Tobacco Documents Online

a JOUfii/ O(Consulung 1nd Cltntal Psvchoaogy CapcT,ght 1993 b, tne ,rn,an a,.cho oR,cai assocuna,. inc tV933 vot.6l, '.o. S.'SI-I60 W_:-606Y.v3,S3 00 I I Pharmacotherapy for Smoking Cessation: Unvalidated Assumptions, Anomalies, and Suggestions for Future Research John R. Hughes ~ I ~ I I I I I I I I I This article questions several assumptions about the rationale for pharmacological therapies for smoking cessation, including whether (a) future smokers will be those more dependent on nicotine and thus in greater need of nicotine replacement or other pharrnacotherapy, (b) transdermal nicotine and nicotine gum work by reducing withdrawal symptoms, and (c) clonidine works by decreasing sympathetic arousal. After describing currently available phartnacotherapies, the article also de- scribes several unexpected findings that need to be taken into consideration by clinicians: (a) trans- dermal nicotine is effective when given without psychological therapy, (b) transdermal nicotine and nicotine gum do not consistently dacrease postcessation weight gain, (c) high level of nicotine de- pendence does not consistently predict better response to transdertnal nicotine, and (d) clonidine is effective in women but not in men. The article poses other questions for future research. Several recent articles have reviewed pharmacotherapies for smoking cessation (Fiore. Pierce, Remington, & Fiore, 1990; Glassman & Covey. 1990: Goldstein, Niaura. & Abrams. 1991; Gourlay & McNeil. 1990: Jarvik & Henningfield. 1988; Nunn- Thompson & Simon, 1989; Prignot, 1989; Sachs, 1991 b; Sachs & Leischow. 1991; Schwartz. 1992). This article differs from those reviews in that it focuses on (a) the empirical basis for common assumptions in pharmacotherapy for smoking, (b) anomalies found in studies of pharmacotherapy, and (c) ques- tions for future research. Role of Pharmacotherapy in the Future Public health efforts and social pressure are decreasing initi- ation of smoking and increasing smoking cessation so that the prevalence of smoking is decreasing about 0.5-1.0%/year (Pierce. Fiore. Novotny. Hatziandreu, & Davis, 1989). How- ever some have hypothesized that those who are quitting are the less dependent smokers (Coambs. Kozlowski, & Ferrence, 1989; Hughes & Glaser, 1993). If this "selection hypothesis" is true, then the remaining smokers will consist of mostly smokers who are heavily dependent on nicotine. Because more dependent smokers appear to be especially benefitted by pharmacotherapy (Fagerstrom & Schneider, 1989), future smoking cessation treatments may especially need to include pharmacotherapies. Although this selection hypothesis is logical, the data to sup- port it are mixed. The best way to test the hypothesis would be to determine the prevalence of nicotine dependence in a popu- lation-based sample and follow it over time. This has not been done. Thus, we are left with examining indirect measures of dependence. This article was supported by Research Scientist Development Award K02 DA-00109. Correspondence concerning this article should be addressed to John R. Hughes, Human Behavioral Pharmacology Laboratory, Department of Psychiatry. Univcrsity of Vermont, 38 Fletcher Place, Burlington, Vermont 05401-1419. Several studies have suggested that heavier smokers are more dependent (Killen, Fortmann, Telch. & Newman, 1988). The proportion of smokers smoking 25 or more cigarettes per day increased frum 24% to 3 1 % between 1965 to 1976 but only in- creased from 3 1 % to 33% between 1976 to 1985 (U.S. Depart- ment of Health and Human Services. 1988, Table 10). The "quit ratio" is the fraction of ever-smokers who are ex- smoken. This could be a measure of dependence i<n that the ability to quit is an important measure of dependence. The quit ratio in the United States is currently about 45% (Centers for Disease Control, 1990). If the selection hypothesis is true, then over time the quit ratio should begin to level out as only the more dependent smokers remain. Data for the iast 10 years shows that quit ratios are continuing to increase with no evi- dence of leveling off (U.S. Department of Health and Human Services. Figure 4, 1988). However, even this result does not necessarily rule out the selection bias. For example, with social pressure dramatically increasing over time, quit ratios could be expected to be accelerating but the selection bias of remaining dependent smokers may have prevented this acceleration. In summary, although the hypothesis that future smokers will be those more dependent on nicotine is logical, empirical data to substantiate it are not available. Determining the validity of this hypothesis is crucial to planning what mix of public health, minimal interventions. and intensive behavioral and pharma- cological therapies are needed. Replacement Therapies Pharrnacotherapy for drug dependencies typically can wor" by means of several mechanisms: replacement, blockade, withL7 drawal relief, making intake aversive, and so forth (Jarvik 8ai~6 Henningfield, 1988). The most developed group of inedications= for smoking cessation are replacement or substitution therapies= The most common rationale for their use is to abate withdrawa= symptoms initially and then to avoid further withdrawal byCD gradually tapering the replacement therapy. O~ ~ 751

I I I I I I I I I I I I I ~ I I JOHN R. HL'G'iES 0 ~ 0 0 0 0 0 0 0 ~ e ~ 0 -- 88 -----8 ---0 - 8 NG NG TN TN alone + PT alone + PT Figure 1. Odds ratios for nicotine gum (NG) and transdermal nico- tine (TN) alone or with psychological therapy (PT). (The horizontal line represents the median odds ratio. References for each data point are available from the author upon request.) Brie/'Review o(Nicotine Replacement Therapies Nicotine polacrilez (nicotine gum: NG). To illustrate the efficacy of NG. I calculated the odds ratio of long-term absti- nence for NG versus placebo (PL) in studies with 6- or 12- month follow-ups. These odds ratios can be thought of as the relative increase in probability of long-term abstinence. Odds ratios were used, as they are probably less biased by sample characteristics. study-setting, and so forth, than percent absti- nent or increase in percent abstinent. Because of the need for brevity, I simply plotted the odds ratios and determined a me- dian rather than conduct a true meta-analysis. Previous studies were divided into two major groups. Several studies examined NG when given with brief advice, usually in medical settings. Other studies examined NG when given with a psychological therapy (PT), usually in a cessation clinic set- ting. The PT ranged from nonstructured group support to structured behavior therapy. When given with brief advice. NG appeared to slightly in- crease quit rates over PL (Figure 1). However, when given with PT. NG increased quit rates twofold over PL gum. This result is consistent with previous narrative and meta-analytic reviews. (Fagerstrom. 1988; Gourlay & McNeil, 1990; Lam, Sze, Sacks, & Chalmer, 1987). Several studies have found NG especially beneficial to smok- ers who are heavily dependent on nicotine (Fagerstrom & Schneider. 1989): for example, NG often increases quit rates in low-dependent smokers by 10-1596 but increases quit rates in high-dependent smokers by 20-30%. Serious side effects from the gum have not been reported (Hughes. 