use Withdrawal symptoms Handling Eating Coffee drinking Stress in difficult situations Marijuana use The most surprising problem to disappear with gum use was the one of "handling" For those people for whom "what to do with their hands" was troublesomee in past attempts to quit, not one experienced the problem in this trial. Although behavioral substitutes and methods were offered, they proved unnecessary. It appears that handling and other behavioral difficulties can be a manifestation of the tension, anxiety and discom- forts of nicotine-specific withdrawal. In general, with the reduction of withdrawal symptoms and with the use of gum per se, subjects reported being better able to handle stress, resist cues to smoke and were less fearful of quitting. In the clinic, the subjects seemed less consumed with withdrawal symptoms and turned their fo- cus to behavioral change during treatment. This is a significant advan- tage over spending treatment time coping with withdrawal. The above list summarizes problems experienced in "cold turkey" at- tempts to quit that are reduced with gum use. In this sense, the marijuana finding may be out of place on this list, i.e., nonrelapse with marijuana may hold true whether or not nicotine gum is used to relieve withdrawal and craving. In the 20 or so individuals who smoked marijuana during the course of treatment, none reported relapse to cigarettes at the time or within short periods following use (e.g., within the week). This was somewhat validated by continued nonsmoking carbon monoxide levels at subsequent visits. While marijuana may differ in taste, aroma and style ofsmoking, there are "cues" such as matches, ashtrays and the visuals of the smoke that might be expected to induce relapse if psychological 242 1 243

factors were paramount. It is suggested that relapse did not occur be- cause nicotine was not present and the effects of inhaled nicotine were not produced. These observations seem to indict tucotine in processes that were previ- ously perceived as nonpharmacological or solely behavioral issues in smoking cessation. It will be interesting to see if a non-self-manipulated form of nicotine replacement, such as the transdermal patch, will sup- port these observations. Continued High Relapse Risks Independent of Gum Use While marijuana did not trigger relapse, alcohol consistently present- ed a problem. A number of cessation problems remained serious, indepen- dent of nicotine replacement: Alcohol Old companions smoking Surrounded by smokers at work or home Stress (acute, prolonged, unanticipated) Celebration or reward Feelings of deprivation Boredom Of the above, the most predictable and constant precursor of relapse was stress in any form. Feelings of deprivation and boredom triggered relapse most often in those who lived alone or had unsatisfactory lifestyles. Resistance to cigarettes in the constant presence of other smokers was a serious problem. Workplace nonsmoking codes may ultimately benefit the smoker who wants to quit. Coping Strategies Coping skills, alternative rewards and problem solving ability were crit- ical for long-term relapse prevention. There were a number of coping strategies that were suggested and reported by subjects as useful during the individual counseling. These were given "hooks" or "iucknames" to promote recall for their use, such as: "fast food" coping, "family feud" and "rote thought." I will mention these briefly. "Fast food" coping was designed to pro- vide subjects with a coping mechanism that could be utilized in seconds or minutes (quick breath, ice water, stretching). Deep breathing, a shower or meditation are unworkable responses to a phone call and many other triggers to smoke. A means of testing levels of coping skills was introduced into our laboratory by Dr. Saul Shiffman; this was converted into a variation of the TV game "Family Feud." Subjects were presented with a potential relapse crisis and given several minutes to list the most reasonable alter- natives to smoking. Then they were asked to identify for themselves potential relapse episodes and continue the game by responding to each. In the beginning of cessation, responses were minimal and often unfeasi- ble; as cessation proceeded, individuals became more adept at provid- ing alternatives to smoking. The aim of this test, to develop and rehearse personalized coping strategies, is well served by an individualized treat- ment approach. Because of non-use of previously learned coping skills in crisis situa- tions observed here and also reported by Shiffman (11), a technique called "rote thought" was instituted. The idea was to program the individual "rotely" (defined as something learned by memorizing) to go to a fixed thought to stop relapse. Most often the individual rehearsed in advance the idea that when he/she thinks "to hell with it;" they will stop them- selves. While very simple, the preprogramnung of resistance provided a behavior stopping technique in crisis. The individual could then pro- ceed to use already developed coping strategies. A Note on Prediction of Relapse The "promise" was discovered in questioning subjects about delayed relapse episodes. Many subjects revealed they had made themselves a promise that "if things got bad enough they could smoke." The existence of this promise was the single best predictor of long-term relapse in the study. In the beginning of treatment, subjects were allowed to think in terms of one day at a time. Later on (2-4 months), they were encouraged to get rid of the promise of relapse in crisis. The acknowledgement of this "promise" helped to decrease its occurrence. Individual vs. Group Treatment If the findings of this trial can hold up to empirical test, they should apply to groups as well as to individual treatment. As a technique for quit- ting per se, the question is this: is individual therapy with nicotine gum a useful and necessary approach? There arc definite advantages. It brings in smokers who want individual care and who would not participate in group treatment. It allows for well- guided use of gum and development of personalized coping strategies. It enhances the potential for relapse prevention through access to ongo- ing nlaintenance. 244 1 245

However, this kind of individual treatment, viz., many hour-long ses- sions, is not cost effective, lacks support systems (buddies, groups) and may not be necessary. For example, the outcome results in this trial will not exceed 50% at one year. Killen and colleagues (4) found a 50% suc- cess rate with fewer sessions, conducted in groups, indicating that a com_ bined group/pharmacological approach may be more cost effective. A word of caution is needed in that the groups may not be comparable and thus it may not be reasonable to compare outcomes. Table 4 Research Needs Withdrawal: nature, course, duration Nicotine-specific Withdrawal: Which symptoms? Dose effects? Parameters for nicotine gum use: Dose (& changes over time) Fixed vs. Ad Lib or Both? How Long? For Whom? Support: What Kind? At What Stage? Nicotine in long-term relapse prevention. Success rates: (post-treatment vs. post-cessation) Route of nicotine: gum, transdermal, nasal Speed of reinforcement Steady-state vs. stimulus-response Compliance by route Relief and Success Nicotine Effects: True action or relief? Research Needs My intent in this open trial was to identify, for future study, both the common and esoteric nature of withdrawal and the effectiveness of and problems with nicotine replacement in gum. I am surprised at the lack of information on the parameters of withdrawal and of nicotine replace- ment while focus proceeds on efficacy testing. Table 4 summarizes some of the research needs in this area. Basics such as dose/scheduling func- tions should probably preceed or be tested concurrently with outcome. Conclusion It has been shown that nicotine gum can be a valuable adjunct in treat- ment for smoking cessation. However, problems with use continue and may be undermining the potential of this preparation as an aid in smok- ing cessation. With more precise guidelines for use and better compli- ance, success rates for physician intervention and for a combined phar- macological/behavioral clinic approach could be . enhanced. The effectiveness and efficacy of the higher dose of nicotine gum (4 mg) as well as other forms of nicotine replacement should be fully explored in treatment. References 1. Fagerstrom K-O, Melin B. Nicotine chewing gum in smoking ces- sation: Efficiency, nicotine dependence, therapy duration and clini- cal recommendations. Pp. 102-109 in J Grabowski, S Hall (1;1is.) Phar- macological Adjuncts in Smoking Cessation: NIDA Monograph Research Series #53. Washington DC: US Government Printing Office, 1985. 2. Hatsukami D, Hughes J, Pickens R. Characterization of tobacco withdrawal: Physiological and subjective effects. Pp. 56-67 in J Grabowski, S Hall (Eds.) Pharmacological Adjuncts in Smoking Cessa- tion: NIDA Monograph Research Series #53. Washington DC: US Government Printing Office, 1985. 3. Hughes J, Hatsukami DK, Pickens RW, Krahn D, Malin S, Luknic A. Effect of nicotine on the tobacco withdrawal syndrome. Psy- chopharm. 1984; 83:82-87. 4. Killen JD, Maccoby N, Taylor CB. Nicotine gum and self-regulation training in smoking relapse prevention. Beh. Ther. 1984; 15:234-248. 5. Krivokapich J, Schneider NG, Child JS, Jarvik ME. Cardiovascular effects of nicotine gum and cigarettes assessed by ECG and echocardi- ography. Pp. 42-55 in J Grabowski, S Hall (Eds.) Pharmacological Ad- juncts in Smoking Cessation: NIDA Monograph Research Series #53. Washington DC: US Government Printing Office, 1985. 6. Russell MAH, Fcyerabend C, Cole PV. Plasma nicotine levels after cigarette smoking and chewing nicotine gum. Brit. Med. J. 1976; 1:1043-1046. 246 1 247

7. Russell MAH, Merriman R, Stapleton J, Taylor W. Effect of nicotine chewing gum as an adjunct to general practitioners' advice against smoking. Brit, Med. J. 1983; 287:1782-1785. 8. Russell MAH, Raw M, Jarvis MJ. Clinical use of nicotine chewing gum. Brit. Med. J. 1980; 280:1599-1602. 9. Schneider NG, Jarvik ME. Nicotine gum vs. placebo gum: compar- isons of withdrawal symptoms and success rates. Pp. 83-101 in J Grabowski, S Hall (Eds.) Pharmacological Adjuncts in Smoking Cessation: NIDA Monograph Research Series #53. Washington DC: US Govern- ment Printing Office, 1985. 10. Schneider NG, Jarvik ME, Forsythe AB, Read LL, Elliott ME, Schweiger A. Nicotine gum in smoking cessation: A placebo- controlled double-blind trial. Add. Beh. 1983; 8:253-262. 11. Shiffman S, Read L, Maltese J, Rapkin D, Jarvik ME. Preventing relapse in ex-smokers: A self-management approach. Pp. 472-520 in GA Marlett and JR Gordon (Eds.) Relapse Prevention. New York: Guil- ford Press, 1985. 12. West RJ, Jarvis MJ, Russell MAH, Carruthers ME, and Feyerabend C. Effect of nicotine replacement on the cigarette withdrawal syn- drorne. Brit. J. Add. 1984; 79:215-219. 13. Wilhelmscn L, Hajlmarsson A. A smoking cessation experience in Sweden. Canadian Fam. Physician 1980; 26:737-743.. 248 Nicotine Gum in the Workplace: Prelim.inary Report of Two Randomized Trials Stephen Sutton, Ph.D. Robert Hallett, Ph.D. Addiction Research Unit Institute of Psychiatry London Introduction Nicotine chewing gum has been evaluated in a number of different set- tings, including smoking cessation clinics and general medical practices, with promising results (1,2). This paper reports preliminary findings from two controlled studies conducted in a workplace setting. Both studies were randomized trials of the offer of a brief treatment course based on nicotine gum. There is already substantial evidence that the gurri has more than simply a placebo effect (1,2), and since our aitn was to evaluate the offer of treatment, not to ascertain whether any treatment effect observed could be attributed to the pharmacological effect of the gum, we did not include a placebo control condition. It should be stressed that the find- ings reported here are preliminary and that fuller reports of both studies are in preparation. Method The first study was carried out at the London head office of a large oil company. One hundred and sixty-one cigarette smokers who had watched a videotape about smoking but who were still smoking at the time of the three-month follow-up stage were randomly divided into two groups. The subjects in the first group were sent a personal invitation from the occupational physician to take part in a brief treatment program based on the use of nicotine chewing gum. No invitation was sent to smokers in the second group who constituted a randomized non- intervention control group. The consultations were given by the occupational health nurses, one nurse being present at each consultation. I3efore the treatment began, the nurses were given a typed sheet which contained recommendations on 249

