Tobacco Documents Online

0 C0 0 O]

PROCEEDI~IGS OF THE FIFTH ~DRLD CONFERENCE ON SMOKING AND HEALTH WIBNIPEG~ CARADA~ 1983 OEGANIZERS Canadian Council on Smoking and Health SPONSORS The Canadian Cancer Society The Canadian Heart Foundation The Canadian Lung Association T108350675

© Canadian Council on Smoking and Health ISBN 0-969-2331-0-8 ISBN 0-969-2331-I-6 TI03350676

PROCEEDIHGS OF TRE FIFTH WORLD CONFERENCE ON ~MOKING AND HEALTH Edited by William F. Forbes, Ph.D., D.Sc. Richard C. Frecker, M.D., Ph.D. David Nostbakken, Ph.D. Canadian Coun=il on Smoking and Health 725 Churchill Avenue Ottawa, Ontario, Canada KIZ 5G7 With contributions from Canadian Cancer Society Canadian Heart Foundation Canadian Lung Association Health and Welfare Canada Merrell-Dow Pharmaceuticals (Canada) Inc. TI0~,350677

The 5th World Conference on Smoking and Health was held July I0-15, 1983 at Winnipeg, Manitoba, Canada. There were about i,i00 participants from almost 80 countries. A full list of participants can be obtained from the Canadian Council on Smoking and Health, 725 Churchill Avenue, Ottawa, Ontario KIZ 5G7, Canada. This publication contains ~ost of the invited papers and some of the other papers which were presented at the Conference. This forms Volume 1 of the Proceedings. It is planned to produce a second volume which will contain many of the remaining papers. We regret that all the papers could not be published at the same time but, for a variety of reasons, which arose because the authors are scattered throughout the world and other delays~ and also because of the large number of papers and the desirability of publishing at least some of them as soon as possible, the decision was made to publish the Proceedings in two volumes. Abstracts of the contributed papers that were presented at the Conference, have already been published in the book entitled, "ABSTRACTS, Fifth World Conference on Smoking and Health, Winnipeg, Canada, July 10-15th, 1983", which was given to each participant in the Conference. A limited number of abstract books are still available from the office of the Canadian Council on Smoking and Health. It is hoped that the two volumes will represent a comprehensive review of current knowledge in all fields related to smoking and health. The Editors would also llke to express their gratitude to the various reviewers (Dr. L. Kozlowski, Dr. R. Lauzon, Dr. L. Pederson, Dr. G. Piper and Ms. P. Zipchen), and particularly to Dr. J.A. Jackson for editorial assistance, and to Miss Redi~a Caracaz for very competent word processing which greatly facilitated the production of this volume. William F. Forbes, Ph.D, D.Sc. Richard C. Frecker, M.D., Ph.D. David Nostbakken, Ph.D. TI08350678

iii TABLE OF C~TS INI~ODUC'EION ............................... i PLENARY ADDRESSES Summary of Report of WHO Expert Couznlttee on Smoking Control Strategies in Developing Countries ......................... 1 Abdul Rahman AI-Awadi S~oking and Wo~en ............................ 7 Mary Jane Ashley Address to the Fifth World Conference ou Snaking and Health ..... Monique BEgin .25 Legislation and Political Activity ................ Kjell Bjartveit • . 31 Public Information Programmes ...................... 47 Mike Daube Smoking in Developing Countries ..................... 59 D. Femi-Pearse Presentation to the Fifth World Conference on Smoking and Health .... 71 George Godber The Social and Economic Implications of Tobacco Use ........... 77 Nigel Gray Introductory Remarks at the Opening Plenary Session ........... 93 Samuel Hynd Message to the Fifth World Conference on S~oking and Health ....... 95 The Role of ~EO in S~oking Control ................... 99 R. Masironi Tl08350879

HEALTH CONSEQUENCES The Risk of Smoking During Pregnancy. Chromatiu Changes in Cells of U~bilical Arteries from Newborns Delivered by Mothers Smoking more than I0 Cigarettes Per Day ....................... I03 Inger Asmussen Pharmacokinetlc Considerations in Understanding the Effects of Cigarette Smoking and Smoking Behavior ................. 107 Neal L. Benowitz Pregnancy and Patterns of Tobacco Use: Shifts and Direction ...... 117 Virginia Cresswell-Jones Eye Movement Measurement and the Pharmacodynaudcs of Tobacco Dependence ........................... 121 R.C. Frecker The Interaction of Pbarmacologlcal and Psychological Determinants of Tobacco Use ............................. 131 Seymore Herling and Lynn T. Kozlowskl Passive Smoking and Lung Cancer, Nasal Sinus Cancer, Brain Tumor and Ischendc Heart Disease ....................... 137 Takesh~ Hirayama Influence of Smoking on the Health State of Working Wo~en in Slovakia .............................. 143 Leona Huba~ov~, Miloslav Huba~, Franti~ek Strelka and Imrich Borsk~ Health Consequences of Smoking in two Socioeconomic Classes in Bombay, India ........................-- 149 Cigarette Smoking and Excess Mortallty in Shanghai, China. Evidence from a Prospective Study ................... 157 Li Wan-Xian The Life-Expectancy of Non-Smoking_ Men and Women ............ 165 G.H. Miller and D.R. Gerstein T108350680

TABLE V Smoking and Various Physical C~plaints ................ 171 Hiroshi Ogawa, Suketami Tominaga and Kunio Aoki The Effect of Cigarette-Smoklng on the 2.414 l~Run of 155 Adult Males in Singapore ........................... 177 Teck Chin, Ong and Pui Yong, Tan Tobacco Addiction ~nd Other Drug Abuse Among American Youth ...... 183 R.T. Ravenholt and William Pollin The Dosimetry of Passive Smoking .................... 191 James L. Repace Passive Smoking ~nd the Lungs: A Eeview of Effects other than Malignancy ............... 199 Jonathan M. Samet and Frank E. Speizer Effect of Smoking on Gastrointestinal Hormone Secretion ........ 207 Yutaka Seino, Kinsuke Tsuda, Kozaburo Mori, Shozo Li, Jiro Takemura, Shigeru Matsukura and Hiroo Imura Irritating and Annoying Effects of Cigarette Smoke ........... 213 Roy J. Shephard Smoking, Physical Activity and Health: Findings from the Canada Fitness Survey ................ 217 Thomas Stephens and Linda Pederson Smoking and Subarachnold Haemorrhage .................. 225 A. Taha, K.P. Ball and R.D. Illingworth The Smokers' Dependence on Nicotine and the Fate of Nicotine During Tobacco Curing and Smoking and its Reduction .......... 229 T.C. Tso, J.D. Adams, N.J. Haley and D. Hoffman Lung Cancer Risk and Tar Yields of Cigarettes Smoked ........ Christian Vutuc and Brigitte Gredler .239 TI08350681