1993b). Minor side effects are due to chewing (sore jaw, gas) and to nicotine (nausea). Dependence on the gum (use for >6 months) occurs in about 20% of those who quit with the gum (Hughes, Gust, Keenan, Skoog, Fenwick, & Higgins, 1991). The major problem with the gum has been poor compliance that is due in part to side effects, the need for significant instruc- tions in its use (when to chew, how to chew, avoid acidic bever- ages, etc.). and social acceptability (Hughes, 1986). In addition, ad-lib use of the 2-mg dose of NG does not reliably decrease craving (West & Schneider, 1987), and whether it decreases weight gain varies across studies (Gritz, Klesges, & Meyers, 1989: Klesges. Meyets, Winders. & French. 1990). Scheduled gum use (e.g.. one piece per hour). higher doses (e.g., 4 mg), or more frequent use of gum may decrease craving (Sachs. 199Ia) and weight gain (Gross, Stitzer. & Maldonado, 1989; Leischow, Sachs, Bostrom. & Hansen. 1992). Transdermal nicotine (TN: nicotine patch). TN has several substantial advantages: once-daily dosing, social acceptability, and more steady and greater nicotine replacement. Nicotine patches reliably decrease withdrawal, including craving, but their effect on weight gain is unclear (Fagerstrom. Hurt. Sawe, & Tonnesen, 1992; Hughes & Glaser, 1993: Palmer & Faulds, 1992). To examine the efficacy of TN, I calculated a similar set of odds ratios. In this analysis, when given with brief advice, TN increased quit rates over PL patches by a factor of 2.5 (Figure 1). When given with PT, TN again appeared to increase quit ratios by a factor of 2.5; however, only three such studies are available. Other nicotine replacements. Nicotine inhalers (Hajek. Jar- vis. Belcher, Sutherland, & Feyerabend, 1989) and nasal sprays (Jarvis, Hajek, Russell, & West. 1987; Perkins, Grobe, Stillet; Fonte, & Goettler, 1992; Sutherland et al., 1992) are used to more closely mimic the bolus doses of nicotine from cigarettes, as some believe replicating the rapid absorption of nicotine may improve the efficacy of a replacement therapy (Pomerleau & Pomerleau, 1984). When fully developed, these products will probably be used as transitions to, or adjuncts with, TN. Initial results with the nasal sprays alone are encouraging (Jarvis et al., 1987: Sutherland et al., 1992); however, some data suggest that abuse liability and long-term use may be problems (Hughes, 1989: Jarvis et al., 1987; Sutherland et al., 1992; see also Hen- ningfield & Keenan. 1993). The side effects of these products include local irritation (e.g., runny nose) and social acceptabil- ity; however, few subjects have had to drop out because of side effects. One other replacement therapy is lobeline, a chemical analog of nicotine used in over-the-counter preparations. Its efficacy has not been demonstrated (Kozlowski, 1984). Anomalies ojNicotine Replacement Studies Effect ojdose. Dose is a major determinant of the effects of nicotine (Henningfield & Woodson, 1989), including therapeu- tic effects (Fagerstrom, 1988), yet only one TN outcome study has examined more than one dose (Transdermal Nicotine Study Group, 1991). Interestingly, this study found that abstinence rates increased linearly with dose (Figure 2, upper panel). Thus, an important study is to determine whether even higher doses of TN would produce greater abstinence rates. This is especially important given the low incidence of side effects with TN (Fag- erstrom et al., 1992; Hughes & Glaset; 1993; Palmer & Faulds, 1992). TN versus NG. The clinical experience of many physicians and investigators (myself included) is that TN is more effective than NG. However, a comparison of studies of TN plus PT ver- sus studies of NG plus PT suggests that TN is not substantially I

I I I I I I I I I I I I I I I I I I SPECIAL SECTION: PHARMACOTHERAPY FOR SMOKiNG CESSATION 7j3 100-1 ~ Q 60, H 1.. 401 a> a ~ 0 Q tn 20 2i R 3 ~ .0 E 0 U 1 0 0 0 1 7 T 14 0 O 00 0 0 7 14 21 Dose (mg/24hr) Figure 2 Data from Study A of the Transdermai Nicotine Study Group (Transdermal Nicotine Study Group, 1991). (The upper panel represents end-of-treatment [ 12 weeks later] verified, continuous absti- nence rates. Each data point represents results for one of five sites in the study. The lower panel represents combined withdrawal symptom scores for the first 6 weeks. Each data point represents results for I of 6 weeks. ) superior to NG (Figure 1). Cross-study comparisons, as that shown in Figure 1, are always dangerous because ofconfounds; for example, perhaps the TN trials entered more dependent smokers or used less effective PT. In addition, only three long- term studies of TN plus PT are available. Thus far, only one study has directly compared TN plus NG. This study reported 3-month abstinence rates of 55% for TN plus PT versus 43% for NG plus PT (Meier-Lammermann, Meyer, & Boleskei, 1990). Nicotine replacement and postcessation weight gain. Despite the robust data with animals that nicotine suppresses eating and weight gain (Grunberg, 1990), neither NG nor TN have consis- tently reduced postcessation weight gain in humans (Fagers- trom et al., 1992; Hughes & Glaset; 1993; Palmer & Faulds, 1992; see also Perkins, 1993). The inconsistent effects in NG studies could have been due to poor compliance; however, cotn- pliance with TN is good. but its effects on weight gain are also inconsistent. Several explanations are possible. First, whether the doses tested in TN and NG trials are comparable to doses that sup- pressed weight in animals is unclear. Second, perhaps abstinent smokers were actively suppressing weight gain, masking drug effects; however, weight gain in most trials was substantial (Fag- erstrom et al., 1992; Palmer & Faulds. 1992). Third, perhaps nicotine must be given as a bolus during hunger pangs to reduce eating and weight. However, nicotine reduced weight in animals when given as a continuous, noncontingent infusion. TN in medical settings. Whether NG substantially increases quit rates over PL when given with brief physician advice is con- troversial (Gourlay & McNeil, 1990: Lam et al., 1987). In con- trast to NG. TN is clearly more effective than PL when given with brief physician advice (Fagerstrom et al., 1992; Hughes & Glaser, 1993: Palmer & Faulds, 1992). Historically, NG's lack of efficacy in medical settings has been attributed to the lack of an adjunctive PT (Hughes. 1991: see also Orleans. Kristeller. & Gritz, 1993). However, the success of TN in medical settings where PT was not given suggests that some other factor may be operating. One possibility is compliance. Direct empirical comparison of compliance between NG and TN is difficult be- cause of the different routes of administration; however, most investigators believe that compliance is greater with TN than with NG (Hughes & Glaser, 1993). Thus, perhaps with better compliance in medical settings, NG would also be effective. Consistent with this notion, most studies in which NG was pre- scribed to be taken on a regular schedule report gAater nico- tine-PL differences (Cooper & Clayton, 1990; Goldstein, Ni- aura, Follick, & Abrams, 1989: Killen, Fortmann, Newman, & Varady, 1990, Sachs. 1991 a). Nicotine dependence and response to NG and TN. Another anomaly is that the degree of dependence on nicotine as mea- sured by the Fagerstrom Tolerance Questionnaire (FTQ) reli- ably predicts differential response to NG (Fagerstrom & Schnei- der, 1989) but not to TN (Hughes & Glaset; 1993; Palmer & Faulds. 