® the schedule and content of the consultations. The recommended schedulc was four consultations over a 12-week period. Each consultation was to take a maximum of 15 minutes, so that the total titne per patient for the full treatment course would be no more than one hour. In the first consultation, the rationale for the gum was explained and in- structions were given on how it should be used. Subjects were given a pri- vate prescription for one box (105 pieces) of 2 mg Nicorette, which they were told would cost about six pounds sterling, and a copy of the manu- facturer's booklet. Subsequent consultations were used to give encourage- ment and support, to check that subjects were using the gum correctly, to deal with any problems, and to give further prescriptions for the gum. Thirty-two of the invited group attended at least one consultation. There was a small amount ofcontamination between groups in that five members ofthe control group also asked for and were given treatment. One year after the treatment course began, subjects were interviewed about their smoking, and ifthey reported having stopped smoking their expired-air carbon mon- oxide (CO) level was also measured (3). If reported stoppers had a CO of greater than 10 ppm they were considered to be continuing smokers. The second study was carried out at the London head office of a large retailing company. Three hundred and thirty-four cigarette smokers were recruited by means of a questionnaire survey of the work force. We used the same design in the second study as in the previous onc. Of the 334 volunteers, a randomly chosen 270 were sent a personal invitation to take part in the treatment course; the remaining 64 constituted a randomised control group. Of the 270 who were invited, 172 attended for treatment. In this study, the course consisted of only two consultations two weeks apart. The consultations were given by the authors (both psychologists), not by the occupational health staff. We decided to do the consultations ourselves because our experience had suggested that occupational health staff would not have time to treat a large number of smokers in a short time. A maximum of 30 minutes was allowed for the first consultation and 15 minutes for the second. CO levels were measured on both occa- sions. As in the first study, there was a small amount of contamination between groups itt that four members of the control group asked for and were given treatment. All subjects, including those who did not attend for treatment and those in the control group, were followcd up one year after the start of treatment. Results The outcome results for both studies are shown in the table. There were effectively three groups (A, B, and C). Group A are the people who were 250 invited and who attended at least one consultation; Group B are those who were invited but who did not take up the invitation; and Group C are those who were not sent an invitation (the control group). We used two criteria of sustained abstinence. According to the "lenient" criteri- on, successes were those who claimed to have smoked no more than 20 cigarettes or five cigars in total throughout the follow-up period. "Strict" abstainers were those who claimed to have been totally abstinent from tobacco throughout the follow-up period. All those who were success- ful according to either definition had one-year CO levels no higher than 10 ppm, giving a "deception" rate of zero. Table 1 Outcome Results for Two Randomised Studies of the Offer of Nicotine Gum Treatment. Study Outcomea Invited Control Significance of criterion Attended Did not (A + B) vs. C attend Chi2 Fisher'sb (A) (B) (C) exact test Study 1 N= 32 N= 47 N= 82 Lenient 19% 2% 2%` P<.10 ' p = .075 (6) (1) (2) Strict 16% 0"%> 2%` P<.30 p = .206 (5) (0) (2) Study 2 N= 172 N= 98 N= 64 Lenient 13% 2"/<, 2%o'l P<.05 p=.029 (22) (1) (1) Strict 12% 11% 2`%," P<.08 p=.052 (20) (1) (1) Pooled N= 204 N= 145 N = 146 Lenient 14% 2%) 2"/<, P<.01 p=.00 (28%,) (3) (3) Strict 12% 1'% 2% P<.02 p=.012 (25) (1) (3) -'See text for definitions of "lenient" and "strict" 6One-tailcd. `'Five people in the control group asked for and were given trcauncrit. One of these was a long-tcrm abstainer and is classified as a control group succcss. "Four people in the control group asked for and were given treatment. One of these was a long-term abstainer and is classified as a control group success. 251

As the table shows, the pattern of results in the two studies was very similar, the sustained abstinence rates in Groups B and C being extremely low in both cases. Of those who attended for treatment (Group A), on the other hand, 16% and 12% respectively were totally abstinent through- out the follow-up period. The table gives the significance results for the comparison of the invited group as a whole (A + B) and the randomised control group (C). This comparison was statistically significant only for the second study and then only for the lenient criterion; for the strict criterion it was marginal. Pooling the results of the two studies yielded clearly significant differences for (A + B) versus C with regard to both criteria. The comparison of the treated group (Group A) and the control group (Group C) was highly significant across each criterion and each study. These tests are conservative in that those in the control group who "un- officially" received treatment are included in the analysis. If they are ex- cluded, the comparison of (A + B) and C is significant for each com- parison except for that involving the strict criterion in Study 1. Conclusion To our knowledge, these studies are the first controlled evaluations of a workplace smoking intervention program using nicotine gum. The results of the two studies were consistent and suggest that the offer of treatment based on nicotine gum increases the long-term success rate in this setting. Furthermore, the pattern of results (a virtually zero absti- nence rate except among those who attended for treatment) suggests that the higher success rate among the treated group was due to the treatment itself rather than to self-selection into treatment. Although subjects were treated on an individual basis rather than in groups, the treatment was not intensive, particularly in the second study where it consisted of only two consultations with a maximum of 45 minutes spent with each pa- tient. In our view, these results arc encouraging and suggest that this ap- proach is worth further investigation. Acknowledgments The two companies involved in our studies were: Shell UK Limited, Shell-Mex House, London (Dr. R. Houston, 1Jr. K.H.M. Young, Prof. D.H. Elliott, Ms. K. Roll, Mrs. A. Allard; Sisters P. Nowland and N. Teaklc administered the nicotine gum treatment); and Marks and Spencer plc, Michael House, Baker Street, London (Dr. F.G. Taylor, Dr. A.C. Houston, Dr. J. Haslehurst, Mrs. J. Burr). We thank the employces who partici- pated in our studies and the occupational health staff who collaborated with us. We also thank our colleagues Michael Russell and Martin Jar- vis for their helpful comments on an earlier version of our paper. The Medical Research Council, London, England provided financial support. References 1. Grabowski J, Hall, SM. Pharmacological Adjuncts in Smoking Cessation. Rockville, MD: U.S. Department of Health and Human Services, NIDA Research Monograph 53, 1985. 2. Hughes JR, Miller SA. Nicotine gum to help stop smoking. J. Am. Med. Assoc. 1984; 252:2855-2858. 3. Jarvis MJ, Russell MAH, Saloojec Y Expired Air Carbon Monox- ide: A Simple Breath Test ofTobacco Smoke Intake. Brit. Med. J. 1980; 281:484-5. 252 1 253

Discussion: Integrating Pharmacologic and Behavioral Approaches Chair: M.A.H. Russell, M.R.C.P. Institute of Psychiatry, London Presenters: Edward Lichtenstein, Ph.D. University of Oregon, Eugene Douglas M.C. Wilson, M.D. and J. Allan Best, Ph.D. Chedoke-McMaster Hospitals, Hamilton, Ontario and University of Waterloo, Ontario Stephen Sutton, Ph.D. Institute vf Psychiatry, London Peter Hajek, Ph.D. Institute of Psychiatry, London Nina G. Schneider, Ph.D. University of California, Los Angeles DR. RUSSELL: Okay, the first person at the mike. CHRIS STEELE: I'm speaking as a family physician in Manchester, and as a doctor who runs two stop-smoking clinics based in hospitals. I have actually treated over 1,500 patients with Nicorette, and with that ex- pericnce behind me, I would like to endorse some points that have been made by Nina Schneider, by Peter Hajek and also possibly address some of the problems that John Hughes raised yesterday. I'd advise physicians to follow what I would call a basic three-point plan ifyou want to help your patients to quit in the confines of the office. Choose your patients correctly, teach them to chew the gum correctly, and, for God's sake, follow them up correctly. I think a very simple way to find out how addicted is a patient is to use the Fagcrstr6m questionnaire but take it right down to one question. Ask that patient, "Flow early in the day do you have your first cigarette?" That will give you an indication straight away as to how addicted that patient is. And then a second question I would ask the patient as a predictor is, "I)oes your nearest-and-dearest smoke?" If the nearest-and-dcarest does smoke and is not willing to quit at the same time as the patient, then that is a predictor of possible negative outcome. I would like to just say one or two words about the use of the gum. The problems that John Hughes mentioned and which have been high- lighted in this congress-problems affecting the stomach and digestion, acid regurgitation, hiccups, the local problems in the mouth of ulcers, aching jaws, sore tongue-are all problems of overuse of the gum and ovcrchewing. It must be stressed to patients that they ought to chew very slowly, indeed. If they continue to have problems when they chew slowly, simple guidelines would be: if the gum is still causing you problems, halvc it and take half a piece each time instead of a full piece, mix it with ordi- nary chewing gum to dilute the flavor or park it in the corner of the mouth-stop chewing it and just leave it there. Although this will release nicotine more slowly, it is very often enough to help the patient get through the withdrawal phase. Finally, in follow-up, I would like to stress that patients should be seen within the first week of quitting. And, I think the gum should be used in far higher quantities. Patients who say that the gum is no use, it's not helping them, are very often only using five or six pieces per day when they'd been smoking 30 cigarettes per day, and, therefore, there is a place here for the 4 mg gum. Doctors shouldn't be afraid of using the 4 mg dose and patients shouldn't be afraid of using it, either. DR. SCHNEIDER: I want to make one response to what Chris said about mouth ulcers. Mouth ulcers have been reported as a function of smoking cessation in a lot of the literature. So to consider it a side effect of gum is not necessarily true at all. SPEAKER: I would like to put a question to Dr. Wilson. Did you try to stratify your patients, your results, according to the fact that the doc- tor, the family doctor, is a smoker, an cx-smoker, never-smoker? DR. WILSON: Yes, we looked at quite a number of physician charac- teristics, one of which was current smoking and ex-smoking, and stratified on the basis of only two or three key variables across all three groups. There were not significant differences among the physicians in all three groups. There are a lot of ex-smokers in Canada who are physicians. Biochemical validation RUTH MASON: It would be very helpful to members of the press if each of the presenters could tell us the dates of their studies; and my second question is, for Dr. Sutton: Is it my understanding that it's usual to measure CO levels when you measure smoking cessation levels at the end of the study? Was your doing that based on the researcher who paid a surprise visit to people after six months and measured their CO levels and found that a lot of people had lied about the fact that they had quit smoking? 254 1 255

DR. SUTTON: This is becoming routine in smoking research. I sup- pose about ten years ago it would be rare in outcome studies for smok- ers to actually be tested biochemically to see whether they had in fact stopped smoking. But these days it's becoming routine and most studies that are looking at the efficacy of different treatment programs should include some sort of biochemical validation to check that people are in fact not smoking. DR. RUSSELL: Would anyone want to respond to that request for the dates of their programs? DR. WILSON: The project that I reported among family physicians in Canada is currently ongoing. Recruitment started earlier this year; training, in 1985. The one-year outcomes will be reported about a year from now. DR. SCHNEIDER: Mine is ongoing. It started in 1984, ending in April, 1986. DR. SUTTON: The two studies I talked about were done in the last two years, not yet published. DR. HAJEK: My study is currently in press and will be published by the end of this year. Patients refusing gum DR. RUSSELL: I see Judy Ockene is at the microphone. Judy has done so very much work in this field and suffered slightly from very tight pro- gram pruning, so I'm going to make an exception in her case. If she has got anything more than just a question or a brief comment ...? JUDITH OCKENE: That's very kind of you, Michael. As Michael is noting, we also have been active in a physician-delivered smoking in- tervention study, very generously funded by the National Cancer Insti- tute. We are training residents in family practice and general medicine, and so far we have trained 70 out of 80 residents to counsel patients on smoking cessation. We are actually training them in how to advise pa- tients to stop smoking, also how to counsel patients to stop smoking, and the third arm is how to counsel and prescribe nicotine gum to help their patients stop smoking. We've been very encouraged by the physi- cians' training outcomes, which we have evaluated. Each of them has de- veloped very good skills in counseling their patients and also in prescrib- ing Nicorctte gum. The one outcome that I'd like to question the panel on and one which I think is similar to what Alan Best and Doug Wilson are reporting from their study is the percentage of patients refusing gum. Even though the physicians are trained very carefully in how to prescribe Nicorette glus- of those patients who are actually randomized into the counseling-p Nicorette gum, about 50% of them actually refuse the use of the gum which is being offered free to them. So although we see that it could be a very useful intervention, I think it's still important that, no matter how well we're training physicians - and I think that we have a luxury in the amount of time that we're able to use to train the physicians-patients are still not electing in at least half the cases to actually try the gum. DR. SCHNEIDER: I would like to say first that we need motivated patients. But, second, without meaning to offend anyone, I find that phy- sicians usually insult the patients and don't explain what they need to know. They need to know something about buccal absorption and why it doesn't yield ten times or a hundred times their usual dose. So you can give someone gum and say, chew slowly, do this, do this, but if you don't explain why, they're going to go home and get scared. That's what I found, every time they call. DR. OCKENE: But, in fact, our physicians are trained to describe ex- actly what's happening with the use of the gum. DR. SCHNEIDER: Then I refer to what Chris said about follow-up the first week and correcting use, because I found that redundancy is crit- ical. You can explain everything and patients come back in the first week and it's almost as if you hadn't said anything. DR. WILSON: The physicians in our experimental gum-plus group were trained to discuss all of the possible side effects and implications of the Nicorette. I thought there might be a difference in the gum users in favor of the gum-only physicians because that's really all we told them to offer their patients, was advice plus Nicorette, as opposed to the trained physicians who were offering the patients a choice of one of three things in the contract. One is that they would continue to smoke and if they continued to smoke, they would be given the opportunity to come back in two or three weeks to discuss that again. Or they would choose to quit but they could choose to quit with the offer of the Nicorette gum or without. So it was in a sense engaging the patient, negotiating with the patient in a contract sense. It's interesting that even among the trained physicians, the amount of gum use is higher, and the cessation is higher. DR. LICHTENSTEIN: One of the things that I'm acutely aware of in the smoking cessation literature generally is that we're rather ostrich- like. We've done almost no market research at all in terms of what it is that people out there want, what factors facilitate or serve as barriers to treatment. You're right, there are a lot of patients who apparently may not want the gum for some reason and we don't know anything about that. One of the side studies that we're doing with this involves work with small groups of patients and study them very intensively as they 256 1 257