TABLE CF ~Oh'TENTS III: PUBLIC EDUCAT10N AND INFORMATION Saoking~umng Children ...... Theodor Abelin ................... 243 Tobacco and the Indian g~msn ...................... 255 Mira B. Aghi Mortality and Morbidity from Smoking-Related Diseases in Parliamentary Constituencies in Scotland - a New Method of Presentation of Data ..................... 259 K.G. Brotherston and'E~ ~r~f;on A Randomlsed-Controlled Trial of Education for Prevention of Smoking in 12-Year Old Children .................... 263 Deborah A. Fisher, Bruce K. Armstrong and Nicholas H. de Klerk The 'Beliefs' and 'activities' of GPs and Health Visitors About Anti-Smsklng Education ...................... 271 Godfrey Fowler and Conrad Jamrozik Monigo. Programme d'interventlon sur le Tabagisme Pour les Enfants de 5 ~ 12 Ann ..................... 281 Jocelyne Gauthier The Tobacco Indus~ry's Response to Scientific Evidence on Involuntary Sm~klng ......................... 287 Stanton A. Glantz Smoking Education for Teenagers .................... 293 E.M. Gray, P. Gammage, M.J. Morgan, and J.R. Eiser The Challenge to Public Education: A Rulti-National Perspective .... 301 Ellen R. Gri~z The Mass Media in Health Education. The Need for Audience Involvement ............ G.B. Hastings and D.S. Leathar • 311 Advertising and Women's Changing Smoking_Habi.~s; A Hi~6~ai Perspectxve ..... ................... 319 Holly L. Howe T108350682

TABLE OF C~h~S vii Creative Uses of Co=~unications ~dia in S=oklng ............ 331 Michael Pertschuk The Experts Meet the Media ....................... 337 Uma Ram Nath Public Education Programs ....... Lars M. RamstrSm ............... 343 Smoking Education in the United Kingdom with Special Reference to England, Wales, and Northern Ireland ................ 355 Donald Reid, John Harris, Mich~al Jacob, Alan Maryon Davis, and Jane Randell The Role of Legislation, Health Education and Social Organizations of a Socialist Country in the Battle against Smoking .......... 361 Adam Tahy Anti-Tobacco Educational Film in the Light of Receivers' Opinion: 13-15 Year Old Youths 367 Stanislaw Wijatkowski IV: CESSATION A ~ndomised Trial of Three Different Anti-Smoking Interventions in General Practice .......................... 371 K.D. Jamrozik, M.P. Vessey, N.J. Wald, and GoH. Fowler The Use of Nicotine Chewing Gum in a Smokers' Clinic .......... 377 M.J. Jarvis l~nediate and Delayed Effects of Postal Advice on Stopping Smoking . Frank Ledwith •383 Smoking Intervention in Pregnancy ................... 389 Barbara Keely Loeb, Gunnar Waage, and Jeffry Bailey -What Quitters Need to Know ....................... 397 Wendy J. Moreton and Robert East TI03350683

viii TABLE S~okin~ Behavior Change Patterns in the Nmltiple Risk Factor Intervention Trial (HRFIT) and their Relatienship to Hortality £rom Coronary Heart Disease (CHD) ................... 403 Judith K. 0ckene, Stephen B. Hulley, and Terance A. Gerace Early Prophylaxis of Smoking ...................... 415 Stanimir G. Penev and Luben G. Penev A Comparison of Behavlour Hodlfication, Health Education and Eypnosis Prograu~es for Cigarette SmoklngCessation: A Random~zed Clinical Trlal ............................. 419 Simon W. Rabkin, Evelyn Boyko, Fred Shane~ and Joseph Kaufert The Role of Chest Physicians as S~oking Cessation Counsellors ..... 427 Martin Raw and James Friend Smokers' Clinics in Britain. A Descriptive Survey ........... 433 Martin Raw and Julia Heller Hinimal Anti-Smoking Intervention by Physiclans and its Enhancement by Nicotlne Chewing Gum .................. 439 M.A.H. Russell, R. Merriman, and A.R. Edwards BUTT OUT! Evaluation of The Canadian Armed Forces Smoking Cessation Program ....................... 445 Ronald P. Schlegel, Stephen R. Manske, and Michael E. Shannon Nicotine C~nn in Smoking Cessation: Outcome and Withdrawal in a Placebo-Controlled Trial ........................ 453 Nina G. Schneider and Murray E. Jarvik Developments in Smoking Cessation: Trends and Observations ....... 461 Jerome L. Schwartz ~aternal Cigarette Smoking in Nova Scotia .............. 473 Kenneth E. Scott, Barbara English, and Yola~de Samson The Socio-economic l~plications of Smoking and the Non-Smokers' Rights Hovement ............................ 477 John F. Banzhaf III TI08350684

Taxation and Cigarette S~oklng in the United States .......... 483 Michael Grossman Smoking and Health Care Costs: Plus or Minus? ............. 489 Robert E. Leu and Thomas Schau5 Taxation as a Means o£ Affecting Tobacco Usage ............. 495 M.E. Thompson and W.F. Forbes Socio-econom~c and Cultural Implications of Health Interventions: The Case of Smoking iu Ethiopia .................... 505 K. Yayehyirad VI: LEGISLATION Legal Action on Smoking and Health ................... 513 John F. Banzhaf III and Paul N. Pfeiffer Sidestream Smoke: A Mainstream Health Problem - the Arizona Response ........................... 519 The Legal Implications of Cigarette Smoking .............. 523 Donald W. Garner Lignes Directrlces pour la Progran,aation d'Initiatives Legislatives, d'Informatlon et d'Educatlon Sanltaire pour l'Action contre le Tabagisme en Italle .......................... 533 L. Giannico The Tobacco Iudustry's Polltical Tactics in California since 1978 . . . 539 Stanton A. Glantz and Paul L. Loveday Strategy of the Tobacco Industry Concerning Legislation on Tobacco Advertising in some Western Europeau Countries ............. 549 Luc Joossens Legi~latlon~nd-Political-Act~vity--in-Sri--Lank~-i~the~Field_ of Smoking and Health ......................... 555 W.A. Karunaratne TI08350685

The Urgent Need to Cont¢ol the Smok~ug Epidesaic in the Third World . . .§61 Martin Khor Kok Pens The Tobacco Industry and the Ban o~ Advertising ............ 567 ~sbjorn Kjonstad Anti klngLegi 1 tl " Egypt 575 --~mo 8 a O~ ~Jl ..... • ....... • • • * ° ° Sherif Omar a~ H~i E1 Sayed Smoking-Control in Developiug Countries - Support by WHO and SIDA (S~edish Intern~tlo~al Development Authority) ........... 581 Lars M. Ramstrom" " ~ Experience in aCountry without Smoking Control ............ 589 Mario Rigatto Combatin~ the Smoking Eplde~c: Why Legislation? ............ 593 Ruth Roemer Non-Smokers' Health R~ghts from 1976 to Today: A Plaintiff's Perspective ................ Donna M. Shimp ...... 603 Achieving a Non-smoking Environment: The Vital Role of Individual Actions ........................... 60~ Donna M. Shlmp Countering the Opposition ....................... 615 David Simpson Smoking in Malaysia - Promotion and Control .............. 623 S.K. Teoh Tobacco Advertising in Develop[ng Countries - Experience in Papua New Cuinea ........................... Martin ToVadek and Konrad Jamrozlk .629 Campaigning for Legislatiou to Ban Tobacco Promotion: A Case Study .............................. 635 Stephen W. Woodward TI083506,2.,6