1992). Both TN and NG studies find that overall absti- nence rates are worse in high-dependent smokers than in low- dependent smokers. However in NG studies, the difference in abstinence rates for NG versus PL is small (0% to 18%) for low- dependent smokers, but is large (+ 18% to +37%) for high-de- pendent smokers. A similar interaction is not consistently seen in TN studies. Usually, discussions of the predictive ability of the FTQ in NG studies focus on identification of high dependent smokers who do poorly unless they receive nicotine replacement. One possible hypothesis to account for differences in the predictive power of the FTQ between NG and TN focuses on the low rather than high-dependent smokers. This hypothesis proposes that low-dependent smokers are less tolerant to the aversive side effects of nicotine replacement and that NG is more likely to produce aversive nicotine side effects than TN (Fagerstrom et al., 1992; Hughes & Glasez; 1993; Palmer & Faulds, 1992). Thus, in this scenario, low-dependent smokers should be less compliant with NG and thus not show much of a NG versus PL effect on quit rates. On the other hand, since TN has few side effects, low-dependent smokers should be compliant with TN and thus should show a TN versus PL difference. I

I I I I I I I I I I I I I I I I I I JOHN R. HUGHES In addition to explering these anomalies. several other studies on nicotine replacement are indicated. Among the more impor- tant questions are as follows: (a) Because many cardiac events occur in the early morning, do 24-hr TN patches increase risk for cardiac events, compared with PL (Hughes, 1992b)? (b) Can TN patches be safely used in pregnant women or those with a recent mvocardial infarction (Hughes, 1992b)? (c) Could TN be used to treat Alzheimer's. Parkinson's, or Tourette's disor- ders or ulcerative colitis (Newhouse & Hughes, 1991)? Clonidine. Clonidine is the only other well-tested pharmaco- logical therapy for smoking cessation (Covey & Glassman, 1991). The major potential advantage of clonidine is that it is nonnicotine therapy, and some clinicians and patients prefer to avoid continued dependence on nicotine postcessation. Cloni- dine reduces alcohol and opiate withdrawal because of its abil- ity to dampen sympathetic outflow. Initial studies found that it reduced tobacco withdrawal and craving as well (Glassman, Jackson. Walsh & Roose, 1985). Clonidine, either as pills or patches, improved abstinence over PL in all nine of the short- term trials (Covey & Glassman. 1991). Only two studies have reported follow-ups of 6 months or longer. One found positive results (Glassman. Stetner. Walsh, & Covi. 1988), and one did not (Prochazka et al.. 1992). Finally, unlike NG and TN, most of the data for the efficacy of clonidine remains in abstracts and, thus, the validity and generalizability of these studies cannot be evaluated. The mechanism of any efficacy by means of clonidine is de- batable. Tobacco withdrawal is not a syndrome of excessive sympathetic activity (Hughes. Higgins, & Hatsukami, 1990); thus, why clonidine wbuld decrease withdrawal is unclear. In addition. an unexpected but consistent finding is that cloni- dine's effects appear limited to women (Covey & Glassman, 1991). A ready explanation of this finding has not been pro- posed. Studies examining why clonidine works could be important because they may discover a new set of variables important in smoking cessation. Similarly, studies of why clonidine is effective mainly in women could provide a clue as to why smok- ing cessation rates are lower in women. Nicotine withdrawal and the effuacy, of pharmacotherapy. The most widely cited mechanism of action for nicotine re- placement is that it reduces withdrawal. However, this mecha- nism has not been demonstrated (Hughes et al., 1990; West, 1992). Evidence that nicotine reduces withdrawal is abundant (Hughes, 1992; Hughes et al-, 1990); however, whether the se- verity of withdrawal is related to the ability to stop smoking is debatable (Hughes et al., 1990; U.S. Department of Health and Human Services, 1990}. In addition, in the multiple-dose study cited earlier (Transdermal Nicotine Study Group, 1991), a low dose of TN decreased tobacco withdrawal. Moderate and higher doses did not significantly improve on this reduction in with- drawal (Figure 2, lower panel). Yet moderate and higher doses did produce higher quit rates than the low dose (Figure 2, upper panel). At least four nonwithdrawal mechanisms could account for the efficacy of nicotine replacement. First, nicotine replace- ment may be effective beca'tse of instructional or "expectancy" factors (Hughes, Gulliver, Amori, Mireault, & Fenwick, 1989). Because detectability is probably dose-related, this could also explain the dose inconsistencies of cessation versus withdrawal noted earlier. On the other hand, several studies have found that nicotine replacement appears to decrease withdrawal or in- crease quit rates independent of subjects' belief of which drug they received (Hall. Tunstall. Ginsberg. Benowitz. & Jones, 1987: Hughes. Gulliver et al., 1989: Hughes. Gust, Skoog, Kee- nan, & Fenwick, 1991; Hughes & Krahn, 1985). Second, by maintaining nicotine levels and tolerance, nico- tine replacement may make cigarettes less reinforcing and thus prevent a slip from becoming a relapse. Consistent with this hypothesis, laboratory studies indicate nicotine replacement decreases liking for cigarettes (Nemeth-Coslett & Henningfield, 1986: Perkins, et al., 1992; Rose, Herskovic, Trilling, & Jarvik, 1985). In addition, one treatment study found that subjects who smoked while on TN reported that their cigarettes did not taste as good, compared with subjects who smoked while on PL (Buchkremer, Benu, Horstmann, Opitz, & Tolle, 1989). Third, because smoking often occurs in distinct settings, oc- curs frequently, and has a short time between response and re- inforcement, the reinforcing effects of nicotine may become conditioned to environmental events or settings (Niaura et al., 1988). By making nicotine intake independent of environmen- tal evenu, nicotine replacement could disrupt this pairing of nicotine intake and environmental events and thereby make it easier for abstinent smokers to not smoke in the presence of these events. Although attractive, this mechanism has not been directly tested. In fact, although repeatedly hypothesized, to my owl- edge, only a few empirical demonstrations of conditioned effects have been published (Payne, Etscheidt, & Corrigan, 1990; Payne, Schare, Levis, & Colletti, 1991). In addition, one laboratory study did not find that nicotine replacement dis- rupted such conditioning (D. K. Hatsukami, personal commu- nication). Further work in this area could be valuable both in undenstanding the mechanism of action of replacement thera- pies and the often-hypothesized conditioning factors in smoking. Other Pharmacotherapies for Smoking Cessation Blocking Agents Blocking agents prevent the reinforcing effects of drugs. Mec- amylamine is a nicotinic blocker that increases ongoing smok- ing (presumably an attempt to overcome blockade) and de- creases the subjective effects of cigarettes (Clarke, 1991; Stoler- man, 1986). Also, because mecamylamine is a noncompetitive blocker that acts by means of a site different from the nicotinic- cholinergic receptoi; it does not appear to precipitate with- drawal. Initial trials with mecamyiamine used higher doses and short-term therapy and found prohibitive side effects. Thus, fu- ture trials should test lower doses or a gradual increase in doses of inecamylarnine and in long-term therapy. Naltrexone, an opioid blocker, has also been tested under the hypothesis that some of the reinforcing effects of smoking are mediated by opioid release (Pomerleau & Pomerleau, 1984). Whether naltrexone changes smoking behavior is unclear (Gor- elick, Rose, & Jarvik, 1989; Karras & Kane, 1980; Nemeth- Coslett & Griffiths, 1986), and at this point, clinical trials are I