go through the cessation process so that we can start to get some hints about some of these problems. I think you're pointing your finger at an important area. One of the particular problems perhaps with this one is that gum is being sprung unawares on them. They've come for something else and all of a sudden the doctor is talking about not only smoking, which is threatening itself, but also some strange new tool. I wonder if you had a more sensitively tailored approach where physicians would especially encourage patients to talk about concerns they had about the gum. But then you're getting into that tradeoff area where you have to get them back and physicians have to do extra, and who knows? DR. WILSON: I think a comparison of interest, though, would be the percentage of patients who make a quit attempt subsequent to the phy- sician intervening with and without the offer of gum. I think it's been indicated clearly that not all patients coming into a doctor's office are going to be interested and it's hard to evaluate what 50% means without some reference point. DR. RUSSELL: Have the panel had their say on that? Thanks very much, Judy. Dosing regimen instruction DR. BENOWITZ: I'd like to just offer a couple of observations and ask a question, coming from a person who has been interested in therapeu- tics with drugs but not involved in any of the trials with Nicorette gum. It seems clear that a number of people are interested in issues of 2 mg gum and 4 mg gum, and Nina began to address some of the issues about dosing schedules and whether people are taking it frequently enough. But in very few of the presentations have I heard anything about how the patients or clients are really instructed about the use of the gum, whether they use it when they want to, whether they're put on a fixed regimen. It seems to me that an issue like this is really critical. The doc- tors get advice in the Merrell Dow handout about how to use the gum and some of them follow that and some of them don't. But most patients will do only what they're definitcly instructed to do, and if it's very vague, then patients will often not comply at all. Is anyone doing a systematic trial of the dosing regimen instructions? Do you give someone 12 pieces a day and tell them they must take one per hour and compare that to taking it on an as-desired basis? DR. SCHNEIDER: It's on my list to do fixed versus ad lib dose and scheduling. DR. BENOWITZ: It seems to me that if this is going to bccorne a legitimate form of therapy that a doctor in his or her office can do, he or she has to have some very clear guidelines about how to do it, and I don't think they're available. DR. WILSON: I think we're giving our trained physicians as clear guide- lines as we understand from the literature. In terms of actual number of pieces per day, there are not absolutely clear guidelines. However, one that we use is that the smoker may want to chew about half the number of pieces of Nicorette per day as he smokes cigarettes, and so that the 20-a-day smoker would chew 10 pieces a day and 30 would chew 15 2 mg pieces. Those are the guidelines we give to our physicians who use that with their patients. DR. BENOWITZ: Do you advise your patients to set themselves a schedule and dose themselves regularly or use it when they have an urge to smoke? I think those are critical differences. DR. WILSON: We encourage the physicians to tell their patients to use that number of pieces per day and to space them accordingly. They may find that there are some situations when they have extreme craving when they would want an additional piece. So it's a little bit of both. DR. SCHNEIDER: Towards the end of treatment I started telling them to use it ad lib but don't go X amount of time without a piece of gum, according to how they were feeling and what stage in treatmerit: Patients' needs change from the first week to the third to the second month. DR. BENOWITZ: Joel Killen and a colleague in another NCI-funded trial, have included that as a comparison in a self-help kind of approach they're using. The role of the general practitioner DR. RUSSELL: Any more comments from the panel on that? Thank you, Neal. The next person, please. DR. BITTOUN: I've been very interested by the presentation of Douglas Wilson which shows the role of the general practitioner because it reaches the community. Three brief questions of him. The first one. Is the rate of success with trained GPs the same as the rate of success in the smok- ing clinic? Second one: Did you calculate the cost of this project with trained GPs in your country? And third: What is your suggestion for countries which have a low number of GPs? DR. WILSON: Were you asking if there was any comparison between the outcomes of the trained physicians and smoking cessation with the clinic? I think it would be somewhat foolish to compare the outcomes 258 '1 259

because of the nature of the populations entered into each of those con- ditions. In our study all cigarette smokers attending a farnily physician's office, whether they wanted to quit or had ever thought about quitting or not, were randomized into entering into the study. Those smokers at- tending clinics would be self-referred motivated smokers who were at- tending because they want to quit. And so because of that difference, I'in not sure that comparison would be meaningfiil. The clinics that I'm aware of in Canada don't go beyond self-reporting, don't include dropouts or those lost to follow-up in their total figures and there is no biochemical validation. So, again, I think it's difficult to compare the 12-month out- comes, as it is difficult to compare all of the studies that have been present- ed here over the past two days. The second question related to cost. We're fortunate in Canada to have a fairly comprehensive health care scheme which reimburses physicians for about 90% of the fees. The fees in Canada I believe are considerably less than the fees in the United States but probably more than the fees in the U.K. The cost of an initial visit which is considered to be an intermediate assessment in Canada-and that would take somewhere between seven and 12 minutes-would be about $15 and the physician would be reim- bursed $15 for that visit. The remaining four visits would be paid as a minor assessment and that would be something like $12. So if the pa- tient attended all visits, the cost is under $75, which is about the cost of room service at the Waldorf-Astoria for a hamburger and french fries. The third question-in countries where there are low numbers of GPs, I think I would suggest that other primary care workers be involved, whether they be nurse practitioners or primary care physicians of other disciplines. I think it's the primary care health professional who can play a key role with smokers in any country. I'm not sure about China. Dosage rotation DR. GRUNBERG: I was particularly interested in something that Nina mentioned on rotating the 2 and 4 mg gums because two phenomena came to mind. One is differential reinforcement schedules used in animal self-administration paradigms of trying to use self-administered drugs of dependence in order to give differential dosage, or altering the dosage from day to day or hour to hour would in effect change the schedule. And secondly, that the rotation of the dosages, 2 and 4 and perhaps 2, 4 and 0, would possibly allow for the 4 mg gum every few days to have a greater pharmacologic action because of lack of habituation to the dosage, thus working from both a behavioral paradigm and of drug self- administration of variable schedules of reinforcement plus the pharmaco- logic action. I was curious if anybody else has attempted that kind of ro- tation of dosage, and that would be the first part. And second, if they haven't, I encourage people to try this, because I think, just on a theoret- ical basis, it might have particular usefulness. Are there any data on this? Has anybody tried this? SPEAKER: Consider the implications, though, of doing that in the usual way that the gum is given, which is, "Here, go take it;" and the patient is never seen again. DR. GRUNBERG: I was thinking, obviously, within the context of a weekly visit in which the package might not be just 4 mg or 2 mg but instead the package would have 10 pieces of gum, 5 of which are 4 mg, 3 of which are 2 mg, 2 of which are 0 mg. The gum is given in a rotated fashion and therefore the drug administration would possibly have a great- er pharmacologic action and the behavioral reinforcement effect, based on any animal self-dependence literature, could theoretically be more in- fluential. ED LICHTENSTEIN: It makes good theoretical sense but, again, you've got a regime-in which the data indicate already-that neither patient nor doctor follow very well. DR. GRUNBERG: But, you see, Ed, from a variable reinforcement schedule, you should get a better compliance because of stronger rein- forcement on a variable schedule. DR. SCHNEIDER: I did it for a different reason than that- I wanted them to stop begging me for 4 rng gum and getting so crazy about it. But if you packaged it so that they were given a box of 2 and 4 at the beginning, not only would that discourage them from trying to use too few a day, but also they could use the 4 ing ad lib at the critical times like in the morning and under stress in the first month. Then you just don't refill the 4 mg prescription unless you think you've got a heavy smoker or whatever you discover. DR. GRUNBERG: Well, either way, I just encourage the considera- tion, particularly in light of Neal Benowitz's comment. DR. RUSSELL: We'11 just take one more question or comment from Saul Shiffman. Physician training DR. SHIFFMAN: Let me ask of Dr. Wilson and Dr. Best: I was very interested in your presentation, but I didn't get a sense of what you ac- tually tell physicians to do, much less what they actually do when con- fronted with a patient. That is, how is the topic introduced? It seems to 260 1 261

me there are a lot of critical process variables which I have the sense you've attended to but I'm not clear on exactly what you teach. Could you give us a brief summary of that? DR. BEST I'm not sure, with the time constraints that we have that it would be fair to try to describe what it is we teach the physicians to do. We do have training videotapes which demonstrate how we think they can engage a smoker coming in for whatever reason into considering quit- ting smoking and then go on to discuss with them their reasons for quit- ting smoking. We have a patient questionnaire which is very helpful and which provides 3 or 4 key items for the physician to discuss with the pa- tient at the first visit. So in a sense the physician feeds back to the patient some of the materials and some of the data that the patient has provid- ed, such as reasons for wanting to quit, support systems at home, con- gratulating the patient on the decision, and so on. It's also interesting that 80% of our 2,000 patients indicated that they would like help with quit- ting smoking and that was very encouraging. DR. RUSSELL: Thank you very much, Saul. I'd like to thank all the audience. It seems that we've gotten away from considering whether the gum works, but we're considering more how to get it to work better in a variety of settings, and, secondly, what is the most cost-effective level of intensity of adjunctive support. I 262

Overview: Economics of Treatment Judith K. Ockene, Ph.D. University of Massachusetts Medical School Worcester, Massachusetts Introduction The economics of smoking treatment is an important consideration both because of the continued high prevalence of smoking and the need for some smokers to receive costly help to stop smoking. Although preventive health services should not necessarily be required to save money, common sense would dictate that prevention yields greater benefits than does attempting to cure disease after onset (5). While it is difficult to assign a monetary value to human lives, there have been a num- ber of attempts to assess the economic costs and benefits of a reduction in smoking prevalence (1,2,6,7,9-11,14). Current best estimates place the annual medical costs related to cigarette smoking at $39 billion to $55 billion in the United States and $400 to $800 in excess costs to employ- ers per smoking employee (1983 dollars) (13). Cost Issues Included in calculations of inedical-care costs ofcigarette smoking are the direct costs of medical care for the diagnosis and treatment of illness and disease and the rehabilitation of patients. Indirect costs result from lost output by reduction or cessation of productivity due to death or dis- ability. There are other non-economic cost issues which must also be con- sidered when using a broad economic perspective, including physical pain, fear and disability, and general quality of life. These costs are important, but it is difficult to place an economic value on them, and they are often not taken into account in cost estimates of smoking (13). By ignoring these important cost intangibles the figures noted above greatly underes- titnate the truee social cost of smoking (16). When the wider non-economic social, health and human benefits are considered the benefit-cost ratio of a reduction in smoking prevalence is so skewed towards benefit that the cost of intervention seems minor (3). The economics of treatment can also be considered from the individual smoker's perspective. Smokers can accrue a tremendous savings over a lifetime after they stop smoking. These savings include the costs of cigarettes, such unpredictable costs as holes burned in clothing and fur- niture and fires set by cigarettes, and out-of-pocket medical care costs. Using these kinds of calculations, it is clear that cigarette smoking treat- ment resulting in cessation is very cost effective from the patient's per- spective. Smoking cessation will likely lead to decreased medical costs only if medical care for the elderly is less expensive than medical care for those more youthful. A death from myocardial infarction at age 40 is often very cost-effective from a medical care perspective - the individual usually dies quickly (often out of the hospital), and society is spared the burden of caring for the elderly person with a multiplicity of expensive, recurrent hospitalizations. If we could by some miracle convince all smokers to per- manently quit tomorrow, there would be a sharp decline in medical costs which would last for a period of some years. But in fact we would have simply been postponing costs, not eliminating them. As this large popu- lation, spared the early smoking-induced deaths, grows older, the inevita- ble degenerative illnesses of aging will lead to large costs associated with their care. Smoking cessation is therefore cost-effective in terms of diminished long-term medical costs only in a system where there are age cut-offs for certain types of therapy, as is seen in certain European coun- tries. In such a system if a given individual's coronary heart disease is moved from age 60 to age 72, medical costs are diminished, since there is a cutoff of age 70 beyond which intensive care units are not utilized and coronary artery bypass surgery is not performed. Instituting such a cut-off system, however, is not a medical decision; rather it is an issue for sociologists, politicians, philosophers, and ethicists, requiring many of us to radically restructure our philosophy of medical care. Denying medical care to someone who might benefit from it contradicts our defi- nition of providers' responsibilities in the United States, although increas- ingly such discussions of resource allocation take place in the context of diminishing medical resources and increasing medical care costs. But even without a change in our attitudes toward health care for the aged, businesses stand to benefit economically from smoking cessation. This would result from the postponement of medical costs that are borne by employers into the age range beyond retirement in which medical costs are borne by the Medicare system. Thus, while overall medical expenses may not be favorably altered, the business community may see smok- ing cessation as a cost-effective approach to reducing its health care costs. Just as confounding factors must be considered when assessing cost- effectiveness from a medical care perspective, there are also confound- ing factors to consider when estimating indirect costs and benefits of 264 1 265