VII: SB~EYSA~T~09S Longitudinal Prediction of the Onset and Change of Adolescent Smoking ......................... 641 Dennis V. Ary, Anthony Biglan, Cheri L. Gallison, Wendy Weissmann, and Herbert H. Severson The Waterloo Smoking Prevention Project: Nethodological Advances, Results, and E~pirical Guidelines for Disse~nation to the Schools . . .649 J. Allan Best, Katherine B. Ryan, K. Stephen Brown, Shelagh. M.J. Towson, and Brian R. Flay Priorities for Social Science Research on Smoking. Report o£ the Fifth World Conference Working Croup on Social Science and Program Related Research ...................... 657 Is Cigarette Consumption Declining in Canada? ............. 667 Neil E. Colllshaw Group Effects in Smoking Research: Statistical Considerations ..... 679 Annette J. Dobson and Gregory R. Hardes Smoking and Women's Emancipation; the Developed World ......... 687 Ingrid Eide Smoking Habits am~ug Registered Nurses, Auxiliary Nurses and Nurse Aides ............................ 693 Steinar Yolger~, lngerma Brofoss, Per Morten L~chsen, and Kjell Bjartve~t Quantitative Models of Lung Cancer Nortallty for the United Kingdom, Canada and Australia ................. 701 R.W. Gibberd, E. Doyle, K.S. Brown, and W.F. Forbes Prediction of Lung Cancer Incidence in Finland: Appraisal of Different Approaches ................... 707 Timo R. Hakulinen, Eero I. Pukkala, and Esa M. L~r~ Planning, Development and Evaluation o~ a Special Smoking and Health Pro.gramme for Pup_il's Aged_l-2r/~.~ear~ . ~ ~ ~ -_ _- ~ ~ ~ ~ . .7/1 A. Hauknes, P.M. L~chsen, and L.E. Aar4 TI08350687

xii TA.~LE O~ ~ Changes in S~oking Habits Between the 1976 and 1981 New Zealand Population Censuses .................... 725 D.R. Hay APropos de Quelques Aspects Soclo-culturels du Tabagisme en Alg~rie ............ M. Khellaf and B. Bensmail ............. 733 SmokingAmong Alaska Native Youth ................... 737 John F. Lee A Decade of Smoking ~ong High School Students in Hobart Tasmania " " 743 fan C. Lewis, Kent J. Rayner, and Klaus M. Schwarzenholz Comparisons Between the United States and Western Europe in the Health Effects of Smoking and Related Factors .......... 751 Cornelius J. Lynch The Social Context of Smoking During Adolescence ............ 757 Michael Murray A Study of Adolescent Smoking in the Greater Dublin Area ....... .;763 Desmond J. O'Byrne Smoking Patterns of Students in Higher Institutions of Learning in Nigeria ................ . . . . . . . . .773 B.O. Onadeko, A.A. Awotedu, and M.O. Onadeko Smoking Trends in Thunder Bay, Ontario High School Students Richard S. Stanwick, Yarn Sawatzky, and David A. Legge • 781 The S~oking Habits of Native Canadians ................. 785 Margaret P. Thomson Smoking Behavlour in the Netherlands from 1958-1982 .......... 789 Jan van Reek T[08350688

CLOSING ~DRESS Smoking or Health Activities: Lessons from the Past, ~mplications of Present Experience~ Challenges for the Puture N.C. Delarue ........ 797 CONFERENCE RECOMMENDATIONS RECOMMENDATIONS THROUGH RAPPORTEURS Health Consequences ........................ 805 Public Education and Information .................. 807 Cessation .............................. 809 Economics .............................. 811 Legislation ............................. 8~3 RECOMMENDATIONS PROM DELEGATES .................... 815 RECOMMENDATIONS FROM THIRD WORLD DELEGATES ............... 819 TI08350689

PLENARY ADDRESSES T[083506£0

SUMMARY OF ~OET OF k~EO EXPERT COK~TTEE ON S]~KING CONTROL STRATEGIES IN DEVELOPING COUNTRIES GENEVA, 22-27 NOVEMBER 1982 The Honourable Abdul Rahman AI-Awadl Minister of Public Health Ministry of Public Health P.O. Box 5, Kuwait The WHO Expert Committee on Smoking Control Strategies in Developing Coun- tries met to draw up guidelines that could help governments and health institutions, in developing countries, in their efforts to stem the spread- ing of the tobacco smoking epidemic in the Third World. The tobacco indus- try is indeed exerting increasing pressure to expand its markets among un- suspecting populations in such countries, thus adding the burden of increas- ing rates of coronary heart disease and other smoking-related diseases to the existing problems of malnutrition and communicable diseases. WHO's first clear commitment to anti-smoking action came in 1970, when the World Health Assembly, in a resolution, called on governments to take action in the field of smoking control. Since 1974, two WHO Expert Committees have been convened to advise the Organization, the WHO International Clearing- house on Smoking and Health Information has been established, and WHO co- sponsored the Third, Fourth and Fifth World Conferences on Smoking and Health. The Organization also collaborated with FAO on a study of the eco- nomic benefits and social and medical costs of tobacco production and con- sumption. More recently, a WHO Action Programme on Smoking and Health was set up in accordance with resolution WHA33.35, adopted by the Health Assem- bly in 1980. Under this programme, the Organization has analysed the con- tent of cigarettes from developing countries, conducted educational semi- nars, collaborated with developing countries in the implementation of national smoking control programmes, and published numerous reports on smok- ing trends and related data. The most recent activity was the convening of a WHO Expert Committee on Smoking Control Strategies in Developing Coun- tries, whose main findings and recommendations are summarized here. The Expert Committee reviewed not only the harmful health effects of differ- ent types of tobacco use which characterise developing countries (i.e., cigarette smoking, bidis or hooka smoking, tobacco chewing, etc.), but also the adverse effects of tobacco use on the economy of the countries on account of smoking-related diseases and work absenteeism. It advised on the objectives of smoking control programmes, including data collection, educa- tion and information, legislation, smoking cessation, and the role of medical, political, social and religious leaders; the role of WHO, UN Agen- cies and NGOs; research on smoking behaviour; and evaluation of programme efficacy. The Committee provided guidance on how to counteract the tobacco industrx!~S_~ ~rguments, and provided several reco~me.nda.t~o.ns__f~r ac~on~ including collaboration at TCDC level. Tl08350691

2 AL-AWAD I More than one million people around the world still die prematurely every year because of cigarette smoking. In developed countries, it is generally understood that smoking causes lung cancer, coronary heart disease, chronic bronchitis, and other respiratory disorders, and major campaigns have been launched to reduce the rate of smoking. In developing countries, however, the situation is extremely serious because the public is not aware of the dangers to the same extent, nor are educational, legislative and other measures being taken to combat the smoking epidemic. These were sore of the conclusions reached by the recent WHO Expert Committee on Smoking Control Strategies in Developing Countries. The Committee called for firm steps to be taken now "to prevent the most unnecessary of modern epidemics", failing which the prospects for future generations are bleak. The Expert Committee noted that the tobacco-related diseases are on the rise in developing countries. The rate of lung cancer deaths has been increasing steadily in Hong Kong; rates for women there are now the highest in the world. In China, lung cancer cases doubled from 1963 to 1975 whereas mortality rates for cervical and uterine cancer decreased two fold in the same period. Other interesting points emerge from recent studies made in developing coun- tries. In India, it has been shown that filter and non-f~ILe~ cigarettes have identical tar and nicotine yield. Smoking filter-tips does not, there- fore, reduce the danger to Indian smokers. Many developing countries have cigarettes on sale with high yields of tar and nicotine. Compare a tar yield of 21-23 mg in India, China, and the Philippines with that of 0.5-20 mg in most industrial countries. Tobacco cultivation has spread to about 120 countries (63% of world produc- tion now comes from developing countries compared to 58% in 1972 and only 50% in 1962), thus becoming a substantial source of employment and creating new vested interests in the status quo. Overall, however, the costs demon- strably outweigh the "benefits". For example, tobacco taxes are "politically comfortable", that is, easy to administer and generally acceptable to smokers, thus commending themselves to many governments in developing countries. However, these taxes do not contribute to national wealth but merely redistribute wealth that is produced. They cannot, in any case, offset the economic losses caused by tobacco: health service expenditure on smoking-related diseases, disablement and work absenteeism, domestic and forest fires, use of scarce fuel to cure tobacco, and reduced food production. Indeed, while tobacco production has been growing in the developing world, per capita food consumption in many of these countries has remained stagnant or even declined. And, of course, malnourished people are particularly i11 equipped to withstand respiratory and other smoking-related diseases. Smokers have higher annual rates of medical care utilisation than non- smokers. In Canada, for instance, health care costs and losses in produc- tivity due to smoking together are estimated to account for at least "The Committee carefully examined the case sometimes presented to support tobacco cultivation and production and also evidence presented by FA0 on the T108350692