~ SPECIA%L SECTION PHARMACOTHERAPt FOR SMOKING CESSATIOti I I I I I I I I I I I I I I I lacking. Gken the few side effects of naltrexone. such trials ap- pear indicated. Agents That Make Intake Aversive Some therapeutic medications work b~ making drug use cause unpleasant effects. An example in smoking is silver ni- trate lozenges or gum (Jensen. Schmidt. Pedersen. & Dahl. 1991). Silver nitrate combines with smokers' saliva to produce silver chloride which imparts a bad taste in the mouth. Al- though face valid. both old and recent trials have not found sil- ver nitrate to increase abstinence (Jensen et al.. 1991). In a similar vein. it has been surprisingly difficult to show that analogous drugs for alcohol (e.g., disulfiram ) and opioids (e.g., naltrexone) are effective (Fuller et al., 1986). Much of the lack of efficacy of blockers appears to be due to noncompliance (Brewer. l 990): thus. studies of the efficacy of silver nitrate when embedded in a behavioral program to establish and maintain medication use are indicated. Agents to Replace the Reinforcing Effects oJNicotrne Several nonnicotine medications have been tested because they may mimic effects of nicotine that are putatively associated with nicotine reinforcement (e.g.. improved performance, mood stabilization. stress relief. weight reduction. etc.) (Pomer- leau & Pomerleau. 1984). Thus. the rationale here is that such medications will provide a temporary replacement for some re- inforcing effect of smoking until abstinent smokers learn non- pharmacological means to achieve these same effects. Medications that mimic pharmacological effects of nicotine that have been tested include anxiolytics (Schwartz & Dubitsky, 1968: West. Hajek. & McNeill, 1991), stimulants (Low. Jones. Carter. & Cadoret. 1984) and anorectics (Klesges. Klesges. Mey- ers. K1em. & Isbell. 1990: Spring. Wurtman, Gleason. Wurt- man. & Kessler. 1991). Some of these appear promising. but none have had long-term positive results. Agents to .-lbate kithdrawal Because withdrawal symptoms have been thought to prevent cessation. medications to abate the specific symptoms of to- bacco withdrawal may be useful. For example, nicotine with- drawal produces anxiety, difficulty in concentrating, and hunger (Hughes et al., 1990): thus, anxiolytics, stimulants, and anorec- tics may be worthwhile. As indicated earlier, long-term trials of such agents have not beeD published. Agents to Treat Associated Psl+chopathologv Smoking is associated with an increased prevalence of a past and present history of psychiatric disorders (Breslau. Kilbey, & Andreski. 1991: Glassman et al.. 1990). In addition. both a present and a past history of depression (Anda. Williamson, Escobedo, Mast. Giovino, & Remington, 1990: also see Hall, Munoz. Reus. & Sees. 1993) or alcohol dependence (Hughes, 1993a) predict worse outcomes for smoking cessation. At present. very little is known about how to treat smokers with associated psychiatric (Goldstein et al., 1991) or witn al- cohol or drug abuse (Hughes, 1993a) problems. Antidepres- sants have been studied as a treatment for unselected groups of smokers (Edwards. Murphy. Downs, Ackerman. & Rosenthal. 1988: Jacobs, Spiker. Norman, Wohlberg, & Knapp. 1971: Sell- ers, Naranjo, & Kadlec. 1987): however. their usefulness for smokers with a past history of depression has not been tested. Such a test is especially needed given recent clinical reports that smoking cessation can precipitate a depression in such subjects (Glassman. Covey, & Stetner, 1989). In most studies. nicotine replacement decreases depressive symptoms (Hughes et al., 1990): thus, a test of the efficacy of TN in such smokers is also indicated. In terms of alcohol or drug abuse, a post hoc finding in one study was that nicotine replacement was especially ben- eficial to smokers with a past history of alcohol or drug prob- lems (Hughes, 1993a). Because there is no proven therapy for smoking cessation in recovering alcoholics, a prospective test of this finding is indicated. Sensory replacement. Although not a true pharmacological therapy, innovative work by Rose and his colleagues indicate reproduction of some of the sensory effects of smoking (e.g., the "throat scratch") by means of citric acid ( Levin. Rose. & Behm, 1990: Rose & Hickman. 1987) and other aerosols (Behm, Levin. Lee. & Rose. 1990: Levin. Behm. & Rose. 1990: Rose & Behm. 1987) decreases craving and improves short-term cessa- tion. Further clinical tests of these therapies are indicated, in- cluding tests of combined sensory and pharmacological thera- pies. In addition, these sensory therapies presumably work by means of gustatory and olfactory cues. Gustatory and olfactory stimuli are especially likely to become conditioned to pharma- cological effects (Seigel, 1983); thus, citric acid spraysand sim- ilar devices would appear to be not only promising therapies but promising research tools to study conditioning effects of smoking. Combined Pharmacotherapy and Psychotherapy When pharmacotherapy is used for a psychiatric disorder typically psychotherapy also occurs (Hollon & Beck. 1987). This is especially true for drug dependencies. For example, treatment of alcohol dependence with disulfiram alone (i.e., without counseling) (Brewer, 1990) or opioid dependence with methadone alone (General Accounting Office, 1990) is cur- rently not considered adequate treatment. On the other hand, unlike alcohol or drug abuse patients, the large majority of smokers appear to have effective coping skills and do not appear to have significant psychological problems: thus, the notion has arisen that perhaps a simple prescription of a medication with- out adjunctive psychotherapy may be feasible. In fact, as de- scribed earlier, the recent TN medical practice trials suggest that this is true. This raises the question of whether physicians should recorn- mend adjunctive PT with TN. One argument for such a recom- mendation is that trials with NG indicate that adding PT to NG substantially increases quit rates (Hughes, 1991). One argu- ment against such a recommendation is the poor feasibility ofZZ PT; often the smoker must delay cessation until a group is start-C") ing, pay money that will not be reimbursed by insurers to attend14" PT, and attend several sessions of PT (Hughes & Glaser. 1993). ~ Randomized trials of adding PT to TN appear to be espe- cially ~ cially important. As described earlier, simply comparing TN "*O ~ . ~