smoking cessation treatment. For example, smokers who die early place less of a burden on the social security system, and the gap left in their work environment creates a job for someone else. To put all of this in perspective, surely life and health have greater benefits than death and disability. To evaluate cost and benefits only in terms of dollars and cents is shortsighted and devalues the meaning of life and health. Cost-benefit analysis must encompass issues spanning many fields-not only medicine, hut also business economics, the so- ciology of aging and the value society places on the aged, philosophical concerns about allocation of resources, the opportunities of individuals, and the ethics of restricting access to medical care. If the total benefit of improving the quality of life was made a priority by nations such as the United States and those in western Europe, economic resources would be allocated to smoking treatment and "econ(,i+>ic concerns" would not be an issue (4). Policy Implications Several policy implications flow from the economics of smoking treat- ment. At present therc is a low rate of reimbursement from insurors and/or government for smoking intervention. While there may be indirect ways to help deal with this problem, lack of economic support often deters physicians, psychologists and other health care providers from treating smokers. If insurors and government were convinced that treatment is cost-effective, nicotine dependence would likely be subject to the same reimbursement policies as other medical problems in terms of treatment and prescription costs. Reimbursement policy for smoking treatment has major implications for health care providers and smokers. A greater likeli- hood that there would be payment to health care providers such as psy- chologists and physicians would probably lead to an increase in the nurn- ber of treated individuals, resulting in a greater prevalence of cessation. Because of the many factors affecting the economics of smoking it is difficult to limit considerations of treatment reimbursement to narrow cost-effectiveness. It may therefore be more realistic to consider rcim- bursement policies in the context of the many benefits and costs of smok- ing and smoking treatment in our society (15). Reimbursement policy also has implications for other health promo- tion and disease prevention services beyond smoking treatment. If in- surors and government determine that smoking-cessation services are cost-effective then they may be willing to reimburse for other disease prevention services. A more vigorous approach to disease preventior) could enhance people's general health and increase their satisfaction with life. Ultimately, governmental or insuror support for smoking treatment will impact health and disease among our citizens. Although the overall impact of intervention may be modest, given the size of the smoking population, behavior changes become substantial in terms of absolute magnitude. If one of even every four smokers were to become a quitter as a result of treatment, the premature deaths of almost half a million Americans could be averted (17). It is difficult to put a price on these lives saved. On the other hand, it is also important to be aware that medical practices and technology may improve in the future so that many now- fatal heart and lung diseases will become curable. If this were to occur, the potential savings of lives might actually decrease (16). Morbidity im- plications are less easy to discern, as avoidance of death could potential- ly lead to increased illness and disability. While effective smoking control has short-term positive economic out- comes, it could ultimately have a negative impact if no policy changes arc made in the mechanism for financial support of the retired popula- tion with a longer life expectancy (11). Policy will need to be adapted for the older, healthier individuals who are able to continue to lead produc- tive lives. It may require the younger, working population to pay more in taxes to support the elderly unemployed, but the former will subse- quently reap the same benefits (12). Assessing the Economics of Treatment The two papers in this section help shed light on the economics of treat- ment. To assess the impact of physician intervention for smoking cessa- tion using nicotine-containing gum as an adjunct, Dr. Gerry Oster and his colleagues at Policy Analysis, Inc., have calculated the economic costs of smoking over the lifetime of an individual smoker and compared these to costs saved when cessation occurs in a percentage of smokers receiv- ing intervention. Using this prospective, analytical, incidence-based ap- proach, Dr. Oster estimates the expected lifetime costs of smoking in present valued dollars. These estimates include direct health care costs and lost income costs from death or long-tcrm disability. Dr. Oster dcmonstrates that in terms of its costs and clinical benefits, treatment with nicotine gum is a cost effective strategy when compared with other widely-accepted medical practices. Dr. David Sachs presents his estimates of niedical care cost savings based on the potential costs of recurrent tnyocardial infarction among smokers in a coronary care unit and the 266 1 267

savings realized if a certain proportion of these smokers quit smoking. Dr. Sachs compares these savings to the costs of various levels of inter- vention, the most intensive of which involves the prescription of nicotine- containing chewing gum. With his calculations Dr. Sachs demonstrates that intensive intervention among smokers with coronary disease is in- deed cost-effective and in the long run would provide a substantial sav- ings to third party payors. Conclusion The two papers in this section illustrate some of the considerations that are included in a determination of the economic costs and benefits of treat- ment for smoking cessation. They lend support to the idea that inten- sive interventions for smokers are economically sound. We are already aware that these interventions are morally and ethically sound, and at least as cost-effective as treating such diseases as lung cancer, emphysema and heart disease. References 1. Atkinson A, Townsend J. Economic aspects of reduced smoking. Lan- cet 1977; 2:492. 2. Cady B. Cost of smoking. N. Engl. J. Med. 1983; 308(18):1105. 3. Catford JC, Nutbeam D, Woolaway MC. Effectiveness and cost- benefits of smoking education. Community Med. 1984; 6:264-272. 4. Fein R. Fiscal and economic issues. Bull. NY Acad. Med. 1975; 51:235-241. 5. Fielding JE. Preventive medicine and the bottom line. J. Occup. Med. 1979; 21(2):79-88. 6. Kristcin MM. Econotnic 6:252-264. issues in prevention. Prev. Med. 1977; 7. Laing ET, Taylor DG. The economics of smoking. Health Educa- tion Council, London, 1983; (unpublished). 0'1 8. Leu RE, Schaub T Does smoking increase medical care expenditure? ~-~ Soc. Sci. Med. 1983; 17(23):1907. 9. Luce BR, Schweitzer O. Smoking and alcohol abuse: A comparison ~a of the economic consequences. N. Engl. J. Med. 1978; 298:569-571. 10. Ramstrom LM. Economic losses to the society due to smoking. P. 112 in LM Ramstrom (Ed.) Proceedings of the Fourth World Confer- ence on Smoking and Health, Stockholm, June 1979, Almquist and Wik- scll International. 11. Richter 13J, Gori GB. Demographic and economic effects of the prevention of early mortality associated with tobacco and related dis- eases. In GB Gori, FG Brock (Eds.) Banbury Report: A Safe Cigarette? Cold Spring Harbour Laboratory, 1980. 12. Schelling TC. Economics and cigarettes. Preu Med. 1986; 15:549-560. 13. Schultz JM. Perspectives on the economic magnitude of cigarette smoking. NY State J. Med. 1985; 302-306. 14. Thompson ME, Forbes WF. Costs and benefits of cigarette smok- ing in Canada. Can. Med. Assn. J. 1982; 127:831. 15. Walsh DC. Corporate smoking policies: A review and an analysis. J. Occup. Med. 1986; 26:17-22. 16. Warner KE. The econotnics of smoking: dollars and sense. NY State J. Med. 1983; 1273-1274. 17. Warner KE. Smoking and health implications of a change in the fed- eral cigarette excise tax. J. Am. Med. Assn. 1985; 255(8):1028-1032. 18. WHO Expert Committee. Smoking Control Strategies in Developing Countries. Technical Report Series 695. Geneva: World Health Or- ganization. 268 269

Cost-Benefit Analysis of Tobacco Dependency Treatment David P. L. Sachs, M.D. Snioking Cessation Research Institute Palo Alto, California and Stanford University School of Medicine Stanford, California Introduction Health care, or more appropriately illness care, costs a substantial amount of money every year. The present spending level in the United States is about $1 billion per day-$365 billion per year (6,9). Smoking induced illnesses, including lung cancer, chronic obstructive lung disease, and coronary heart disease, account for nearly 18% of health care costs in the United States (9). Although 33% of the American popu- lation smoked in 1983, they accounted for a grossly disproportionate share of health care costs: 59% (2). Therefore, in the United States or any coun- try where people smoke, a very effective tool for long-term health care cost containment is smoking cessation. By balancing the costs saved by the medical illnesses avoided through smoking cessation against the cost of that treatment, we can arrive at a per person dollar treatment cost that is economically and scientifically justified. This approach, of course, looks strictly at medical costs saved vs. treatment costs generated and does not consider any other issues, such as quality of life enhancement. The Cost of Treating a Heart Attack The biggest contributor to the rise in health care costs in the United States has been the proliferation of Medical Intensive Care Units (MICU) and Coronary Care Units (CCU) (6). Coronary Care Units, in contrast to Medical Intensive Care Units, have a relatively high survival rate-about 83%, (6). This high survivorship does not mean merely surviving the hospital stay to be discharged; it refers as well to a high quality of life for at least one year following the acute myocardial infarction. Because this kind of treatment has a cost, it is ap- propriate to compare it against the cost of preventing the acute heart attack in the first place. I shall present a cost analysis model of treating acute myocardial in- farction that is based on health care costs in the United States as of 1984. To generate this, I am drawing on six and a half years as an attending phy- sician and two and a half years as Associate Director of the Medical In- tensive Care Unit at University Hospitals of Cleveland. (All monetary data are expressed in 1984 dollar terms.) The model that follows is a conservative one because it only considers the costs generated during the acute CCU stay, and does not include the costs generated when the patient is transferred out of the coronary care unit to a "step down" bed and then to the general medical floor. Even af- ter a simple heart attack, with only a five-day CCU stay, the patient will typically be in the hospital in a general medical setting for an additional five to ten days. The cost accounting model is conservative also because it does not in- clude the more extensive rehabilitation costs in extended care facilities for those patients who had complicated myocardial infarctions. If these legitimate treatment costs were included in the model, then the total treat- ment costs would be higher and any potential savings generated by preventive treatment would be correspondingly greater. This model is conservative in yet another way: other categories of critically ill patients, such as respiratory failure patients, are not considered at all. Stich patients have much higher mortality rates and generate substantially greater hospi- tal costs because of the increased length of MICU stay and the extent and type of high technology services provided (1). Finally, the model is conservative because of rising health care costs since 1984. For example, comparable charges in 1986 at Stanford University Hospital are nearly twice those that I have used in my model. Nonetheless, the fundamental conclusion from this model is that in the United States, just in terms of decreasing myocardial infarction incidence and decreasing CCU utili- zation, we can justify a surprisingly high per patient cost for smoking cessation services. Simple vs. Complicated Heart Attack. For simplicity we shall con- sider only two kinds of myocardial infarctions: simple and complicated. The simple myocardial infarction patient has an average CCU stay of five days. A simple infarction indicates that while such patients have not had either significant arrhythmias or congestive heart failure in the immedi- ate post-infarction period while they were in the CCU. Therefore, their medication and high technology monitoring needs in the CCU are minimal. In contrast, patients with a complicated myocardial infarction spend ap- proximately 14 days in the CCU. A diagnosis of complicated myocardial 271 270 j

Figure 1. HEART ATTACK COSTS average length of stay in days CORONARY CARE UNIT (CCU) 0 5 Simple Complicated SIMPLE COMPLICATED medical cha(geslday 0 TOrAL CHARGES $1000 10 F 15 $2000 $3000 0 $20000 $40000 This illustration presents a relatively conservative model showing the health care costs generated in the United States, as of 1984, for a simple vs. complicated myocardial infarction. Only length of stay in the CCU is considered; step-down and regular medical bed usage, which always follows, is not added in. infarction includes either significant arrhythniias or congestive heart failure during the immediate post-infarction period. Because of this, invasive monitoring is frequently required, including insertion of arterial lines, Swan-Ganz catheters, or transvenous pacemakers. Each must be connect- ed to analog or digital electronic bedside monitoring equipment, which is usually connected to a central computer analysis system. Additional- ly, various respiratory therapy modalities beyond oxygen therapy are fre- quently required. Occasionally, patients such as these will need to be placed on an artificial respirator (mechanical ventilator). These additional monitoring, diagnostic, and therapeutic services not only have capital equipment costs, but more importantly, their use requires extensive at- tention frotn highly trained personnel, such as nurses and respiratory ther- apists. Typically, a patient with a complicated myocardial infarction must be cared for by a registered nurse with a caseload of up to three patients. In contrast, a patient on a general medical floor may be one of 20-25 pa- tients in the RN.'s caseload. Additionally, such units not uncommonly have a full-time respiratory therapist. All these factors raise costs of care for a simple myocardial infarction. Moreover, complicated myocardial infarction patients also generally require more sophisticated, and hence more expensive, laboratory testing, medications, and other ancillary services. Heart Attack Costs-Simple. Costs are sumtnarized in Figure 1, com- paring simple vs. complicated heart attacks. The medical costs per day are expressed in terms of three basic categories: 1) hospital bed charges, 2) hospital ancillary services charges, and 3) physician charges. Note that the hospital's basic bed charge remains identical for the two types of heart attacks-$800 per day. This includes nursing care services and the so- called hotel functions: basic room, board, housekeeping, maintenance, etc. Hospital ancillary service charges include such items as respiratory therapy, pharmacy services, laboratory services, occupational therapy, physical therapy, and central supply services. As Figure 1 shows, for a simple myocardial infarction this will total an additional $1,000 per day. Thus, total hospital costs will be just under $2,000 per day for an un- complicated heart attack. The physiciaris professional fees would be some- where in the range of $100-150 per day in the United States, bringing the total daily costs to about $2,000 per day. Thus, for the five-day CCU stay total medical care costs would be $9,750. Heart Attack Costs -Complicated. The cost for a complicated my- ocardial infarction is substantially higher. What creates the greatest incre- ment in total health caree costs is the rise in hospital ancillary charges from 272 1 273