.~ry ly en r, le .1 O g role and scale of tobacco production ~,-~rldwide. It concluded that while action is justified primarily on health grounds,~ even economic arguments ulti~tely militate against tobacco production and cannot logically- and should not - be used to oppose implementation of s~oking control progr an-~es. The Committee expressed particular concern at reports that imported ciga- rettes sold and promoted in developing countries had been shown to yield much higher levels of tar and nicotine than similar brands (and even brands of the same name) sold in developed countries, including the countries of origin. Most of the work undertaken so far on emission products has been carried out with western-style cigarettes. There is, however, disturbing evidence that developing countries have been subjected to tobacco products with even greater toxicity than those available in the developed world. Recent evidence indicates also that some traditional forms of smoking, such as the bidi, are no less harmful. The Committee noted that cigarettes with high nicotine (and tar) yields may be more addictive than other brands; the promotion of such brands in developing countries is to be deplored, and governments should be alerted to the dangers they face in this area. Governments of developing countries are urged to take the necessary politic- al, social, and educational measures without delay. Systematic and contin- uous action, through a central agency or similar body, is necessary if long-term results are to be expected. Action against smoking can be inexpensive, yet effective, the experts point- ed out. Health warnings should be placed on cigarette packets. In fact the double standards by which cigarettes of the same brand carrying health warnings in developed countries are sold without these warnings and with much higher tar content in developing countries, should and can be ended by appropriate national legislation. In the short term, taxation of cigarettes can slow down consumption. Where no tobacco industry exists, developlng countries should be encouraged to maintain the status quo. Where such an industry does exist, efforts should be made to reduce its role in the national economy and to explore alternative uses for land and labour. The Committee recommended unequivocally that all tobacco advertising and sales promotion be prohibited. In the special case of most developing coun- tries, where literacy is low, the impact of misleading messages associating tobacco with health and educational, financial and personal success is far greater than elsewhere. A ban on advertising, particularly when associated with sporting events, would show the government's determination to act and publicise the dangers of smoking. This action may be strenuously opposed by the tobacco industry - although if, as industry claims, advertising does not induce people to take up smoking, there should be little reason to object to a ban. Legislation should not, however, be taken on its o~ but must be linked to public information and education program_ues. The r~ss media have a crucial role to play in making the public aware of the dangers of smoking and TI0~o5_o9~

in conveying the r-essage that not sr-oklng is natural social behavlour and should be the social norm. Non-smoklng areas - in schools, hospitals, and public transport - make it clear that non-smoking is the norm and help to protect the health and rights of non-smokers. As an example of smoking control programs, in the State of Kuwait these are being carried out with the participation of government and private organizations and the following measures have been taken: I) raising taxation on tobacco; 2) regulation of tar and nicotine limits to a maximum of 15 mg. tar and I mg. nicotine in imported cigarettes; 3) banning of tobacco advertisements in streets and public places; 4) establishment of Kuwait Smoking and Cancer Prevention Society to carry a wider campaign against smoking; 5) establishing a special laboratory for smoking to do future analysis of imported cigarettes; 6) a special watchdog committee has been formed from several governmental and community organizations to follow-up the implementation of the smoking control regulations which have been passed by the Kuwait Council of Ministers; 7) recently a smoking cessation clinic has been opened in Kuwait to help smokers to quit smoking; 8) many workshops, seminars and educatlonal smoking control programmes have been organized at local, regional and international levels; 9) at the same time there are ongoing educational activities for school children and youth clubs; I0) studies and researches are organized by many organizations to cover some aspects of smoking patterns among the population. At the international level, the Expert Committee hoped that WHO, as a lead- ing agency, could seek the help of other organizations in the United Nations system, according to their field of competence, in dealing with this world- wide problem. The Committee also recommended chat WHO should not engage in a dialogue with the international tobacco industry until the latter publicly accepts the scientific fact that smoking is "a major avoidable cause of death and disease." The Committee considered two main categories of smoking control measures: (1) those leading to changes in practice among those engaged in the manufacture, promotion, or sale of cigarettes (e.g. promotlonal bans, reduction in tar, nicotine and carbon monoxide yield); (2) those leading to changes in practice among smokers (e.g. restictions on smoking in public places). It was the view of the Committee, as of previous Expert Committees, that measures in the first category would almost invariably require legislation if they were to be satisfactorily implemented. Measures in the second category, however, might often be achieved through voluntary controls. The Cc~--=~ittet~elt--s~-ron~ly--rh~%-;--wh-i-~e---rhef~--i-s--~very reason to C~rb--t~e activities of those engaged in the sale and promotion of tobacco, it is not desirable to infringe the liberty of individual smmkers (who should, how- T[08350694

ever, he fully informed about the dangers of smoking) so long as their smoking does not infringe the liberties of non-smokers. The Committee considered arguments often raised by tobacco interests to the effect that any legislation in this area infringes individual liberty. The Committee felt that the legislation it recommended infringed no liberty other than that of the tobacco manufacturer to sell and promote the main avoidable cause of lung cancer, heart disease, and bronchitis. Further, the Committee felt that such legislation was aimed, particularly~ at preventing the onset of smoking among the young, and as such would increase the freedom of people to lead a healthy life. "In most countries where action on smoking has been implemented, this has resulted from pressure exerted initially by leaders of the medical profession. An important starting point might be a survey on smoking habits amongst doctors in a particular country. If there is still low awareness among doctors of the dangers of smoking, then the first target clearly will be doctors and health professionals themselves, who must set an example if they are to expect the government to take action. When it can be seen that the medical profession is setting an example for the rest of the community, the government will be more amenable to appropriate pressures. The major reports responsible for action on smoking in developed countries have come from prestigious medical sources (e.g. the Royal College of Physicians of London, the United States Surgeon General, the Norwegian Cancer Society, the Swedish National Smoking and Health Association). The major responsibility for recommending and promoting action on smoking rests in each country with the leading medical bodies." "The full case for a ban on tobacco promotion is set out in the report of the 1978 WHO Expert Committee. It is important to recognize that a ban on tobacco promotion should be: Complete: partial bans lead to parti~l results, and any loop- holes can be exploited. Seen, not as a panacea in isolatlon, but as part of the comprehensive government programme. Fully implemented: an advertising ban that is ill-enforced is likely to have much less impact than one that is adequately monitored and implemented. Long-term: the effects of a ban are to be expected less from cessation among current smokers than from a contribution to a decline in the number of young people taking up smoking. Well publicised: in countries where tobacco advert~slng has been banned, much of the impact has come from publicity surrounding the debate before the introduction of the legislation. A major priority: both in countries where smoking is already commonplace and in countries where it has not yet become popular". T108350895