1 756 JOHN R. HUGHES I I I I I I I I I I I studies that did and did not use PT is inadequate because of multiple cross-study confounds (e.g., samples and training of therapists). Positive results in randomized trials of TN and PT versus TN alone would encourage physicians to work hard at trying to get smokers about to use TN to delay cessation at- tempts and to attend group PT. Small or negative results would encourage physicians to take a stepped care approach (i.e.. first try TN alone and if this fails, then try TN and PT). It is likely such trials of TN plus PT will replicate work with NG (Hughes, 1991) and show that adding PT to TN will im- prove quit rates. Hopefully, such trials will also investigate the behavioral mechanism of any additive or synergistic effects be- tween the two treatments ( Hughes. 199 l). For example, if addi- tive or synergistic effects occur, are they due to complementary effects across individuals (the highly dependent subjects benefit from TN and those with behavioral deficits benefit from PT), across behavioral mediators of relapse (TN abates withdrawal and PT abates psychological factors). or across times (TN helps increase early cessation and PT helps decrease later relapse)? Or are additive effects due to complementary effeLts in compliance (TN increases use of PT or PT increases use of TN) (Hughes. 1991)? Beyond Efficacy As described earlier. the efficacy of NG plus PT and of TN with or without PT is established: however, establishing efficacy of a treatment is only the beginning in establishing the useful- ness of a treatment. One must also examine the acceptability, availability, cost efficacy, indications. and stability of treatment (Hollon & Beck. 1987: Hughes. 1991). Although no data has been gathered on the acceptability of pharmacotherapy. fewer than 10% of smokers seen in a medical practice are willing to attend free PT (Hughes. Gust. Keenan. Fenwick. & Healy. 1989: Jamrozik. Vessey. & Wald. 1984). In the fields of alcohol and mental disorders. studies of "barriers to treatment" and studies to increase acceptance of referral to treatment have begun to identify and intervene on such factors. There is very limited knowledge of why smokers decline phar- macotherapy or PT or what can be done about it (Lichtenstein & Hollis. 1992). Similarly, the availability of pharmacotherapy and of PT var- ies widely across individuals depending on finances, geography, and so forth. Innovative research in health care delivery is needed to learn how to deal with such problems. In fact, this type of research may have a better likelihood of increasing quit rates in a given population than research on new medications. Treatments with effects that are stable (i.e., do not fade with time) are more useful. There are reasons to hypothesize that the rate of relapse back to smoking should be greater soon after an individual stops pharmacotherapy for smoking cessation (Hughes, 199 l). However. existing data for NG do not support this hypothesis (i.e., rates of relapse back to smoking after stop- ping gum use are not greater for NG than PL: Hughes, 1989). To examine if this is also true for TN. I compared rates of relapse to smoking between end of treatment and 3 months posttreatment between TN and PL groups (Table 1). In this calculation, five of the six studies showed no evidence that relapse rates were greater after stopping TN than after stopping PL. Table I Estimated Relapse Rates A - Jier Stopping Nicotine and Placebo Patches Relapse Rate Nicotine Placebo Study Fraction % Fraction % Abelin. Ehrsam. Buhler-Reichert. Imhof. Muller. Thommen. & Vesanen ( 1989) 8/20 40 5/12 42 Daughton. Heatley, Prendergast. Causey, Knowles. Roif. Cheney. Hatklid. Thompson. & Rennard (199 1) 1/39 8 /7 2 Ehrsam. Beuhler. Muller. Mauli. Schumacher. Howald. & Imhof (1991) 13/22 59 7/11 64 Sachs. Sawe. & Leischow (1992) 13/53 25 15/33 45 Tonnesen, Norregaard. Simonsen, & Sawe (1991) 26/59 44 6/ 18 42 Transdermal Nicotine Study Group(1991) 33/98 34' 9/40 23 24/69 35b Note. Fractions and percentages represent the proportion of those ab- stinent at end of treatment (usually 3 months) who relapsed by 6 months postcessation. The fractions may not be exact as estimated from percentages, graphs, and so forth. Percentages cannot be compared across studies because of differences in calculating and reporting absti- nence rates in those studies. ' 21 mg per 24-hr dose. b 14 mg per 24-hr dose. , There are also reasons to hypothesize that relapse rates will be less after the pairing of PT and pharmacotherapy than after pharmacotherapy alone: however, the little data available do not support this notion (Hughes, 1991). That pharmacotherapy for smoking cessation can be cost- effective in the long run is well documented (Cummings, Rubin, & Oster, 1989: Oster. Huse, Delea, & Colditz, 1986). In addi- tion, even intensive PT appears to be cost-effective (Altman, Flora, Fortmann, & Farquhar. 1987). Less intensive treatments are typically used prior to more intensive treatments (Fiore, Novotny et al.. 1990) plus the probability of quitting for any given attempt appears to decrease with successive attempts (Fi- ore, Novotny et al., 1990); thus, initial less costly initial treat- ments can appear to be very cost-effective, compared with lates: more costly treatments. However, intensive treatments may still be shown to be cost-effective when whole treatment algorithms are studied (discussed later). The indications for pharmacotherapy for smoking cessation are not well described (Sachs & Leischow, 1991). Historically, new pharmacotherapies run a course of overzealous use, fol- lowed by a scapegoating of the treatment, finally followed by a more rational assessment of the appropriate indications for the pharmacotherapy. Two common approaches used to decide on indications for a therapy are stepped-care and patient charac- teristic-based schemes. The stepped-care approach assumes that simpler, less intensive, less costly, safer therapies should precede more complex, more intensive, more costly, more dan- gerous therapies. To my knowledge, the efficacy or cost-

I I I I I I I I I I I I I I I I I I I SPECIAL SECTION. PHARMACOTHERAPY FOR SMOKING CESSATION effectiveness o(=such stepped-care approaches to smoking ces- sation has not been studied. The patient characteristic or treatment-matching approach assumes that we can use certain patient characteristics to iden- tifv patients %vho are likely to respond to a particular treatment (see Shiffman. 1993). This latter approach has been well inves- tigated in that the FTQ has usually been found to predict re- sponse. at least for NG (Fagerstrom & Schneider, 1989). Also, given the rationale for use of NG and TN, a history of with- drawal symptoms preventing cessation in past attempts may also be an indicator for use of NG or TN (Hughes & Glaser. 1993). At present. many schemes for determining indicators are, de factor, in effect. For example, in some health maintenance orga- nizations (HMOs) any smoker will be provided TN indepen- dent of whether simpler treatments have been tried or of whether the patient appears dependent on nicotine. Other HMOs will pay for TN only if the smoker is successful, Other HMOs will not pay for TN under any circumstances. Studies comparing such de facto systems with a delineated approach can help health plans and insurance agencies better deliver the best smoking cessation treatment at the least cost (Hughes, Wadland. Fenwick. Lewis, & Bickel. 