$1,000 per day to about $1,500 per day and a stay in the CCU for nearly three times as long. Some of the increased ancillary charges are reflected in the cost of the following disposable items: 1 Swan-Ganz catheter at $85-175 per unit and I transvenous pacemaker at $175-250 per unit. I emphasize that these are simply the costs to the hospital of buying these dis- posable devices. Physician's charges also go up because more time is spent at the bedside and performing more invasive procedures, such as plac- ing an arterial line, placing a Swan-Ganz catheter, etc. Thus, daily phy- sician charges will average about $250 per day. Consequently, the total daily cost for a complicated nlyocardial infarction is about $2,500 per day. Since the patient is in the CCU for 14 days, the total health care costs, jt+st for the CCU phase of the hospital stay, would be $35,700. Although these figures might seem high, I have deliberately been con- servative at each step of the way, in terms of the cost figures used. If an economist performed a cross-sectional analysis of hospitals around the Nonetheless, this gives us a conservative but reasonable background against which to assess the cost effectiveness of smoking cessation treat- ment interventions that would reduce the myocardial reinfarction rate. Cost-Benefit Determination "A medical practice can indeed be said to be cost effective in an abso- lutc sense if it both saves money and provides a health benefit" (4,5). I will show that treating tobacco dependency following inyocardial infarc- tion clearly saves money and also provides a health benefit. Not only is money saved because of the reduced reinfarction incidence and a conse- quent reduction in CCU costs, but also, as the analysis below will show, even after subtracting the costs of smoking cessation treatment for all heart attack patients, there is still a net savings. Clearly, there is substantial benefit in reducing heart attack reoccurrence rate. Thus, post-myocardial infarction smoking cessation treatment not only meets the most appropri- ate medical use of the term cost effective (cost effective = having an ad- ditional benefit worth the additional costs) (4), but it also meets a more stringent business-oriented definition of cost effective: a given practice is regarded as cost effective if and only if it is cost saving (4). This more business oriented definition of cost effectiveness would categorically eliminate many standard medical treatment programs. For example, treat- ment of hypertension clearly does provide a health benefit, but it cer- tainly does not save any money (4); in fact, it requires the net expendi- ture of money. Nonetheless, it is doubtful that the most stringently cost cutting health care administrator, whether in the private or governmen- tal sectors, would recommend abolishing hypertension treatment. Nonetheless, in a contracting economy, it is certainly understandable that third-party carriers adopt this definition and either explicitly or implicitly require that any new treatment result in a net cost saving. Because of the serious, chronic, and often catastrophic medical disorders which cigarette smoking causes, smoking cessation treatment is cost effective in both the more appropriate medical sense, and also in the highly stringent busi- ness sense. That it also provides health benefits besides reduced heart attack reoccurrence rate is an added bonus. Cost-Benefit Analysis of Smoking Cessation 'IYeatment. Assume 100 cigarette smokers continue to smoke after a simple, uncomplicated myocardial infarction. According to such prospective studies as the Gote- berg Trial (10) and the Dublin Trial (8), for the sake of this example, it is reasonable to conclude that 30% or 30 of those 100 patients who have had a heart attack and continue to smoke are going to experience a rein- farction. Thirty reinfarctions x $9,750 (total, acute, CCU health care costs/infarction) = $292,500. If all 100 smokers are provided with a smoking cessation treatment that has only a 15% sustained efficacy rate, then 85% of the patients will still be smoking. Assuming the same 30% reinfarction rate for them, 25.5 (30% x 85 patients) of those individuals will sustain a reinfarction. From the same trials mentioned above, smoking cessation has beeh shown to decrease the rate of reinfarction by approximately 50% (8,10). Therefore, the likelihood of a second infarction, in those who stopped smoking, will drop from 30% to 15%. Thus, for the 15 patients who stop smoking and remain nonsmokers with the postulated intervention, only 2.25 will have a repeat heart attack (15% x 15 patients). Thus the cost of the rein- farctions will be (25.5 x $9,750 = $248,625) + (2.25 x $9,750 = $21,938) = $270,563, or $21,9371ess than if a smoking cessation intervention with 15% sustained efficacy had not been implemented. Looked at another way, by employing an intervention that has a 15%, success rate with those patients who continue smoking following a my- ocardial infarction, 2.25 repeat heart attacks will be avoided at $9,750, for a net savings of $21,937. Therefore, this $21,937 represents the total that can legitimately be spent to treat all 100 patients for tobacco depen- dency following their myocardial inf<arctions (M.I.), and still not increase health care costs. In the first example, where no smoking cessation treat- ment was provided, the medical cost of reinfarction was $292,500 and the cost of smoking cessation treatment was $0.00. In the second case, where treatment intervention with a 15% success rate was provided, the medical cost of myocardial infarction treatment was $270,563. If the entire savings was spent in providing smoking cessation treatment immediately 274 275

following the first infarction, then the cost of smoking cessation services was $21,937, for a total (M.I. + Smoking Cessation Treatment) cost of $292,500. Thus total health care costs for myocardial infarction treatment + smoking cessation treatment remained constant at $292,500. Table 1 A Tale of 100 Smokers Who Still Smoke After Their First Heart Attack- Simple M.I. No Post-CCU Smoking With Post-CCU Smoking Cessation Treatment: Cessation Treatment 15% Cure Rate* 30% Cure Rate 50% Cure Rate # of Smokers at Risk for Rcinfarction 100 85 70 50 Reinfarction Probability Rate 30% 30% 30% 30% Number of Actual Mls 30 25.5 21 15 # of Ex-Smokers at Risk for Rcinfarction 0 15 30 50 Reinfarction Rate - 15% 15% 15% Number of Actual Mls 0 2.25 4.5 7.5 loual Reinfarctions 30 27.75 25.5 22.5 # of Reinfarctions Prevented 0 2.25 4.5 7.5 Medical Cost/Rcinfarction $9,750 $9,750 $9,750 $9,750 Ibtal Cost of Rcinfarctions $292,500 $270,563 $248,625 $219,375 (30 x $9,750) (27.75 (25.5 (22.5 x $9,750) x $9,750) x $9,750) CCU Costs Saved by Smoking Treatment $0.00 $21,937 $43,875 $73,125 ($292,500 ($292,500 ($292,500 - $270,563) - $248,625) - $219,375) *Cure rate nreans sustained, one-year abstinence and reflects continuous abstinence from the end of thc treatment intcrvcntiun through the one-year follow-ttp point. Table 2 A Tale of 100 Smokers Who Still Smoke After Their First Heart Attack-Complex M.I. No Post-CCU Smoking With Post-CCU Smoking Cessation'Ikeatment: Cessation Treatment 15% Cure Rate* 30% Cure Rate 50% Cure Rate # of Smokers at Risk for Reinfarction 100 85 70 50 Reinfarction Probability Rate 30% 30 % 30% 30 % Number of Actual Mls 30 25.5 21 15 # of Ex-Smokers at Risk for Reinfarction 0 15 30 50 Reinfarction Rate - 15 % 15 % 15 % Number of Actual Mls 0 2.25 4.5 7.5 Total Reinfarctions 30 27.75 25.5 22.5 # of Reinfarctions l'revented 0 2.25 4.5 7.5 Medical Cost/Rcinfarction $35,700 $35,700 $35,700 $35,700 Total Cost of Rcinfarctions $1.071 million $990,675 $910,350 $803,250 (30 x $35,700) (27.75 (25.5 - (22.5 x $35,700) x $35,700) x $35,700) CCU Costs Saved by Smoking Treatment $0.00 $80,325 $160,650 $267,750 ($1,071,000 ($1,071,000 ($1,071,000 - $990,675) - $910,350) - $803,250) *Cure rate means sustained, one-year abstinence and reflects continuous abstinence from the end of the treatment intervention through the one-year follow-up point. Thus, we can legitimately spend $219.37 per patient for each of the 100 smoking patients following infarction and still not increase aggregate health care costs in this sphere. These data are presented in tabular form in Table 1(first two columns). If this model is extended to complicated myocardial infarctions, then the potential savings, and hence the maximum, per patient, cost effec- tive smoking cessation treatment increases approximately fourfold (see Table 2, first two columns). The net savings in this case, $80,325, would justify a $803.25 smoking cessation treatment cost for each of the 100 original patients. The last two columns in Tables 1 and 2 show similar analyses for sim- ple vs. complex reinfarctions assuming a 30% sustained abstinence (cure) rate or 50% sustained abstinence. (Note that Hall RG, et al., reported 50% 276 277

sustained, two-year abstinence for a group of cardiopulmonary patients, many of whom fell into this very post-infarction group (7). Patients in that study were treated with multicomponent rapid smoking.) As column four in Table 2 shows, a smoking cessation treatment which had 50% efficacy and cost less than $2,678 would be fully cost effective, in the most stringent sense (4), and would not increase actual health care costs. Ob- viously, if such a treatment does cost less than that, then savings would be "profit" for society (and the health insurance companies). The models shown in Tables 1 and 2 are conservative in another way. The model assumes that, in the case of complex myocardial infarctions, ex-smokers have an equal pi7obability of sustaining a complex reinfarc- tion as a continuing smoker. This, in fact, is not the case. Continuing smokers are more likely to have complex reinfarctions (and also die) (8,10). For example, if we modify the model in Table 2, column four, to as- sume that the proportion continuing to smoke have complex reinfarc- tions, while those who have stopped smoking only have simple reinfarc- tions, then the money saved by providing the smoking cessation treatment amounts to $462,375. In this context, a smoking cessation treatment pro- gram with 50% efficacy costing $4,624 for each of the 100 post-infarction patients would be fully cost effective (4). Such a treatment, though, can be provided at substantially less cost. For example, the cost of the multi-component, rapid smoking treatment program used in the Hall, et al. article (7) was about $1,400/patient. This included a complete medical history and physical examination, 16 treat- ment visits, and four follow-up visits during the first year. Thus, this latter, somewhat more realistic model, would generate a net savings to the health care system of $3,224 ($4,624-$1,400/patient), or a total savings "to so- ciety" in these 100 patients of $322,400. In the more conservative model, that would still produce a net "profit" to the health care system (and health insurance) of $1,278/patient ($2,678-$1,400) or $127,800 for the entire 100 patients in this example. At the other end of the spectrum, a less intensive treatment interven- tion which did not cost so much could still be cost effective, even though it did not have as high an efficacy rate. For example, in a recent, minimal intervention medical trial using nicotine polacrilex, 2 tng, the sustained, one-year abstinence rate was 31% (3). This intervention, including the cost of the nicotine gum would have been $365/patient. Although the savings to the health care system would be far more modest, there would still be a net "profit" of $74/patient, or $7,400 for the entire group of 100. 1 Conclusion Such a model shows us that by treating post-myocardial infarction pa- tients with the smoking cessation "technology" presently available, the American health insurance industry could realize substantial savings. It is paradoxical that health insurance companies in the United States will reimburse treatment such as chemotherapy for various cancers with cure rates that are not as good as the cure rates presented either in this chap- ter or elsewhere in this volume for tobacco dependency treatment. Moreover, health insurers will also reimburse for the secondary compli- cations of chemotherapy, such as gram-negative septic shock, resulting from loss of white blood cells, even when these complications can produce Medical Intensive Care Unit stays in the 1-2 month range, at $9,000 per day or more. In contrast, insurers are reluctant to reimburse existing, scientifically validated, smoking cessation treatment interventions which produce long- term cure rates in the 15-50% range. The preceding analysis underscores the need to reevaluate such policies. Clearly it is in the financial interest of society and third party payors to reimburse for scientifically validat- ed smoking cessation treatment provided by a licensed physician or psy- chologist. References 1. Bone RC. Prospective payment and critically ill patients. Lecture de- livered as part of the Symposium, Federal Policy and Pulmonary Medicine: Patient Care, Teaching, and Research Issues. American Col- lege of Chest Physicians, 52nd Annual Scientific Assembly, San Fran- cisco, CA. 09/23/86. 2. Cady B. Cost of smoking. N. Engl. J. Med. 1983; 308:1105. 3. Daughton DM, Kass I, Fix AJ, Ahrens K, Rennard SI. Smoking in- tervention: Combination therapy using nicotine chewing gum and the American Lung Association's "Freedom from Smoking" manu- als. Prevent. Med. 1986; 15:432-435. 4. Doubilet PM, Weinstein MC, McNeil BJ. Occasional notes: Use and misuse of the term "cost effective" in medicine. N Engl. J. Med 1986; 314:253-256. 5. Doubilet PM, Weinstein MC, McNeil BJ. Use of the term "cost- effective." N. Engl. J. Med. 1986; 314:1645-1646. 6. Goldberg RJ, Gore JM, Alpert JS, Dalen JE. Recent changes in at- tack and survival rates of acute myocardial infarction (1975 through 1981). J. Am. Med. Assn. 1986; 225:2774-2779. 278 279