"The effect of a ban on tobacco promotion derives not only from the absence of advertising, but also from the impact of a government's decision to act. A government decision to ban tobacco promotion will engender =~ch publicity about the dangers of s~oking; it will also serve as a signal to the population at large that the government feels the smoking problem to be of sufficient magnitude to merit such action. A decision to ban tobacco promotion further helps to create a climate in w~ich smoking is no longer seen by young people as a socially desirable activity, and in which health education can flourish, free of sophisticated, expensive, and misleading opposition." T108350696

SMDKII~ ~D ~]MEN Mary Jane Ashley, M.D. Professor and Chairman Department of Preventive Medicine and Biostatistics Faculty of Medicine University of Toronto Toronto, Ontario MbS IA8 Canada There can be little doubt that smoking is the leading preventable cause of premature death, illness and disability in the developed countries. In many of the developing countries it is a rapidly growing public health problem of significance. From a world perspective, the challenge before us now is no less than it was four years ago in Stockholm. Indeed, it is greater. The knowledge base from which we can act to prevent and control smoking has grown at an accelerated pace. We are in a better position than ever before to mount effective strategies against the smoking epidemic in the developed countries, and to prevent or abort the threatening pandemic, with its in- evltable, staggering consequences. The aim of this paper is to address the broad topic "Smoking and Women", and of necessity, my remarks will be confined to a few considerations. Long- term trends and current patterns of smoking among women will be compared with those among men. My perspective will be limited to the developed coun- tries, using, as examples, the United States and Canada. The health conse- quences of smoking for women will be summarized, concentrating especially on those which are of special significance or, indeed, unique to them. Very briefly, two considerations related to prevention and control, specifically, the role of maternal smoking behaviour, and smoking cessation in women, will be addressed. Finally, two issues with regard to smoking and women will be highlighted. SMOKING ~MDNG ~K)~ ~MERIC~ WDMEN During most of this century the risk exposure of North American women to cigarette smoke, the only significant mode of tobacco smoking by women, has been markedly less than that of their male counterparts (1). Long-TermTrends As illustrated by data from various national surveys in the United States (Figure i), the onset of widespread cigarette smoking among women lagged behind that among men by 25-30 years (i). North American men took up ciga- rette smoking rapidly at the beginning of the twentieth century, especially during World War I, and cigarettes quickly replaced other forms of tobacco TI08350697

8 ASHLEY use. By 1925, approximately 50% of adult r~n in the United States were regular cigarette smokers, a prevalence rate which was maintained into the 1960's. In contrast~ widespread cigarette s=.oking ar.ong wo~e_n is of ~ore recent onset. Prior to World War II, less than 20% of American wo~en were regular cigarette smokers, and a prevalence of around 33% was not reached until the 1960's. Because of the sustained high prevalence of smoking in men and the increasing prevalence in women, by 1965 the sex differential in exposure had diminished considerably. During the last 15 years, this sex differential has diminished even further, but not because the prevalence of smoking has continued to increase among women, relative to men. Rather~ the prevalence in both sexes has declined. The sex differential has continued to diminish because, at least to date, this decline has been much marked among men than among women. FIGURE 1 ESTIMATES OF THE PREVALENCE OF REGULAR CIGARETTE SMOKING AMONG ADULTS, UNITED STATES, 1935-1980 50 ~o 30 10 , ~ J t i I I I,. ,I 1935 19~0 1945 19.~0 19.55 |960 196.~ 1970 197,~ 1980 Year Note: Adapted from Public Health Service (1,2). S~oking in Selected Birth Cohorts experienced lower diffusion rates of cigarette smoking than did the corre- sponding cohorts of men (1). Again, data from the United States illustrate these sex differences (Figure 2). Rates of smoking over ti~e are shown for men and women in three ten-year birth cohorts, each sepsrated by 20 years. TI08350698

FIGURE 2 PREVALENCE OF CIGARETTE SMOKING IN l~N AND WOMEN IN THREE BIRTH COHORTS, UNITED STATES 2O 10 Birth cohort 1911-20 Men ~ ~ J~l I I I I Birth cohort 1931-40 ! ! I Birth cohort 1951-60 MeN Women- Note: Adapted from Public Health Service (i). The cohort of men born between 1911 and 1920 experienced the highest diffu- sion rate of cigarette smoking of all male ten-year birth cohorts. The prevalence of smoking peaked at 71% around 1946-48, when these men were, on average, in their early thirties. In contrast, the diffusion of smoking in the comparable female cohort was slower and less extensive. A peak preva- lence rate of about 37-38% was attained around 1959 to 1961, when these women were, on average, in their mid-forties. The mean age of smoking onset in men has been estimated at about 18 years~ while in women, it was around 22-23 years. In the male cohort born between 1931 and 1940, the diffusion rate of smoking was not as great as it had been in the cohort of men born twenty years earlier. A peak prevalence of 61% was attained in 1960-62, when these men were, on average, in their mld-twenties. In contrast, in females the peak prevalence was =ach higher than it had been earlier. Indeed, this cohort of i women experienced the highest diffusion rate of all female birth cohorts, reaching a peak prevalence of 45% in 1966-68, when these women were, on average, in their early thirties. The mean age of onset of smoking in men was about 17, and in women, about 19. The rate of diffusion of smoking was still slower in women than in men. HO~y~ ~m~axed_w_ith the earlier ---~ ......... C0ho-r~-,-the ~ffusion rates were more rapid in both se>:es. [ The striking similarity of smoking diffusion patterns in men and women in the 1951-60 birth cohort stands out in marked contras~ to the previous birth TI08350699

cohorts. Prevalence rates were lower in both sexes, but particularly so in men. Peak prevalence rates of 40% and 38% in men and women, respectively, were attained in 1976, when these persons were about 20 years old on aver- age. The mean age of smoking onset was virtually identical in men and women, about Current P~tterns Data from other developed countries support the conclusion that the smoking patterns of men and women, which have been quite different, are becoming increasingly similar. And at the same time, it is evident, certainly in North America but also in some other developed countries, that the preva- lence of smoking is declining in both sexes. These trends are most evident in adolescents and young adults (Figure 3). The rates of regular smoking*, which had been quite different be=ween the sexes, are now very similar. As well, a pattern of decline is evident in both sexes, due primarily to decreasing rates of initiation of smoking. FIGURE 3 PREVALENCE OF REGULAR CIGARETTE SMOKING AMONG ADOLESCENTS AND YOUNG ADULTS 40¸ 3O 0 ~ ~=---"'¢~'~ a les ~ . ~ 15-19 years L.anaoa / -- -.L-_~_ ~ ~ Females ~ =,~ ~ Males r / ~18 yea s United ~'~"~e m ales ~ States ~ ......... I I ~ ~ ' ' ': ' I I I I I I I I I 1965 1970 1975 1980 Year Note: Adapted from Public Health Service (I) Health and Welfare Canada (3,4,5). In Canada a regular smoker was someone usually sm~klng every day. In the United States regular smokers included those who smoked at least weekly. However, ~n 1979 about 90% of regular smokers used cigarettes daily. TI03350700