1991). Perhaps more im- portantly- such studies can provide guidelines to physicians and patients for the most logical sequence of treatments and for treatment matching. Conclusion This article has taken a critical tone to encourage future re- search on pharmacotherapies for smoking cessation. This tone should not be taken as indicating that pharmacotherapy for smoking cessation is not effective. The large number. large sam- ple sizes. and tight designs, plus the consistency of results across studies of TN with or without PT and of NG plus PT, clearly show that these treatments are effective. This article suggests that further research in pharmacotherapy should not only test new medications but, perhaps more importantly, should focus on testing existing assumptions and anomalies, behavioral mechanisms of action of combinations of pharmacotherapy and PT. nonefficacy outcomes. and delivery systems for phar- macotherapy. References Abelin. T.. Ehrsam. R.. Buhler-Reichert. A., Imhof, P. R.. Muller. P.. Thommen. A.. & Vesanen. K. (1989). Effectiveness of a transdermal nicotine system in smoking cessation studies. Methods and Findings in Expcrrmental and Clinical Pharmacology. 11, 205-214. Altman. D. G.. Flora J. A.. Fortmann. S. P., & Farquhar. J. W. (1987). The cosi-effectiveness of three smoking cessation programs. Ameri- can Journal of Public Health. 77, 162- I 65. Anda. R. F. Williamson. D. F.. Escobedo, L. G., Mast. E. E., Giovino. G. A.. & Remington. P. L. (1990). Depression and the dynamics of smoking: A national perspective. Journal of the American Medical Association. 264. 1541-1545. Behm. F. M.. Levin. E. D.. Lee. Y. K.. & Rose. J. E. (1990). Low-nico- tine regener.d ted smoke aerosol reduces desire for cigarettes. Journal of Substance Abuse. 2. 237-247. Breslau, N.. Kilbey. M. M.. & Andreski. P. (1991). Nicotine depen- 7)7 dence, major depression. and anxiety in %oung adults. .4rchives ot General Pstrhiatrv, 48. 1069-1074. Brewer, C. (1990). Combining phartnacological antagonists and behav- ioural psychotherapy in treating addictions. British Journal of Ps.•- chtatrr. 15'. 34-40. Buchkremec G.. Bents. H.. Horstmann. M.. Opitz. K.. & Tolle. R. (1989). Combination of behavioral smoking cessavon with transder- tnal nicotine substitution. Addictive Behaviors, 14, 229-238. Centers for Disease Control. (1990). Cigarette smoking cessation- United States. 1989. Journal of the American Medical Association. 39,676-680. Clarke. P. B. S. (1991). Nicotinic receptor blockade therapy and smok- ing cessation. British Journal ofAddiction. 86. 501-505. Coambs- R. B.. Kozlowski, L. T., & Ferrence. R. G. (1989). The future of tobacco use and smoking research. In T. Ney & A. Gale (Eds.), Smoking and human behavior (pp. 337-348 ). New York: Wiley. Cooper. T. M.. & Clayton. R. R. (1990. April). Behavior modiftcation and nicotine reduction therapy with heavy smokers: Comparison of four diferent dosing strategies. Paper presented at Proceedings of the 7th Annual World Nicotine Conference. Perth, Australia. Covey, L. S.. & Glassman. A. H. (1991). A meta-analysis ofdouble-blind placebo-controlled trials of clonidine for smoking cessation. British Journal ofAddiction. 86. 991-998. Cummings. S. R.. Rubin. S. M.. & Oster. G. (1989). The cost-effective- ness of counseling smokers to quit. Journal of the American Medical Association. 261. 75-79. Daughton. D. M., Heatley, S. A.. Prendergast, J. J., Causey, D.. Knowles, M., Roif, C. N.. Cheney. R. A.. Hatlelid, K., Thompson, A. B.. & Rennard. S. I. (1991). Effect of transdermal nicotine delivery as an adjunct to low intervention smoking cessation therapy: A tan- domized placebo-controlled, double-blind study. Archive~oflnternal Medicine. 151, 749-752. Edwards. N. B.. Murphy, J. K., Downs, A. D., Ackerman, B. J., & Ro- senthal. T. L. (1988). Doxepin as an adjunct to smoking cessation: A double-blind pilot study. Journal of Psyrhiatry, 146, 373-376. Ehrsam. R.. Buehler. A., Mullet: P.. Mauli. D.. Schumacher. P. M.. Ho- wsld. H., & Imhof, P. (1991). Transdermal nicotine as an aid towards abstinence in young smokers. Schweizer Rundschau Medicine, 80. 145-150. Fagerstrom, K.-O. (1988). Efficacy of nicotine chewing gum: A review. In O. F. Pomerleau. C. S. Pomerleau. K.-O. Fageistrom, J. E. Hen- ningfield, & J. R. Hughes (Eds.), Nicotine replacement: A critical eval- uatiort (pp. 109-128). New York: Alan R. Liss. Fagetstrom, K.-O., Hurt. R. D.. Sawe, U., & Tonnesen, P. (1992). Ther- apeutic use of nicotine patches: Efficacy and safety. Journal of Smoking-Related Diseases. 3. 247-261. Fagerstrom, K.-O.. & Schneider N. G. (1989). Measuring nicotine de- pendence: A review of the Fagerstrom Tolerance Questionnaire. Jour- nal ofBehavioral Medicine. 12, 159-182. Fiore. M. C., Novotny, T. E., Pierce. J. P.. Giovino. G. A., Hatziandreu, E. J., Newcomb, P. A., Surawicz. T. S., & Davis. R. M. (1990). Meth- ods used to quit smoking in the United States. Journal of the Ameri- can Medical Associatian, 263. 2760-2765. Fiore, M. C.. Pierce. J. P., Remittgton, P. L., & Fiore, B. J. (1990). Cig- arette smoking: The clinician's role in cessation, prevention, and pub- lic health. Disease-a-Month. 35. 186-249. Fullet: R. K., Branchey, L., Brightwell, D., Derman, R. M., Emrick, C. D., Ibes: F. L., Jamea K. E., Lacoutsiere, R. B., Lee, K. K., LoJ%D wettstam, I., Maany, I., Neiderhiser D., Nocks, J. J., & Shawc S. = (1986). Disul6ram treatment of alcoholism: A veterans adminisua-kQb tion cooperative study. Journal of the American Medica! Associatiat, CTJ 256. 1449-1455. t~ General Accounting Office. (1990). Report to the Chairman. Selectw Committee on narcotics abuse and control. House of Representatives:= IND

I I I I I I I I I I I I I I I 758 JOH:V R. HUGHES ,tlethadone maintenance. (Washington. DC: U.S. Government Print- ing Office. Glassman, A. H., & Covey, L. S. (1990). Future trends in the pharma- cological treatment of smoking cessation. Drugs. 40, 1-5. Glassman. A. H.. Covey. L. S. & Stetner. F. (1989. May). Srnoking ces- satton. depressron, and antidepressants. Paper presented at the 142nd annual meeting of the American Psychiatric Association. Glassman. A. H.. Helzer. J. E.. Covey, L. S.. Cottlen L. B.. Stetner. F.. Tipp, J. E.. & Johnson. J. (1990). Smoking, smoking cessation. and major depresston. Journal ofthe Amertcan Medical Association. 264. 1546-1549. Glassman. A. H.. Jackson, W. K.. Walsh. B. T., & Roose, S. P. (1985). Cigarette craving, smoking withdrawal, and clonidine. Science, 226, 864-866. Glassman. A. H.. Stetner. F., Walsh. B. T., & Covi, A. (1988). Heavy smokers, smoking cessation, and clonidine: Results of a double-blind, randomized trial. Journal of the American Medical Association. 259. 2863-2866, Goldstein. M. G.. Niaura, R.. & Abrams. D. B. (1991). Pharmacologi- cal and behavioral treatment of nicotine dependence: Nicotine as a drug of abuse. In A. Stoudemire & B. S. Fogel (Eds.). Medical-Psy- chratrtc practice (pp. 541-596). Washington, DC: American Psychi- atric Press. Goldstein. M. G.. Niaura. R.. Follick. M. J.. & Abrams. D. B. (1989). Effects of behavioral skills training and schedule of nicotine gum ad- ministration on smoking cessation. Amerrcan Journal of Ps,trhiatry, 146. 56-60. Gorelick. D. A.. Rose. J. E.. & Jarvik. M. E. (1989). Effect of naloxone on cigarette smoking. Journal ofSubstance Abuse. 1. 153-159. Gourlay. S. G.. & McNeil. J. J. (1990). Antismoking products. Medical Journal o(Atutral ia. 153. 699-707. Gritz E. R.. Klesges, R. C.. & Meyers. A. W. (1989). The smoking and body weight relationship: Implications for intervention and post-ces- sation weight control. Annals of Behavioral Medicine, 11, 144-153. Gross. J,. Stitzer. M. L.. & Maldonado. J. (1989). Nicotine replacement: Effects on postcessauon weight gain. Journal of Consulting and Clin- ical PstrholnAt: 5 7. 