7. Hall RG, Sachs DPL, Hall SM, Benowitz NL. Two-year efficacy and safety of rapid smoking therapy in patients with cardiac and pulmo- nary disease. J. Consult. Clin. Psychol. 1984; 52:574-581. 8. Mulcahy R, Hickey N, Graham IM, MacAirt J. Factors affecting the five-year survival rate of men following acute coronary heart disease. Am. Heart J. 1977; 93:556-559. 9. National Center for Health Statistics. Health, United States, 1983. (DHHS Publication # PHS 84-1232), Public Health Service. Wash- ington, D.C. U.S. Government Printing Office, 1983. 10. Wilhelmsson C, Vedin JA, Elmfeldt D, Tibblin G, Wilhelmsen L. Smoking and myocardial infarction. Lancet 1975; 1:415k-419. The Cost-effectiveness of Nicotine Chewing Gum as an Adjunct to Physiciari s Advice Against Cigarette Smoking in a Primary Care Setting Gerry Oster, Ph.D Daniel M. Huse, M.A. Thomas E. Delea, B.A. Policy Aiiai}%sis, inc. Brookline, Massachusetts Graham A. Colditz, M.B.B.S., M.P.H. Channing Laboratory Department of Medicine Harvard Medical School and Brigham and Women's Hospital and Department of Epidemioloyy, Harvard School of Public Health Boston, Massachusetts Introduction Since the publication in 1964 of the first Surgeon General's Report on Smoking and Health (21), scientific evidence of the adverse effects of smoking has continued to mount. Today, cigarette smoking is generally regarded as the single most important preventable cause of death and dis- ability in the U.S. (22). Significant attention has therefore been directed toward methods that can assist sniokers in giving up smoking. One of the newest such methods involves the use of nicotine chewing gum, which is designed to produce serum nicotine concentrations similar to those that occur during cigarette smoking (11). While the efficacy of nicotine gum has been demonstrated in clinic-based smoking cessation programs (6,7,10,15,16,28), its value among patients seen in a primary care setting recently has been questioned, largely because of a belief that the cost of nicotine gum is high relative to its effectiveness in this set- ting (1,2,11,13,17). In this study, we examine the cost-effectiveness of The authors would like to thank Clarice Brown of the Office on Smoking and Health, U.S. Department of Health and Human Services, for her assistance in providing us with data. This study was supported by a researchgrant from Merrell Dow Pharmaceuticals, Inc., and partially by grants CA 37088 and HL 24074 (f the National Institutes of Health. 280 281

nicotine gum as an adjunct to physician's advice against cigarette smok- ing during routine office visits. We estimate the effectiveness of nicotine gurn in a primary care set- ting on the basis of a meta-analysis of data from published reports of clin- ical trials. We compare the costs of nicotine gum therapy against its benefits, measured in terms of years of additional life expectancy for those smokers who would not have quit smoking (or would have done so only later in life) had they not used the gum. We calculate cost-effectiveness in terms of the amount that must be spent to save an additional year of life. Methods Calculation of Cost-Effectiveness. The cost-effectiveness of nico- tine chewing gum is examined among a hypothetical group of 250 pa- ticnts who are smokers and are seen during routine office visits. We as- sumed that a prescription for the gum (but not the gum itself) would be offered to all 250 patients, but that only a fraction of these patients would actually use the gum in an effort to quit smoking. The cost of this intervention is based on the amount of physician's time that would be spent prescribing nicotine gum to 250 patients, as well as the amount of gum that would be purchased by those patients who ac- tually use it, whether or not they succeed in giving up smoking. We mea- sure effectiveness in terms of the difference between the number of pa- ticnts who would give up smoking if given a prescription for the gum along with advice against smoking, and the number who would quit on the basis of physician's advice alone. The benefits of intervention are ex- pressed in terms of the added years of life expectancy for these addition- al patients who stop smoking. In measuring the gain in life-years due to intervention, we take account of the fact that some of the patients who stop smoking while using nicotine gutn would have done so on their own at a future date. Cost-effectiveness is expressed in terms of the cost (in 1984 dollars) per year of life saved among all patients who are offered a prescription for the gum, not just those patients who purchase it or those who succeed in giving up smoking. Compliance with Physician's Advice to Use Nicotine Gum. In clin- ical trials, approximately 50%, of smokers seen in primary care practice have accepted an offer of nicotine gurrr when it has been provided free of charge (13,26). This number is likely to be much lower when patients must purchase the gum with their own frtnds; we know of no data, however, on rates of patient compliance with physician's advice to usc nicotine gum under these circumstances. Therefore, we assruned that only 25% of patients who are offered a prescription for the gum will actually use it. Wc examined the sensitivity of our results to changes in this as- suniption. Effectiveness of Physician's Advice with and without Nicotine Gum. 1o determine a rate of smoking cessation for patients who receive only physician's advice against smoking, we identified all English- language reports of trials of this intervention using a computerized liter- ature search (MEDLINE), which we supplemented with a review of ci- tations from the retrieved articles. We then used only those trials that reported a 12-month rate of smoking cessation, and excluded any trial that enrolled only patients with smoking-related diseases. IZcportcd rates of smoking cessation that were based on patient self-reports of quitting were adjusted downward to account for an expected rate of deception. We estimated this deception rate on the basis of trials that reported both biochernically-validated and self-reported cessation rates (14,26,27), us- ing the rnean ratio of the former to the latter. Rates of smoking cessation one year after intervention were pooled using techniques of meta-analysis (3>H) • Because of limited data, we had to use an indirect method to estimate a rate of smoking cessation for patients who use nicotine gum.in a primary care setting. We first identified all published reports of placebo-controlled trials of rucotine gum, and then used only those trials that utilized ran- domized assignment to treatment groups and also reported a 12-month rate of smoking cessation. The effectiveness of the gum in each trial was thcn expressed in terms of the ratio of the cessation rate for patients ran- domized to active gum divided by the corresponding rate for patients randomized to placebo gum. We calculated a weighted average of these ratios by assigning each a weight proportional to its precision (i.e., to the inverse of the variance of the ratio) (25). Finally, to estimate a one-year cessation rate for patients using nicotine gum in a primary care setting, we rnultiplied the advice-only cessation rate by this weighted average ratio. Smoking Relapse. Among patients who are abstinent at 12 months, smoking relapse is relatively uncommon (5,12) although at least one study has found it to occur at a rate as high as 5%, per year after year one (18). In our analysis, we assumed that a patient who has not smoked for a period of one year will not begin to smoke again later in life. We examined the sensitivity of our results, however, to changes in this assumption. Future Changes in Smoking Behavior in the Absence of Interven- tion. In mcasuring the benefits ofintcrvention, it must be recognized that a patient who quits smoking while using nicotine gum may not have 282 283

smoked a lifetime in the absence of intervention. Data from US Health Interview Surveys suggest that, in any given age cohort, the prevalence of smoking declines by about 1% annually (unpublished data, Office on Smoking and Health, United States Department of Health and Human Services). Therefore, in estimating cost-effectiveness, we assumed that 1% of remaining smokers would quit smoking during each year after intervention. We examined the sensitivity of our findings to changes in this rate. Increase in Life Expectancy due to Smoking Cessation. We cal- culated annual mortality rates for cigarette smokers and nonsmokers be- tween the ages of 35 and 105 years using unpublished 1985 life table data based on the American Cancer Society's (ACS) 25-state Cancer Preven- tion Study. For patients of every age between 35 and 69 years who stop smoking, annual mortality rates were estimated for the years between the age at cessation and age 105 years, using Doll and Peto's (4) estimates of the relative decline in excess mortality in the years after quitting in con- junction with smokers' and nonsmokers' mortality rates. We then calcu- lated life expectancies for smokers of every age between 35 and 69 years, and for patients who stop smoking at each of these ages, using annual probabilities of survival derived from corresponding mortality rates. The difference, at any given age, between the life expectancies of patients who smoke and those who stop smoking represents the increase in life ex- pectancy due to cessation of smoking at that age. Cost ofNicotine Gum Therapy. The average selling price of nicotine chewing gum (Nicorette; 2 ing gum; box of 96 pieces) was determined through a survey of 12 retail pharmacies in the Boston metropolitan area. Based on reported ad libitum gum consumption in clinical trials (10,15), we estimated that patients who use nicotine gum will consume six pieces per day. We assumed that those who use the gum and are able to give up smoking will require a four-month supply, while patients who use the gum but do not subsequently quit smoking will use only a one-month supply. Since nicotine gum must be purchased in whole boxes, we round- cd up our estimates of patients' gum use to the nearest number of whole boxes to determine cost. We also assumed that five minutes of physician's time would be required for every patient who is given a prescription for the gum. We examined the sensitivity of our results to changes in this assumption, as well as to adjustment of the net costs of intervention for changes in the lifetime med- ical care costs of patients who quit smoking. Discounting. Future annual probabilities of survival (used to calculate life expectancies) and future medical care costs were both discounted; the rationale for this has been discussed elsewhere (32). A 5% annual rate of discount was used throughout the analysis. Results Effectiveness of Nicotine Gum as an Adjunct to Physician's Advice. Examination of results reported from six trials (9,14,20,26,27,29) indicates that 4.5% of smokers (95% confidence interval, 4.0%, 5.1%) receiving routine physician's advice and counseling against smoking in a primary care setting will be abstinent at one year. On the basis of seven placebo- controlled trials (2,6,7,10,15,t6,28), we estimate that nicotine gum will increase the one-year rate of cessation by 35.3% (confidence interval, 6.6%, 71.1%). Combining these results, we estimate that 6.1% (i.e, 0.045 x 1.353) of patients seen in primary care practice who use rricotine gum will quit smoking. Assuming a 25% compliance rate, we calculate that only 63 patients out of a total of 250 who are prescribed the gum will actually use it. Of these 63 patients, four (6.1 %) will give up smoking; only three of these patients (4.5 %) would have done so with physician's advice alone. There- fore, among a group of 250 patients, we estimate that one additional pa- tient (12 versus 11) will give up cigarette smoking when a prescription for nicotine gum is given along with physician's advice against smoking. Cost ofNieotine Gum Therapy. The average retail price of nicotine chewing gum is $20.18 per box. The cost of the gum for patients who use 2 boxes per month and quit smoking, assuming four months of gum use, is thus $161.44. For those who use the gum but do not stop smok- ing, the cost of a one month's supply is $40.36. Therefore, the total cost of nicotine gum for the 63 patients who use it is $3,027 (i.e., [4 x$161.44] + [59 x$40.36]). The value of a physician's time in offering a prescrip- tion for the gum to 250 patients must be added to this amount. Assurn- ing a rate of$50 per hour and five minutes per smoker, this adds $1,042 to the cost of nicotine gum therapy. The total cost of intervention is there- fore $4,069. Increase in Life Expectancy Due to Smoking Cessation. Estimates of the increase in life expectancy for men and women who stop smok- ing between the ages of 35 and 69 years are presented in Table 1. Although we discounted all changes in life expectancy when calculating cost- effectiveness, discounted and undiscounted figures are both presented here to illustrate the effect of discounting on the estimated gain in life-years. On an undiscounted basis, the increase in life expectancy is highest for 284 285

youngest patients, and declines markedly with age. Discounting substan- tially reduces the estimated gain in life-years, particularly for younger age groups, and flattens the profile of benefit. , Table 1 Increase in Life Expectancy due to Smoking Cessation, by Sex, Age at Quitting, and Discount Rate. Increase in Life Expectancy (Years) due to Smoking Cessation Undiscounted Discounted Age at Quitting* (0%) (5%) Men 35-39 5.08 0.99 40-44 . 4.60 1.07 45-49 4.00 1.10 50-54 3.32 1.07 55-59 2.60 0.97 60-64 1.90 0.83 65-69 1.32 0.66 Women 35-39 3.18 0.54 40-44 2.94 0.60 45-49 2.64 0.64 50-54 2.28 0.65 55-59 1.85 0.63 60-64 1.40 0.56 65-69 0.97 0.45 *Mcdian age used for purposes of calculation. Cost-Effectiveness of Nicotine Gum as an Adjunct to Physician's Advice. Estimates of the cost-effectiveness of nicotine gum, used as an ~ adjunct to physician's advice against smoking in primary care practice, ~ are presented in Table 2 by age at intervention, for men and women be- tween tween the ages of 35 and 69 years. Costs per year of life saved range from ~ $4,113 to $6,465 for men and from $6,880 to $9,473 for women. Cost- effectiveness is generally highest for patients between 45 and 54 years of age, and is lowest for those in the oldest age group. W O1 ~ 00 Sensitivity Analysis. We examined the sensitivity of our results to changes in a number of our key assumptions and parameters. Reducing the rate of compliance with physician's advice to use nicotine gum to 10%, 286 the expected number of additional patients who stop smoking declines to one in every 625 patients who are prescribed the gum. With a decline in patient compliance, cost-effectiveness ratios increase, ranging from $5,693 to $8,948 for men and from $9,523 to $13,112 for women. Con- versely, a 50% rate of compliance results in one additional patient quit- ting smoking for every 125 patients who are prescribed the gum. Costs per year of life saved decrease in this instance, ranging from $3,586 to $5,637 for men and from $5,999 to $8,260 for women. Table 2 Cost-Effectiveness of Nicotine Chewing Gum as an Adjunct to Physician's Advice Against Cigarette Smoking in Primary Care Practice, by Sex and Age at Intervention. Age at Cost per Life Year Saved Intervention Men Women 35-39 $4,748 $8,996 40-44 4,303 7,846 45-49 4,113 7,187 50-54 4,167 6,880 55-59 4,498 7,073 60-64 5,222 7,842 65-69 6,465 9,473 We also exatnined the sensitivity of our results to changes in the pre- sumed effectiveness of nicotine gum, using the upper and lower confi- dence limits alternatively to calculate the increase in the one-year, advice- only cessation rate that occurs when patients use the gum. Using the lower limit of a 6.6% increase in this rate, cost-effectiveness ratios range from $21,998 to $34,578 for men and from $36,798 to $50,666 for women. With the upper limit of 71.1 %, costs per life-year saved range from $2,042 to $3,210 for men; the corresponding range for women is from $3,416 to $4,703. We next recomputed our results using alternative estimates of the likeli- hood that a patient who quits smoking while using the gum would have continued to smoke in later years. Assuming that no patients would have quit smoking later in life, we find that cost-effectiveness improves slightly. Costs per life-year saved range from $3,706 to $6,159 for men and from $6,218 to $8,974 for women. Conversely, if we assume that 4% of pa- tients will quit smoking in each future year, cost-effectiveness ratios range from $5,292 to $7,415 for men and from $8,910 to $14,076 for women. 287