Rates of Heavy Several other s=oklng patterns have prevailed in the past, which have favoured a lower risk exposure in women. The intensity of s~oking has been less. In the United States the rate of "heavy" smoking, for example 25 or more cigarettes per day, was lower in women than in men at all points bet- ween 1965 and 1979 at which comparable measurements were made (i). However, in women the absolute rate of heavy smoking has been increasing, whereas it has remained relatively stable in men; thus, the sex differential in expo- sure risk has further diminished. However, the proportion of heavy smokers relative to all smokers has increased in both sexes. In 1965, about I in 4 and i in 7 male end female smokers, respectively, smoked 25 or more ciga- rettes per day. By 1979, the corresponding figure were about 1 in 3 and i in 4.5. Use of Filter-Tip Cigarettes Data from the United States also indicate that past cohorts of women smokers have favoured filter-t~p cigarettes more than men (i). This preference would tend to diminish risk exposures, at least with regard to some health consequences. But again, differences between the sexes have almost dis- appeared. The use of "filter-tlps" has quite overtaken the use of non- "filter-tlps" in both men and women smokers. Use of Lo~ 'Tar" Yield Cigarettes Women smokers also have shown a greater preference for low "tar" ciga- rettes. The proportion of smokers of both sexes, who smoked cigarettes yielding less than lO mg "tar", increased steadily throughout the 1970's. None-the-less, the sex differential is clear. Although there is some evi- dence that the smoking o~ low "tar" yield cigarettes may be less hazardous with regard to some health consequences, it is recognized that compensatory changes in the style of smoking may offset or substantially minimize any potential risk reduction (6). In summary: in past decades women incurred lower cigarette risk exposure than men. Trends in smoking and patterns of cigarette usage, however, have become much more alike between the sexes end, today, ~he r~sk exposures of men and women are essentially the same. Fortunately, the prevalence of - smoking in both sexes is declining. If these trends continue, risk exposure of both sexes will be less than those of the past. None-the-less, at the beginning of this decade about 52 million Americans and 6 million Canadians were cigarette smokers. For some years to come, significant numbers of North Americans, of both sexes, will suffer unnecessary morbidity, dis- ability, and mortality. ~_ALT~ CONSEQUENCES OF SMOKING FOR hDMEN That smoking is a major health hazard for women is irrefutable. In prospective studies, the overall mortality risks of women stokers have T108350701

been found to be 1.2 to 1.3 rites those of ~o=en non-stokers. The mortality risk increases with the number of cigarettes sm=ked per day, with an earlier age of beginning stoking, with a longer duration of smoking, with inhalation of cigarette smoke, and with the tar and nicotine content of cigarettes stoked. For example, in the largest prospective study, involving more than S60,0D0 American women from 25 states, the mortality ratio, which overall was 1.3 increased to 1.6 in women smoking 40 or more cigarettes daily, and to 2.2 in women smoking this amount, who indicated that they inhaled moder- ately or deeply (1). In these studies, the overall mortality ratios of women smokers were some- what less than those of men smokers. This is a reflection of factors contributing to the lower risk exposures of women in the past, such as a later age at starting to smoke, a lower intensity of smoking, and a lower tar and nicotine content of cigarettes smoked. As women and men smokers become more alik~ in their smoking characteristics, their mortality expe- riences will do likewise. Norbldity Compared with women who have never smoked, women smokers report more chronic conditions, includin~ bronchitis, emphysema, sinusitis, peptic ulcers, and arteriosclerotic heart disease (I). For most of these chronic conditions, there is a dose-response relationship between cigarettes smoked per day and the frequency of reporting. The age-adjusted incidence of acute conditions, such as influenza, is 20% higher for women who have ever smoked, than for non-smokers (I). Currently employed women, who smoke cigarettes, report more days lost from work due to illness and injury than working women who do not smoke. Limitation of activity is reported more commonly among women under the age of 65 who have ever smoked, than among those who have never smoked. Disability The burden of disability attributable to smoking among Canadian women has been calculated recently from data collected in the Canada Health Survey, 1978-79 (7). In women aged 15-64, the relative risk of experiencing dis- ability was 1.25 times higher in current or former smokers, compared with non-smokers. Almost 12% of total disability days were attributable to smok- ing. Based on these data, it can be calculated that smoking accounts for 19.4 million disability days annually or 2.65 days per year for each woman in this age group. ~a~cer Smoking in women is a cause of cancer of the lung~ larynx, oral cavity, and esophagus. It is a risk factor for cancer of the urinary bladder, kidney, renal pelvis, ureter and pancreas. In the United States, it was estimated for the year 1986 that ci~a_K~t~_e di~g-no~ cancers in women and for one-quarter of all cancer deaths (I). Data from three prospective studies indicate that the relative risk of lung cancer mortality in women cigarette smokers is about 2 to 5 times that of women non-smokers (Figure 4). The lower relative risks in women compared TI08350702

13 with men reflect the sex differential in risk exposure previously des- cribed. As women and men become more alike in their smoking behaviours, so will their risks of lung cancer. FIGURE 4 LUNG CANCER MORTALITY RATIOS* IN THREE PROSPECTIVE STUDIES 14.0 Men r'~ Women 10.1 8.2 British American Swedes doctors volunteers Age- adjusted Note: Adapted from Public Health Service (I). This is i11ustrated by lung cancer incidence rates in white Americans in the age groups 35-44 during the period 1969 to 1978 (Figure 5). In 1969 the incidence in men was more than twice that in women. Ten years later, the incidence rates in the two sexes were virtually identical, a direct result of a gradually decreasing risk in men and a rapidly increasing risk in women (S). Chronic Obstructive Lung Disease Smoking in women, as in men, produces lung damage, manifested by cough, sputum, shortness of breath, wheeze, diminished lung function, and destruc- tion of lung tissue. Whether women suffer the same frequency of these out- comes as men, given similar risk exposures to cigarette smoke, is still a matter of debate; there are a number of studies which suggest that they do. What is clear, however, is that chronic obstructive lung disease mortality rates in ....... " --'-n-- wo~en are_aporoachiDg--~ze.se i m~ f differential in smoking exposures. For example, in Canada between 1971 and 1980, the age-adjusted mortality rates from bronchitis, emphysema, and asthma, combined, increased 34% in women, while the corresponding increase in men was 9% (9). TI0~3350703

FIGUP~ 5 LUNG CANCER INCIDENCE IN ~'~ITE A~RICAt~S* AGE 35-44, 1969-1978 20 0 0 0 0 0 "- ~0 L - : 6.3 I, I I I 1969 1970 1971 1972 1973 1974 1975 1976 1977 t978 From five areas Note: Adapted from Horm and Asire (8). Coronary Hear~ Disease For coronary heart disease, the leading cause of death in both sexes, ciga- rette smoking is a major independent risk factor in women, as it is in men. In general, cigarette smoking in women increases the r~sk of coronary heart disease by a factor of two (I). However, data from a case-control study of non-fatal myocardial infarction in women under the age of 50, conducted in the United States, suggest that the impact of cigarette smoking on risk is particularly marked in younger women, especially in those who smoke heavily (i0). In the age group 45-49, the risk among women smoking 1-14 cigarettes daily was 1.8 times greater than that among women who had never smoked. This risk increased with amount smoked to 3.9 times among those smoking 25-34 and also 35 or more cigarettes daily. Among younger women, aged 30-44, the relative risk in the lightest smoking group was also ].8. However, the risk increased markedly, to 9.4 and 13 times in those smoking 25-34 and 35 or more cigarettes daily, respectively. Ci~=-rette s=o;img-a~n-g-wom6n--a~Dio ~ncreases £h~~f~e~i~heral vascular disease, suharachno~d hemorrhage, and malignant hypertension, findings similar to those among men (i). TI08350704