87-92. Grunberg. N. E. (1990). The inverse relationship between tobacco use and body weight. In L. T Kozlowski. H. Annis. H. D. Cappell. F. B. Glaser. M. Goodstadt. Y. Israel, H. Kalant. E. M. Sellers. & E. R. Vingilis ( Eds. ), Research advances in alcohol and drug problems (pp. 273), New tbrk: Plenum Press. Hajek. P.. Jarvis. M. J.. Belcher. M.. Sutherland G.. & Feyerabend. C. (1989). Effect of smoke-free cigarettes on 24 h cigarette withdrawal: A double-blind placebo-controlled study. Ps}rhopharmacology: 97, 99-10:. Hall. S. M.. Mueoz. R. F., Retu, V. L. and Sees. K. L. (1993). Nicotine, negative affect, and depression. Journal of Coresulting and Clinical Pstrhrrhgt: 6/. 761-767. Hall. S. M.. TunstalL C.. Gin3berg. D.. Benowitz, N.. & Jones, R. T. (1987). Nicotine gum and behavioral treatment: A placebo-con- trolled trial. Journal of Carsulttrg and Clinical Psyrholog): 55. 603- 605. Henningfield. J. E.. & Keenan. R. M. (1993). Nicotine delivery kinetics and abuse Itab+hty. Journal ofCaasultutRand Clinical Ps)rhology, 61, 743-750. Henning5eld. J. E.. & Woodson. P. P. (1989). Behavioral arsd physiolog- ical aspects of nicotine dependence: The role of nicotine dose. Jour- nal ofSubuance.dbuse. 1. 301-317. Hollon. S. D.. & Beck, A. T. (1987). Psychotherapy and drug therapy: Comparisons and combinations In S. L. Garfield & Bergin (Eds.), Handbook ol pstrhotherajn• and heha+•irn change (pp. 43'-490). New York: Wiley. Hughes. J. R. (1986). Problems of nicotine gum. In J. K- Ockene (Ed.), Pharmacologic treatment oJ'tobacco dependence: Proceedings nfthe kbrld Congress, .ti"ovember 3-5-. 1985 (pp. 141-147). Cambridge. MA: Institute for the Study of Smoking Behavior and Policy. Hughes. J. R. (1989). Dependence potential and abuse liability of nico- tine replacement therapies. Biomedicine and Pharmacotherapr. 43. 11-17. Hughes. J. R. (1991). Combining psychological and pharmacological treatment for smoking. Journal ofSubstanceAbuse. 3. 337-350. Hughes, J. R. (1992). The nicotine withdrawal syndrome: A brief review and update. International Journal of Smoking Cessation, 1. 21-26. Hughes, J. R. (1993a). Nicotine dependence in smokers with and with- out a past history of alcohol/drug problems. Journal of Substance Abuse Treatment. 10. 181-187. Hughes, J. R. (1993b). Risk/benefit of nicotine replacement in smoking cessation. Drug Safety: 8, 49-56. Hughes, J. R.. & Glaser, M. (1993). Transdermal nicotine for smoking cessation. Health Values. 17, 24-31. Hughes, J. R., Gulliver, S. B.. Amori, G., Mireault, G., & Fenwick, J. (1989). Effect of instructions and nicotine on smoking cessation, withdrawal symptoms and self-administration of nicotine gum. Psy- chopharmacology, 99. 486f391. Hughes, J. R., Guat. S. W., Keenan. R. M., Fenwick, J. W., & Healy, M. L. (1989). Nicotine vs placebo gum in general practice. Journal of the American Medical Association,161. 1300-1305. Hughes, J. R.. Gust. S. W.. Keenan. R. M.. Skoog, K., Fenwick, J. W.. & Higgins. S. T. (1991). Long-term use of nicotine versus placebo gum. Archives oflnternal Medicine. 151. 1993-1998. Hughes, J. R.. Gust, S. W., Skoog, K.. Keenan, R.. & Fenwick, J. W. (1991). Symptoms of tobacco withdrawal: A replication and exten- sion. Archives of General Psyrhiatry, 48, 52-59. Hughes, J. R., Higgins, S. T., & Hatsukami, D. K. (1990). ER'ects of abstinence from tobacco: A critical review. In L. T. Kozlowski. H. Annis, H. D. Cappell, F. Giase>: M. Goodstadt, Y. Israel, H. Kalant, E. M. Sellers, & J. Vingilis (Eds.), Research advances in alcohol and drug problems (Vol. 10, pp. 317-398). New York: Plenum Press. Hughes, J. R.. & Krahn, D. (1985). Blindness and the validity of double- blind drug trials. Journal of Clinical Psyrhopharmacology, 5, 138- 142. Hughes, J. R.. Wadland, W. C., Fenwick. J. W., Lewis. J., & Bickel, W. K. (1991). Effect of cost on the self-administration and efficacy of nicotine gum: A preliminary study. Preventive Medicine. 20, 486- 496. Jacobs. M. A., Spiker, A. Z., Norman, M. M., Wohlbetg, G. W., & Knapp, P. H. (1971). Interaction of personality and treatment condi- tions associated with success in a smoking control program. Psyrho- sanatic Medicine, 6. 545-556. Jamrozik, K., Vessey, M., & Wald, N. (1984). Placebo-controlled trial of nicotine chewing gum in general practice. British Medical Jourrtal, 289. 794-797. Jarvik, M. E.. & Henningfield, J. E. (1988). Pharmacological treatment of tobacco dependence. Pharmacology, Biochemistry & Behavior, 30, 279-294. Jarvis, M. J., Hajek, P., Russell, M. A. H., & West. R. J. (1987). Nasal nicotine solution as an aid to cigarette withdrawal: A pilot clinical trial. British Journal ofAddiction, 82, 983-988. Jensen, E. J., Schmidt, E., Pedersen, B., & Dahl, R. (1991). Effect on smoking cessation of silver acetate, nicotine and ordinary chewing gum. Psyrhopharmacology, 104, 470-474. Karras, A., & Kane, J. (1980). Naloxone reduces cigarette smoking. Life Sciences. 27. 1541-1545. Killen, J. D., Fortatann, S. P., Newman, B., & Varady, A. (1990). Evat- uation of a treatment approach combining nicotine gum with self- guided behavioral treatments for smoking relapse prevention. Jour- nal ofCoasulting and Clinical Psyrhology, 58, 85-92.

I I I I I I I I I I I I SPECIAL SECTION: PHARMACOTHERAPY FOR SMOVING CESSATION Killen, J. D., Fotzmann, S. P., Telch. M. J., & tiewTttan, B. (1988). Are heavy smokers different from light smokers? A comparison after 48 hours without cigarettes. Journal of the American Medical Associa- tion. 260, 1581-1585. Klesges. R. C.. Klesges, L. M.. Meyers, A. W.. Klem, M. L.. & Isbell, T. ( 1990). The effects of phenyipropanolamine on dietarv intake, physi- cal activity, and body weight after smoking cessation. Clinical Phar- macology and Therapeutics, 47, 747-754. Klesges, R. C.. Meyers, A. W., Winders, S. E., & French, S. N. (1990). Determining the reasons for weight gain following smoking cessation: Current findings, methodological issues, and future directions for re- search. Annals of Behaviora/ Medicine, 11, 134-143. Kozlowski, L. T(1984). Pharmacological approaches to smoking mod- ification. In J. D. Matarazzo, S. M. Weiss, J. A. Herd. & N. E. Miller (Eds.), Behavioral health: A handbook othealth enhancement and dis- ease prevention (pp. 713-728). New York: Wiley. Lam, W. L., Sze. P. C.. Sacks. H. S., & Chalmer, T C. (1987). Meta- analysis of randomized controlled trials of nicotine chewing gum. Lancet, 2, 27-29. Leischow. S. J., Sachs. D. P. L., Bostrom. A. G., & Hansen, M. D. (1992). Effects of differing nicotine replacement doses on weight gain after smoking cessation. Archives of Family Medicine. 1. 233-237. Levin. E. D., Behm, F. M., & Rose. J. E. (1990). The use of flavor in cigarette substitutes. Drug and Alcohol Dependence, 26, 155-160. Levin. E. D.. Rose. J. E.. & Behm. F. M. (1990). Development of a citric acid aerosol as a smoking cessation aid. Drug and Alcohol Depen- dence. 25, 273-279. Lichtenstein, E.. & Hollis, J. E(1992). Patient referral to a smoking cessation program: Who follows through? Journal of Familv Practice. 34. 739-744. Low. R. B.. Jones, M., Carter. B., & Cadoret, R. J. (1984). The effect of d-amphetamine and ephedrine on smoking attitude and behavior. Addictive Behaviors, 9. 335-345. Meier-Lammermann, E.. Meyer, M., & Boleskei. P. L. (1990). Differ- ences between transdermal nicotine application and nicotine chew- ing gum in combination with behavioural group therapy in smoking cessation. European RespiratoryJournal, 3(Suppl. 