Abandoning our assumption that patients who have been abstinent for a period of one year will not subsequently relapse, we assumed that 10% of these patients would eventually return to smoking, gaining none of the health benefits of smoking cessation. The benefits of intervention arc therefore reduced, and costs per year of life saved increase, ranging from $4,570 to $7,183 for men and from $7,644 to $10,526 for women. We also altered our estimate of the amount of physician's time that would be required to offer a prescription for the gum to each of 250 pa- tients who smoke. Reducing this to two and a half minutes per patient, the total cost of intervention declines to $3,548, and cost-effectiveness improves slightly; cost-effectiveness ratios range from $3,586 to $5,637 for men and from $5,999 to $8,260 for women. Conversely, increasing physician's time to ten minutes per patient, the total cost of gum therapy increases to $5,111, and cost-effectiveness declines; costs per life-year saved range from $5,166 to $8,121 for men and from $8,642 to $11,899 for women. We next recomputed our results using alternative rates of discount. At a 30% annual rate, the cost-effectiveness of intervention improves slight- ly; costs per life-ycar saved range from $2,516 to $4,995 for men, and from $4,249 to $7,114 for women. At a 7% discount rate, cost- effectiveness ratios range from $6,141 to $8,214 for men and from $10,299 to $16,317 for women. Finally, we recalculated cost-effectiveness incorporating, as part of the net cost of intervention, changes in the lifetime medical care costs of pa- tients who stop smoking. Reductions in the expected lifetime cost of treat- ing smoking-related diseases were based on results reported by Oster et al (19) and were subtracted from the cost of nicotine gum therapy. Con- versely, increases in the lifetime cost of routine medical care, which were calculated by multiplying increases in discounted life expectancy by 1984 national per capita personal health care expenditures not attributable to smoking (24), were added to the cost of intervention. These changes do not substantially affect our results; cost-effectiveness ratios range from $3,871 to $6,854 for men and from $6,938 to $10,100 for women. Discussion We have examined the cost-effectiveness of nicotine chewing gum as an adjunct to physician's advice and counseling against cigarette smok- ing during routine office visits. Our results indicate that the cost per year of life saved with this intervention ranges from $4,113 to $6,465 for men, and from $6,880 to $9,473 for women. Our findings therefore suggest that nicotine gtun therapy compares favorably with other widely-aceepted medical practices such as the treatment of hypertension or hyper- cholesterolemia. For example, the pharmacologic treatment of moder- ate hypertension (diastolic pressure = 110 mtn Hg) in men between the ages of 40 and 50 years has been estimated to require an expenditure of $6,250 (in 1976 dollars, or about $11,300 in 1984 dollars) for every year of life saved (30). Similarly, treatment of men aged 45-50 years with se- rum cholesterol levels greater than 265 mg/dL with cholestyramine res- in, a cholesterol-lowering drug, has been estimated to cost $126,000 (1984 dollars) for every additional year of life gained (31). In comparison, we have estimated that, among male cigarette smokers aged 40-49 years, the cost of nicotine gum therapy is about $4,200 for every year of life saved. We should note that these comparisons may even understate the cost- effectiveness of nicotine gum relative to antihypertensives or antihyper- lipemics since they do not reflect changes in the quality of life that may be associated with treatment. Measures of outcome should ideally in- corporate these changes along with changes in life expectancy; quality- adjustment of life-years is one method by which this may be accomplished (32). Because the treatment of hypertension or hypercholesteroletnia in- volves a lifelong regimen of drug therapy, the side effects of treatment may partially offset the value that patients place on gains in life expec- tancy. In contrast, the period of treatment with nicotine gum is typical- ly brief, and side effects are relatively minor. Consequently, comparison of treatment strategies in terms only of their respective costs per (unad- justed) year of life saved may somewhat understate the relative cost- effectiveness of nicotine gum therapy. We emphasize, however, some of the methodologic limitations of this study. For one, limited data forced us to estimate the effectiveness ofnico- tine gum in a primary care setting using data principally drawn from trials conducted in smoking cessation clinics. Patients enrolled in these clinics were volunteers, and therefore likely to be more compliant and motivated to stop smoking than most patients who would be seen in a physician's office. We used data from these trials, however, only to estimate the phar- macologic effect of nicotine gum relative to placebo gum. Motivation and compliance should not bias this measure, since they would affect equally patients using placebo and active gum. Furthermore, by mea- suring efficacy relative only to that of inactive gum, we have necessarily ignored any placebo effect that may contribute to the actual effective- ness of nicotine gum in clinical practice. The conservative nature of our estimates may be suggested by the largest advice-gum trial in primary care practice, which reported one-year cessation rates for these interven- tions of 4.1% and 8.8"/<) respectively (26). Nonetheless, the necessity to 288 289

use an indirect estimate of the effectiveness of nicotine gum in a primary c:aree setting is a clear limitation of our analysis, and should be borne in mind. Limited data also forced us to estimate the costs of nicotine gum ther- apy. While we have based our costs on reports of ad libitum gum use in clinical trials, this may be different in a primary care setting and cost- effectiveness may vary accordingly. We also have not taken account of the possibility of patient addiction to nicotine gum. Although the propor- tion of patients using freely provided gum at one year has been reported to be no higher than about 5% (2,10,15,23), this possibility may increase the actual costs of therapy and adversely affect cost-effectiveness. Since our study draws upon the findings of earlier studies, it is also important to acknowledge the possibility of publication bias in the clinical literature, which may distort the results of our meta-analysis. Any ten- dency by clinical investigators not to report the results of trials that find no significant difference between active and placebo gum would result in our overestimating the efficacy and cost-effectiveness of nicotine gum. Although we made no attempt to quantify this possible source of bias, we believe that incomplete reporting of trials is unlikely. Finally, we should note that, due to limited data, we did not take ac- count of any possible relationship between the likelihood that a patient will quit smoking and the number of cigarettes typically smoked, the age at which a patient began to smoke, the length of time a patient has stnoked, or the age at which a patient attcmpts to quit smoking. These may also affect the cost-effectiveness of treatment with nicotine gum. Conclusion We believe that our study has important implications for physicians and the treatment of tobacco dependence. Our results suggest that, in terms of its costs and clinical benefits, nicotine gum therapy compares quite favorably with other widely-accepted medical practices. Despite relatively modest cessation rates for patients who are prescribed nico- tine gum in a primary care setting, pharmacologic treatment of tobacco dependence is a viable and cost-effective patient-care strategy. Our results may therefore encourage physicians to take a more active role in treat- ing their patients who smoke. References 1. Blum A. Nicotine chewing gum and the medicalization of smoking. Ami. Intern. Med. 1984; 101:121-123. 2. British Thoracic Society. Comparison of four methods of smoking withdrawal in patients with smoking related diseases. Br. Med. J. 1983; 286:595-597. 3. Cochran WG. The combination of estimates from different experi- nients. Biometrics 1954; 101-129. 4. Doll R, Peto R. Mortality in relation to smoking: 20 years observa- tion on male British doctors. Brit. Med. _J. 1976; 2:1525-1536. 5. Evans D, Lane DS. Long-term outcome of smoking cessation work- shops. Am. J. Pub. Health 1980; 70:725-727. 6. Fagerstrom KO. A comparison of psychological and pharmacolog- ical treatment in smoking cessation. J. Beh. Med. 1982; 5:342-351. 7. Fee WM, Stewart MJ. A controlled trial of nicotine chewing gum in a smoking withdrawal clinic. Practitioner. 1982; 226:148-151. 8. Glass GV Primary, secondary, and meta-analysis of research. Educa- tional Researcher 1976; 5:3-8. 9. Handel S. Change in smoking habits in a general practice. Postgrad. Med. J. 1973; 49:679-681. 10. Hjalmarson Al. Effect of tucotine chewing gum in smoking cessa- tion. A randomized, placebo-controlled, double-blind study.J. Am. Med. Soc. 1984; 252:2835-2838. 11. Hughes JR, Miller SA. Nicotine gum to help stop smoking. J. Am. Med. Assoc. 1984; 252:2855-2858. 12. Hunt WA, Bespalec DA. An evaluation of current methods of modifying smoking behavior. J. Clin. Psych. 1973; 30:431-438. 13. Jarnrozik K, Fowler G, Vessey M, Wald N. Placebo controlled trial of nicotine chewing gum in general practice. Brit. Med. J. 1984; 289:794-797. 14. Jamrozik K, Vesscy M, Fowler G, Wald N, Parker G, ct al. Controlled trial of three different antismoking interventions in general practice. Brit. Med. J. 1984; 288:1499-1503. 15. Jarvis MJ, Raw M, Russell MA, Feyerabend C. Randomiscd controlled trial of nicotine chewing-gum. Brit. Med. J. 1982; 285:537-540. 16. Jarvik ME, Schneider NG. Degree of addiction and effectiveness of nicotine gum therapy for smoking. Am.J. Psychiat. 1984; 141:790-791. 17. Nicotine chewing gum (editorial). Lancet 1985; 1:320-321. 18. OckeneJK, Hymowitz N, Sexton M, Broste SK. Comparison ofpat- terns of smoking behavior change among smokers in the multiple risk factor intervention trial (MRFIT). Preu Med. 1982; 11:621-638. 19. Oster G, Colditz GA, Kelly NL. The economic costs of smoking and the benefits of quitting for individual smokers. Prev. Med. 1984; 13:377-389. 291 290

20. Pincherle G, Wright 1-113. Smoking habits of business executives. Prac- titioner. 1970; 205:209-212. 21. Public Health Service. Smoking and health: Report of the Advisory Committee to the Surgeon General of the Public Health Service. DHEW Pub. no (PHS) 1103. Washington, DC: Government Print- ing Office, 1964. 22. Public Health Service. The health consequences of smoking: Chronic obstructive lung disease: A report of the Surgeon General. DHHS Pub. no. (PHS) 84-5025. Washington, DC: Government Printing Office, 1984. 23. Raw M, Jarvis MJ, Feyerabend C, Russell MA. Comparison of nico- tine chewing-gum and psychological treatments for dependent smok- ers. Brit. Med. J. 1980; 281:481-482. 24. Rice DP, Hodgson TA, Sinsheimcr P, Browner W. Tobacco's economic costs. Paper presented at the 113th annual meeting of the American Public Health Association, Washington, DC, Nov. 18, 1985. 25. Rothman KJ, Boicc JD. Epidenuologic analysis with a programma- ble calculator. DHEW Pub. no (NIH) 79-1649. Bethesda, MD: Na- tional Institutes of Health, 1979. 26. Russell MA, Merriman R, Stapleton J, Taylor W. Effect of nicotine chewing gum as an adjunct to general practitioners' advice against smoking. Brit. Med J. 1983; 287:1782-1785. 27. Russell MA, Wilson C, Taylor C, Baker CD. Effect of practitioners' advice against smoking. Brit. Med. J. 1979; 2:231-235. 28. Schneider NG, Jarvik ME, Forsythe AB, Read LL, Elliott ML, ct al. Nicotine guin in smoking cessation: a placebo-controlled, double- blind trial. Add. Beh. 1983; 8:253-261. 29. Stewart PJ, Rosser WW. The impact of routine advice on smoking cessation from family physicians. Can. Med. Assoc. J. 1982; 126:1051-1054. 30. Weinstein MC, Fineberg HV Clinical Decision Analysis. Philadelphia, PA: WB Saunders Co., 1980. 31. Weinstein MC, Stason WB. Cost-effectiveness of interventions to pre- vent or treat coronary heart disease. Ann. Rev. Pub. Health 1985; 6,41-63. 32. Weinstein MC, Stason WB. Hypertension: A Policy Perspective. Cam- bridge, MA: Harvard University Press, 1976. Discussion: Economics of Treatment Chair: Judith K. Ockene, Ph.D. University of Massachusetts Medical School, Worcester Presenters: David P.L. Sachs, M.D. Stanford University School of Medicine, Stanford, California Gerry Oster, Ph.D. Policy Analysis, Inc., Brookline, Massachusetts DR. OCKENE: Since this conference is about to close and we are run- ning late, we have only a few minutes for any questions. FLOOR: Dr. Oster, a question. Did I understand your last statement to mean that the probability of a Type II error was 60%? DR. OSTER: What the results of our power analysis suggested was that if there was a treatment effect of the size that we estimated has been demonstrated in the trials, namely a 35% increase in the rate of cessa- tion, then the British Thoracic Society study had only a 40% chance of detecting that treatment effect. FLOOR: Okay, so that is a Type 11 error of 60%. DR. OSTER: Yes, it is. But I'd rather not speak in terms of Type I and Type II. DR. SCHNEIDER: Is there some reason why the cost of cigarettes is not subtracted from some sort of figure when you estimate cost- effectiveness? DR. OSTER: Well, we didn't look at the cost of cigarettes. We wanted to look just at how much would be spent on the intervention and how many lives would be saved given those dollar expenditures. DR. SCHNEIDER: It's a different target audience for whom it's cost- effective, then. You could even take your results and make them better for the patients spending money by subtracting what they would save from quitting successfully, is that correct? DR. OSTER: Presumably you could use that as an offset against the cost of therapy, yes. DR. OCKENE: I'd like to thank our speakers for very fascinating anal- yses, and I hope that one day they present them to the various insurers so that we can proceed with the much needed provider reimbursements in the United States. Thank you. 292 293