Reprodu=tlon-RelatedHealth Hazards With regard to cardiovascular disease, however, there is a special consid- eration for women, that is, the powerful synergistic risk interaction between oral contraceptives and smoking in relation to myocardial infarc- tion and subarachnoid hemorrhage. Data from Sartwell and Stolley (ii), illustrate this interaction. Among women who smoked heavily, but did not use oral contraceptives, the risk of myocardial infarction was seven times higher than that of the reference category, that is, women who neither smoked nor used oral contraceptives. Similarly, among women who used oral contraceptives, but did not smoke, the relative risk was significantly elevated by a factor of 4.5 times. However, among women who smoked heavily and used oral contraceptives, the risk was elevated 39 times. Other studies have demonstrated that this represents an unusually dangerous combination of risk factors for myocardial infarction, and, as well, for subarachnoid hemorrhage. This hazard is but one of a number of health consequences of smoking in women, which are related to the reproductive system and its functions, which will be considered next. They have been divided into two groups, those which are health hazards for the smoking woman, and those which affect primarily her offspring. Hazards for the Smoking Woman Several epidemiologic studies have linked smoking with decreased fertility in women, and mechanisms by which this effect could occur have been postu- lated (I, 12). There is evidence, also, that smoking is associated with an increased frequency of menstrual disorders. In a community survey of general health status reported in 1983, which involved more than 1300 randomly selected Los Angeles county women between the ages of 18 to 44, it was found that the five-year prevalence of such disorders, attended by a physician, was higher in smokers than in non-smokers (13). Smoking was a significant independent discriminator between women who did and did not report menstrual disorders. Women who smoked one-and-a-half or more packs of cigarettes daily were twice as likely to report past menstrual disorders, as women who were non-smokers. These findings are in keeping with previous reports from Finland, Australia and the United States. There is substantial evidence that smoking lowers the age of natural meno- pause. Recently, Willett et al. (14) reported on a prospective study of more than 66,000 US registered nurses, between the ages of 30-35 years, who were premenopausal at entry to their study. At a two year follow-up, just over 5000 had become post-menopausal. Age-specific, weight-adjusted incidence rate ratios for current versus never smokers were all statis- tically significant, indicating an association between smoking and meno- pause, which was independent of body weight. Based on studies of urinary estrogens in smoking and non-smoking premeno- pausal women, MacMahon et al. (15) recently postulated that a reduction of estrogen stimulus by smoking would appear to be the most_ likel_y_me_¢h~ism_ .underlying the-o~erved-ea-rl~--m~rfdpause of smokers. They postulated, also, that a reduction of endogenous estrogens might account for the increased risk of osteoporosis and osteoporotic fractures reported in women who TI08350705

16 AShLeY s=~ke. MacMahon et al. actually Ferforced their study to determine whether or not differences in estrogen profiles of s=oking and non-s=oking women might possibly be a mechanism to account for a so=ewhat lower risk of breast cancer in smokers than in non-s=okers. ~owever, as they, and otherss have pointed outs the evidence suggesting a decreased risk of breast cancer ~n wo=en who stroke, compared with those who do nots is inconclusive. It would be pre=ature to attribute any benefit to wozen's smoking in this regard. There is growing evidence of independent associations of smoking with invasive cancer of the cervix, carcinoma in situ and cervical dysplasia. The mechanism underlying these associations is not clears although several have been suggested (16). Smoking by women also increases the risk of a number of adverse maternal outcomes of pregnancy. From an extensive review of the literature, Mclntosh (17) has identified adverse outcomes of pregnancy, which occur more often in smokers than in non-smokers. From the available data "summary" relative risks were estimated, along with attributable risk proportions. Mclntosh categorized the seven outcomes in Table 1 as adverse maternal effects. The relative risks for smokers compared with non-smokers varied from 1.7 for each of premature rupture of the membranes and placenta praevia, to 1.6 for abruptlo placentae, to 1.2 for early pregnancy bleeding and I.I for spontaneous abortion. The attributable risk proportion, that is, the proportion of these unfavourable pregnancy outcomes in smokers, which are attributable to smoking, varied from more than 40% for premature rupture of the membranes and placenta praevia, to 11% for spontaneous abortion. TABLE I. ADVERSE MATERNAL OUTCOMES OF PREGNANCY RELATED TO SMOKING Outcome Estimated Relative Risk (smokers/non-smokers) Attributable Risk Proportion (%) Premature rupture of the membranes Placenta praevia Abruptio placentae Any pregnancy bleeding Late pregnancy bleeding Early pregnancy bleeding Spontaneous abortion 1.7 1.7 1.6 1.4 1.4 1.2 I.i 42 41 36 29 26 16 11 Note: Adapted from Mclntosh (17). TI08350706

Adverse e££ects on o££aprinE With regard to reproduction-related risks that primarily effect the off- spring, Mclntosh (17) detailed a number, including the i=z=ediate pregnancy- outcome hazards shown in Table 2. AEain, the risks vary, from 1.8 for fetal Erowth retardation, to 1.3 for pre-term delivery, to 1.2 for each of peri- natal and neonatal death, and stillbirth. The attributable risk propor- tions, likewise, vary from 45% to 16%. TABLE 2. Outcome ADVERSE PREGNANCY OUTCOMES IN OFFSPRING RELATED TO MATERNAL SMOKING Estimated Relative Risk (smokers/non-smokers) Attributable Risk Proportion (%) Fetal growth retardation (birth weight < 2500 g) Pre-term delivery (gestation < 38 weeks) 1.8 45 1.3 22 Perinatal death 1.2 19 Neonatal death 1.2 17 Stillbirth 1.2 16 Note: Adapted from Mcln~osh (17). The adverse health effects of maternal smoking on cffspring, however, go far beyond these immediate outcomes of pregnancy. Both the quality and quantity of breast milk may be compromised. The risk of sudden infant death syndrome is increased. Maternal smoking may adversely effect various aspects of the child's long-term physical growth and function, intellectual and emotional development, and behavlour. As well, there is growing evidence that maternal smoking is associated with increased childhood morbidity and mortality, especially from respiratory conditions, but as well, from a variety of other causes (l). In some instances, it is not clear whether these risks are incurred because of maternal smoking during pregnancy, or as the result of passive exposure to maternal smoking during childhood. There is no doubt, however, that maternal smoking is a health hazard for offspring, and that this hazard manife~ts_~tself_in _many ~ay_%. One~__w.~v~y_be--~.hr~ug,h- ~he-assoc-iation--~f maternal smoking bahavlour with the smgking behavlour of offspring. TI08350707