10), 1685. Nemeth-Cosiett, R., & Griffiths, R. R. (1986). Naloxone does not affect cigarette smoking. Ps.trhopharmacologyt 89, 261-264. Nemeth-Coslett, R.. & Henningfield. J. E. (1986). Effects of nicotine chewing gum on cigarette smoking and subjective and physiologic effects. Clinical Pharmacology and Therapeutics. 39, 625-630. Newhouse, P.. & Hughes. J. R. (1991). The roie of nicotine and nicotine mechanisms in neuropsychiatric disease. British Journal of Addic- tion, 86. 521-526. Niaura, R. S.. Rohsenow, D. J.. Binkoff, J. A.. Monti. P. M.. Pedrara, M.. & Abrams, D. B. (1988). Relevance of cue reactivity to under- standing alcohol and smoking relapse. Journal ofAbnormal Psyrhol- ogy, 97, 133-152. Nunn-Thompson. C. L. & Simon, P. A. (1989). Pharmacotherapy for smoking cessation. Clinical Pharmacy, 8, 710-720. Orleans, C. T., Kristellet; J. L., & Gritz, E. R. (1993), Helping hospital- ized smokers quit: New directions for treatment and research. Jour- nal of Covtsulting and Clinical Psychology, 61. 778-789. Oster, G., Huse. D. M., Delea, T, E., & Colditz, G. A. (1986). Cost- effectiveness of nicotine gum as an adjunct to physicians' advice against cigarette smoking. Journal of the American Medical Associa- tion, 256. 1315-1318. Palmer K. J., & Faulds, D. (1992). Transdermal nicotine: A review of its pharmacodynamic and pharmacokinetic properties, and therapeutic use as an aid to smoking cessation. Drugs, 44, 498-529. Payne, T. J., Etscheidt. M., & Corrigan. S. A. (1990). Conditioning ar- bitrary stimuli to cigarette smoke intake: A preliminary study. Jour- nal ofSubstanceAbuse, 2, 113-119. 759 Payne. T J.. Schare. M. L.. Levis, D. J.. & Colletti. G. (19911. Exposure to smoking-relevant cues: Effects on desire to smoke and topographi- cal components of smoking behavior. Addictive Behaviors. 16, 467- 479, Perkins. K. A. (1993). Weight gain following smoking cessation. Journal o,! Consulttng and Clinical Pstrhologt: 61. 768-777. Perkins. K. A.. Grobe. J. E.. Stiller, R. L.. Fonte. C., & Goettler. J. E. (1992). Nasal spray nicotine replacement suppresses cigarette smok- ing desire and behavior. Clinical Pharmacology and Therapeutics, 52. 627-634. Pierce, J. P., Fiore, M. C.. Novotny, T. E.. Hatziandreu. E. J., & Davis, R. M. (1989). Trends in cigarette smoking in the United States. Jour- nal oftheAmerican Medical Association, 261, 61-65. Pomerieau, O. F.. & Pomerleau, C. F. (1984). Neuroregulators and the reinforcement of smoking: Towards a biobehavioral explanation. Neuroscience and Biobehavior Reviews. 8. 503-513. Prignot, J. (1989). Pharmacological approach to smoking cessation. European Respi rator y Journal. 2. 5 50-560. Prochazka, A. V., Petty, T. L., Nett, L-. Silvers, G. W., Sachs, D. P. L.. Rennard. S. I., Daughton, D. M., Grimm. R. H.. Jr., & Heim, C. (1992). Transdermal clonidine reduced some withdrawal symptoms but did not increase smoking cessation. Archives of Internal Medi- cine, 152, 2065-2070. Rose, J. E., & Behm. F. M, (1987). Refined cigarette smoke as a method for reducing nicotine intake. Pharmacology, Biochemistry & Behav- ior. 28, 305-3 10. Rose, J. E., Herskovic, J. E.. Trilling, Y., & Jarvik, M. E. (1985). Trans- dermal nicotine reduces cigarette craving and nicotine preference. Clinical Pharmacology and Therapeutics. 38. 450-456. Rose. J. E.. & Hickman. C. S. (1987). Citric acid aerosol as a potential smoking cessation aid. Chest, 92, 1005-1008. A Sachs, D. P. L. (1991 a). Individual patient differences: Dosing strategies and time course considerations. In J. E. Henningfield & M. L. Stitzer (Eds.), New Developrnents in Nicotine-Delivery SyStenu.• Proceedings of a Conference, John Hopkins University, September 24. 1990 (pp. 53-61). Ossining, NY: Cortlandt Communications. Sachs, D. P. L. (1991 b). Advances in smoking cessation treatment. Cur- rent Pulmonology, 12, 139-198. Sachs, D. P. L„ & Leischow, S. J. (1991). Pharmacological approaches to smoking cessation. Clinics in Chest Medicine, 12, 9-10. Sachs. D. P. L.. Sawe, U., & Leischow. S. J. (1992). Nicotine transda- mai patch, smoking cessation and gender. In L. S. Harris (Ed.), Prob- lems of drug dependence, 1991. NIDA Research Monograph 119 (pp. 367). Washington, DC: U.S. Department of Health and Human Ser- vices. Schwartu, J. L. (1992). Methods of smoking casation. In M. C. Fiore (Ed.), The medical clinics ofNorth America (pp. 451-476). Philadel- phia: W. B. Saunders Company. Schwartz, J. L., & Dubitsky, M. (1968). One-year follow-up results of a smoking cessation program. Canadian Journal of Public Health. 59, 161-165. Seigel, A. (1983). Classical conditioning, drug tolerance, and drug de- pendence. In R. G. Smart, Y. Israel. F. B. Glaser, H. Kalant, R. E. Popham, W. Schmidt, & R. G. Smart (Eds,), Research advances in alcohol and drug dependence, Vol. 7(pp, 207-246). New York: Ple- num Press. Sellers, E. M., Naranjo, C. A., & Kadlec, K. (1987). Do serotonin up- take inhibitors decrease smoking? Observations in a group of heavy drinkers-Journal ofClinical Psyrhapharmacology, 7,4 17-420. Shiffman, S. (1993). Assessing smoking patterns and motives. Journal of Cmsulting aai Clinical Psyrhology. 61, 718-728. Spring, B„ Wurtman, J., Gleason, R, Wurtman, R.. & Kesslet; K. (1991). Weight gain and withdrawal symptoms after smoking cessa-

I JOHN R. HUGHES I I 1 I I I I I tion: A preventive intef~vention using d-fent)uramine. Health Pssrhol- ogti: 10. 216-223. Stolerman. 1. P. (1986). Could nicotine antagonists be used in smoking cessation? British Journal ofAddicuovs, 81, 47-53. Sutherland. G.. Stapleton. J. A.. Russell. M. A. H.. Jarvis. M. J.. Haiek. P.. Beicher. M.. & Feverabend. C. (1992). Randomised con- trolled trial of nasal ntcotine spray in smoking cessation. Lancet. 340. 324-329. Tonnesen. P.. Norregaard. J.. Simonsen. K.. & Sawe. U. (1991). A dou- ble-biind trial of a 16-hour transdermal nicotine patch in smoking cessation. ti'ew England Journal of Medicine. 325, 311-315. Transdermal Nicotine Study Group. (1991). Transdermal nicotine for smoking cessation: Results of two multicenter controlled trials. Jour- nal ofthe American ,•Nedical Association, 266. 3133-3 I38. U.S. Department of Health and Human Services. (1988). Trends in to- bacco use in the United States: Appendix A. In N. L. Benowitz, N. Grunbetg J. E. Henningfield. & H. L. Lando (Ed.), The health conse- quences of smoking: Nicotine addiction. .4 report of the surgeon general (pp• 561-588). Washington. DC: U.S. Government Printing Office. U.S. Department of Health and Human Services (1990). Psychological and behavioral consequences and correiates of smoking cessation. In J. M. Samiet. R. M. Davis, N. E. Grunberg, J. K Ockene, D. B. Petitti. & W. C. Willett. (Ed.). Health benefits ofsmokingcessatron: A report ofthe L`.S. Surgeon General (pp. 521-531). Washington DC: U.S. Government Printing Office. West. R. J. (1992). The "nicotine replacement paradox" in smoking cessation: How does nicotine gum really work? British Journal ofAd- diction.87, 165-167. West. R. J.. Hajek. P.. & McNeill. A. (1991). Effect of buspirone on cigarette withdrawal symptoms and short-term abstinence rates in a smokers clinic. Ps)chopharmacology, 104, 91-96. West, R. J.. & Schneider, N. (1987). Craving for cigarettes. British Jour- nal ofAddiction, 82, 407-4I 5. Received July 29, 1992 Revision received December 3, 1992 Accepted February 8, 1993