VIII: The Future of Pharmacological Treatment for Smoking

The Future of Pharmacological Treatment for Smoking M.A.H. Russell, M.R.C.P. Institute of Psychiatry London Introduction In drawing this conference to a conclusion, it is my brief here to con- sider the future of pharmacological treatments for smoking. The con- cept is not a new one. In 1936, Dorsey suggested the use of lobeline as a substitute for nicotine (3). Various drugs have been tried since then, sometimes in "homeopathic" rather than "pharmacological" doses, but none had been shown to be clinically effective until the development of nicotine chewing gum. At present this is the only drug-based treatment which has proven value in double-blind, placebo-controlled trials. This is not the place for any thorough review or full theoretical ap- praisal of the issues, so I offer no more than a short outline of future pos- sibilities for the next five to ten years. I shall consider first the possibili- ties for improving the product we already have, namely nicotine chewing gum; secondly, the other methods of nicotine substitution which hold promise for the future; and, finally, the potential of other non-nicotine- based pharmacological approaches. Improving nicotine gum and its clinical use The present product, Nicorette, is a considerable improvement on the earlier prototypes that were used in trials in the late 1960s and early 1970s. The flavour has been improved as has the buffering capacity. Despite its success and proven efficacy, it remains a relatively inefficient source of nicotine. Its acceptability to smokers is limited. Many are put off by its taste and irritancy. For some users, the amount of chewing required is irksome, difficult or socially embarrassing, while for others the amount of nicotine absorbed is insufficient to adequately relieve withdrawal. At best, the long-term success rates with the gum remain obstinately at only I thank the Medical Research Council for financial support and Pamela Hancox for secretarial help with the manuscript. 30%-40%. So, there is plenty of room for improvement, both in the for- mulation of the product and the clinical management of those who use it. These two aspects are, therefore, in my opinion, important priorities for future research. Dosage is an obvious unresolved problem. Dosage in the United States is currently restricted to the 2 mg nicotine gum, while in other coun- tries a 4 mg nicotine gum is also available. It is not known whether, or by how much, use of 4 mg versus 2 mg gum would enhance success rates among heavier smokers, and whether or not the higher dose would in- crease the proportion of users who become dependent on the gum. Perhaps there should also be a 1 mg nicotine gum for lighter smokers or for use during initiation and weaning. At the clinical level, for how long should subjects be encouraged to use the gum? One month, three months, six months, etc? At what point would an increase in the propor- tion achieving initial success from longer use be offset by problems with weaning? Should subjects be encouraged to use as much gum as possi- ble to prevent craving or should they chew only when they feel an urge to smoke? All these and many more questions remain to be answered. Apart from questions about improving and providing a choice of flavours and consistency, there are several possibilities for altering the bio- availability of nicotine from the gum. Besides varying the nicotine con- tent, should the gum be altered to enable the nicotine to be released more easily and quickly? This might prevent tired and aching jaws resulting from the heavy chewing necessary to obtain reasonably high buccal nico- tine concentrations, but could also lead to greater wastage of nicotine through swallowing. Should the alkaline buffers be changed? If the pH in the mouth during chewing were raised from pH 8 to pH 9, the propor- tion of nicotine present as absorbable free base would rise from about 50% to about 90%. This would have an effect almost equal to doubling the concentration of nicotine in the mouth, but the rise in pH would prob- ably increase local irritancy. It may be difficult to strike the right balance. There is little doubt, however, that approaches along these lines could lead to the development of more palatable and easily chewed gums of varied flavour and nicotine dose, from which more rapid absorption of nicotine would be reflected in sharper rises in blood nicotine concentra- tions to levels more closely comparable to those obtained from smoking. A wider choice of more acceptable and efficient gums should enable proportionally more smokers to benefit from more effective nicotine replacement with accordingly higher degrees of success. It would be a pity if the manufacturers remain complacent behind the protection of their patent, leaving clinicians and their clients to continue to strive to over- 296 297

come the lirnitations of what is essentially a 1975 model gum. A modest shift in funding from promotion and marketing to research and develop- ment would benefit smokers. It might also be more effective in increas- ing sales in the long-term. Other methods of nicotine substitution As mentioned above, ivcotine chewing gum is a slow and relatively inefficient method of nicotine delivery. The fact that it works so well despite its limitations is very encouraging and suggests that better sub- stitutes may give even better results. During my earlier presentation (this volume), I outlined a range of possible nicotine substitutes together with the principles and potential problems of their use. Only two of them have so far been subjected to even preliminary study and we have had detailed presentations of the latest findings which seem highly promising. The potential of transdermal nicotine was outlined by Jed Rose (this volume). His studies indicate that nicotine-containing skin patches produced measurable pharmacological effects on heart rate and blood pressure and, more importantly, that they also appear to relieve the craving of cigarette withdrawal. I was not convinced, however, of the relevance of his data on salivary nicotine concentrations. The obvious next step is to evaluate transdcrmal absorption with measurements of blood nico- tine concentrations. The other nicotine substitute discussed in some detail was the nasal nicotine solution (NNS) which is used as a kind of liquid nasal snufl: Mar- tin Jarvis (this volume) presented the results of his preliminary studies. He showed that a drop of the solution containing 2 mg nicotine produced peak levels of blood nicotine within ten minutes. This is similar to the time it takes to smoke a cigarette. The absorption of nicotine is clearly nnuch more rapid than it is from nicotine gum which takes about 20 to 30 minutes to produce a very flat peak blood nicotine level. The NNS appears to be reasonably acceptable to the kind of sm: =kers who attend specialised withdrawal clinics and to help them by relieving withdrawal symptoms during cessation. Other than a brief mention in my earlier presentation, the tobaccoless smoke-free cigarette (Favor) has not been covered at this conference, due to the lack of any adequate data on the product (6). Yet, its potential is considerable in that it is only by absorption through the lungs that any form of nicotine substitute could mimic the puff-by-puffpost-inhalation high-nicotine boli obtained from cigarette smoking. For reasons that I mentioned earlier, it is doubtful whether this product will deliver ade- quate nicotine to the lung alveoli. Unlike NNS and the transdermal nicotine patch which are not yet widely available for research, the smoke- free cigarette is already on sale in many parts of the United States and does not require a medical prescription. There is clearly an urgent need for proper evaluation of this product. In view of all that has been said at this conference about the predominant role of nicotine in smoking, starting with Jerry Jaffe's key- note opening address and echoed subsequently by many speakers, and in view of the striking success ofrucotine gum as an aid to cessation, the potential of new and better forms of nicotine substitution is great. It is not unreasonable to hope that in the near future clinical researchers will have access to a range of substitutes suited to the different clinical problems seen in different types of smoker at difrerent stages ofthe cessation process, and intended for use separately or in combination. Such a prospect is an exciting one for researchers and could bring far more effective help to smokers. Other pharmacological approaches As mentioned in the introduction, pharmacological treatments for smoking have been sought for many years and many different drugs have been tried. Until the development of nicotine gum, however, the capac- ity of other drugs to enhance success rates over and above attention- placebo effects has been as ineffective as those of psychological and be- havioural treatments such as hypnosis, acupuncture, electric aversion, cue exposure, rapid smoking, etc. Drugs used to aid smoking cessation can be divided roughly into three main categories: those that act as substitutes for nicotine by reproduc- ing some of its effects; nicotine antagonists to block the effects of nico- tine (extinction model); and drugs which supposedly block or counter- act some of the unpleasant effects of nicotine withdrawal. However, it is not always clear in which of these capacities a given drug may function. Lobeline was one of the first drugs to be used as a potential substitute for nicotine, but its pharmacological effects are weak and it is ineffective as an aid to cessation. The story is similar with other potential substi- tutes. Stimulants, such as amphetamine, increase smoking behaviour rather than diminish it, and sedatives are also ineffective aids to cessation. The subtle dual stimulant and sedative actions of nicotine appear to be unique and animals discriminate the effects of nicotine from those of all other drugs that have been tested (5). It seems, therefore, that no drug is yet available which can substitute for nicotine. In contrast to providing pharmacological substitution for the loss of nicotine action, another approach is the blockade of nicotine action (8). 299 298

A drug that blocks the rewarding effects of nicotine could theoretically be used as an agent of extinction. The effects of nicotine can be reliably blocked by the nicotinic cholinergic antagonist, mecamylamine. In the short-term, mecamylatnine appears to increase rather than diminish ad libitum smoking behaviour, presumably due to an attempt on the part of the smoker to overcome the blockade. Longer term use would be re- quired to test the extinction hypothesis. So far this has been done in only one uncontrolled study with inconclusive results (9). Blockade of some of the indirect effects of nicotine (i.e., those medi- ated by the actions of the neurotranstnitters and hormones released by nicotine action) have potential as extinction agents only so far as the ef- fects blocked contribute to the reinforcing action of nicotine. Thus the beta-noradrenergic blocker, propranolol, does not appear to affect the satisfaction derived from smoking (2) and is consequently ineffective as an aid to cessation (4). It is my opinion, however, that the extinction model based on old-fashioned learning theory has little poten ild i n view of the importance of cognitive factors and the capacity of humans to discriminate between conditions of smoking with and without the blocking drug. Ex- tinction could only be achieved in humans if the blocking drug were ad- tninistered without their knowledge, which would obviously be unjusti- fied on ethical grounds. On the other hand, if blocking agents are unlikely to aid initial cessation, it is still possible, as Stolerman suggests, that they may have a role in helping to prevent relapse (8). Apart from the two approaches mentioned, namely pharmacological substitution for, or blockade of, the reinforcing direct or indirect actions of nicotine, it is possible that craving or other withdrawal effects could be relieved by drugs which affect various subtle "rebound" phenomena in the nervous system which may result from withdrawal of nicotine. Since nicotine affects the activity of many different neurotransmitter systems in the brain, one can expect the effects of its withdrawal to be neurophar- macologically complex. A number of drugs have been found to have pos- sible effects on reducing ad libitum smoking, the desire to smoke or the craving for cigarettes during withdrawal. One example is the alpha-2- noradrenergic agonist, clonidine, which, as we have heard from Sandy Glassman (this volume), may reduce craving during the early stages of withdrawal. Another study has suggested that the opioid antagonist, naloxone, may inhibit stnoking and reduce the desire to smoke (7). Fluox- etine, a 5HT re-uptake inhibitor, may decrease craving and is currently undergoing multi-centre trials in the United States as a potential aid to smoking cessation. Finally, various uncontrolled attempts have been made to aid cessation by correcting some of the hormonal changes which may follow nicotine withdrawal. ACTH injections (1) and use of a synthetic form of vasopressin are two examples. Conclusion The current rapid advance of the neurosciences will no doubt lead to greater understanding of the pharmacology of nicotine and the mechan- isms underlying its actions. This progress may well provide clinicians of the future with a range of highly specific and effective aids to smoking cessation. But, in my view, this is still a long way off. For the near fu- ture, however, the prospect of a range of improved versions of the effec- tive nicotine substitute we already have in the form of nicotine chewing gum, coupled with a range of promising alternative nicotine substitutes such as NNS and the nicotine skin patch, is a realistic hope for clinicians and their clients. In my opinion, this is the path we should pursue to achieve more effective pharmacological treatments for smoking within the next five years. But we should not forget that, whatever new phar- macological treatments become available in the future, supportive ap- proaches will continue to be relevant. We are unlikely to ever see a drug that will cure smoking as an antibiotic cures an infection. References 1. Bourne S. Treatment of cigarette smoking with short-term high- dosage corticotrophin therapy: Prelitninary communication.J. Roy- al Soc. Med. 1985; 78:649-650. 2. Carruthers M. Modification of the noradrenaline related effects of smoking by beta-blockade. Psychological Med. 1976; 6:251-256. 3. Dorsey JL. Control of the tobacco habit. Annals Int. Med. 1936; 10:628-631. 4. Farebrother MJB, Pearce SJ, Turner P, Appleton DR. Propranolol and giving up smoking. Brit. J. Diseases of the Chest 1980; 74:95-96. 5. Hendry JS, Rosecrans JA. Effects of nicotine on conditioned and un- conditional behaviors in experimental animals. Pharm. and Ther. 1982; 17:431-454. 6. Jacobson NL, Jacobson AA, Ray JP. Non-combustible cigarette: al- ternative method of nicotine delivery. Chest 1979; 76:355-356. 7. Karras A, Kane JM. Naloxone reduces cigarette smoking. Life Sciences 1980; 27:1541-1545. 8. Stolertnan IP. Could nicotine antagonists be used in smoking cessa- tion? Brit. J, Add. 1986; 81:47-53. 300 301

9. Tennant FS, Tarver AL, Rawson RA. Clinical evaluation of inecamyla- mine for withdrawal from nicotine dependence, Pp. 239-246 in LS Harris (Ed.) Problems cfDrug Dependence, NIDA Research Monograph 49, Rockville, MD: United States Department of Health, Education, and Welfare, 1984. 362.1 Smoking Behavior 0 and Policy Conference Series oAfiE 188UE0 TO TOB11CCO 'INSTITUTB LIBRARY 302