P~OH AHD COh~F~DL The Role of Maternal S~oking Behaviour It has been shown repeatedly that smoking is =ore co=~n among children if one or both parents smoke. However, maternal smoking may be particularly noteworthy in this regard. The results of four surveys of smoking behaviour in children and adolescents are shown in Figure 6. Two of these surveys, one conducted in 1960 and the other in 1980, involved Winnipeg school children (18). One was a survey of Canadian schoolchildren, conducted in 1972 (19) and the remaining one was a survey of Derbyshire children, conducted in 1974 (20). Compared with girls whose parents did not s~oke, the rate of smoking was elevated if either the father alone or the mother alone smoked. How- ever, the relative ratio of smoking was much higher in association with maternal than with paternal smoking. For smoking in boys, there was also an association with both maternal and paternal smoking, but the differential between mother's and father's smoking was not so marked, with the exception of the Derbyshire findings. Overall, it would appear that the smoking behaviour of parents is linked with that of their children, this association being particularly marked for ~others and their daughters. FIGURE 6 RELATIVE RATIO OF SMOKING* BY PARENTAL BEHAVIOUR '-]Neither F'~2Only. father Only mother parent smokes ~ smokes lille smokes Winnipeg Canada DerbyshIre WInnipeg Ages 10-18 Ages 8-18 Ages 11-12 Ages 10-18 1960 1972 1974 1980 GIRLS 1.8 t8 1.2 BOYS 20 ~ .1.8 1.6 [ 1.~ljI 1. 1.6 • . 1.4 1.4 !.4 1,4 J 17~ "4 ]~0~ 1.3 • Wmmpeg 1960.1980 and Canada.1972-1 + c,garette ! week; De~ ~J h~re-trie~-i~l-b-r-s-moked-reg~la rly Note: Adapted from Harrison (18); Hanley and Robinson (19); Johnson et al., (20). TI08350708

The cessation of cigarette smoking has been a significant factor in the overall decline of smoking prevalence (I). However, there appears to ~e an impression that women cigarette smokers find it more difficult to quit than their ~ale counterparts, h%ile some data from smoking treatment programs may be cited in support of this impression, it must be remembered that the vast majority of both sexes, who quit smoking, do so on their own. Is there evidence at a population level to support this impression? Data on the prevalence of former smokers in both Canada and the United States, clearly indicate that the percentage of former smokers is higher in men than in women (2,4,5). Such data may be the basis for this impression. However, further consideration is necessary. Data on the prevalence of former cigarette smokers by age-group for Canada, 1981 (Figure 7), indicate that in the younger age groups, in which smoking diffusion rates have been much more similar in the sexes, the prevalence of former smokers is also very similar. The male preponderance of former smok- ers is found only in the older age groups, which represent those birth cohorts that were quite different from women in their smoking patterns, including their rates of cessation. FIGURE 7 PREVALENCE OF FORMER CIGARETTE SMOKERS BY AGE GROUP, CANADA, 1981 -~ 40 o 30 L -1 Men ..j Women 20-24 25-44 Age - Group 35.5 27.5 10.2 45 64 65 + Note: Adapted from Eealth and Welfare Canada (5). TI08350709

More direct evidence against this impression cor-es from recent surveys of American sr_~kers, which provide data on attempted and successful smoking cessation efforts in men and women (Figure 8). Among regular smokers, about 30% of both men and women made "fairly serious attempts to quit" within the year prior to interview. There are no clear differences between the sexes, although there may be a slight tendency for more women to make such attempts. The percentages of persons reporting attempting to quit, who also reported success in quitting, again indicate little difference between men and women, although, overall, there may be a slight tendency in favour of men. The "bottom line", however, is the percentage of recent smokers who actually became quitters during the year prior to interview. There are no consistent or marked differences between the sexes. These data, then, provide little support for the impression that women cigarette smokers find it more difficult to quit. FIGURE 8 ESTIMATED RATES OF ATTEMPTED AND SUCCESSFUL QUITTING BY ADULTS*, REGULAR CIGARETTE SMOKERS, UNITED STATES Men r-I Women Attempting to Quit 32.6 Successful Quitters Attempters 1975 1978 1979 1980 1975 1978 1979 1980 1975 1978 1979 1980 Year °1975-21 years and older; 1978,1979,1980-17 years and older Note: Adapted from Public Health Service (1,2). I SSUES_HEGARDIN~ SMO~~MEN Finally, two issues should be raised regarding smoking and women. First, how informed are women about the health risks associated with cigarette smoking and, for that matter, about smoking trends and behaviour? In a 1981 TI08350710

review of data from =may sources, the Federal Trade Co=ission of the United States found serious gaps in consumer knowledge of the specific health hazards of smoking (21). For example, nearly 50% of all women did not know that smoking during pregnancy increases the risk of stillbirth and mis- carriage, and about 30% of women did not know about the relationship between smoking, oral contraceptives and the increased risk of heart attack. As well, some wo¢en may hold an ill-founded perception of lesser vulnera- bility to the risks of smoking. In one American survey, 38% of the women interviewed felt the risk of smoking was greater for men (22). Furthermore, non-smokers, especially young women, may have inaccurate perceptions of their non-smoking status relative to smokers and may overestimate the preva- lence of smoking. In a 1981 survey conducted in Canada (23), at which time the prevalence of non-smokers in the population was about 62%, parents and young people, who were non-smokers, were asked the following question: "Do you feel, as a non-smoker, that you are part of a majority or part of a minority of Canadians? The results were striking. Significant proportions of non-smokers~ both adults and young people, held misconceptions as to their majority position. But, of young girls particularly, only 34% were correct in perceiving themselves to be part of the majority which, in fact, they were. Since the recent publication of the "Black Report" in Great Britian (24), growing attention has been focussed on inequalities in health. The second issue I want to raise is related to such inequalities, which represent an important challenge to policymakers in the health field. In the "Black Report", data on smoking in Great Britain were cited as one indicator of inequalities of health across socio-economic groups. It was noted that between 1972 and 1980, smoking among women in the professional class declined 36%, while among women who were unskilled manual labourers, the decline was only 2%. Such inequitable trends are not unique to Great Britain, nor, for that matter, to women. They are, however, a major concern. In Canada, the prevalence of smoking among women is declining. The overall decline was 7% between 1977 and 1981 (Table 3). However, this decline was not equally experienced among groups of women characterized by differing educational attainment. Among women with elementary or some secondary school education, groups together comprising 75% of the female population, the declines in prevalence were 4% and 3%, respectively. It was double that, 8%, in women with some post-secondary education, who made up an additional 7% of the female population. However, among the 18% of women in the two highest educational groups, those with a post-secondary certificate or diploma and those with a university degree, the decline was very much greater, 25% and 41%, respectively. These inequalities are, indeed, a challenge. T10,2,350711

22 ASHI~v~ TABLE 3. PREVALENCE OF REGUL&R CIGARETTE SMOKING IN WOMEN BY EDUCATION, CANADA, 1977 AND 1981 Education (Percent of total population, 1981) Percent smoking Percent cigarettes change 1977 1981 1977 to I981 Elementary (20.9) 25.4 24.3 Some secondary (54.0) 34.4 33.5 Some post-secondary (7.4) 28.7 26.3 Post-secondary certificate/ diploma (II.I) 32.6 24.5 University degree (6.5) 26.3 15.5 -4 -3 -8 All women (I00.0) 31.1 28.9 -7 Note: Adapted from Health and Welfare Canada (5,25). In conclusion: the pre-eminence of smoking as a health risk among women, as it is among men, in the developed countries, is clear. However, the responses of governments, the professions, the voluntary health agencies, the research community, and indeed, women themselves, have not been of a degree appropriate to the importance of this problem. This tragic imbalance must be redressed. Public Health Service, US Department of Health and Human Services. The health consequences of smoking for women. A report of the Surgeon General. Rockville, Md.: Office on Smoking and Health, 1980. Public Health Service, US Department of Health and Human Services, Smoking and Health Bulletin. 1981 September-0ctober; S-1,2,3. Health and Welfare Canada. Smoking habits of Canadians, 1975. -Non-Med~l--U~c--cr~-Drug~--IY~rar~-i977--D~c~er, " Health and Welfare Canada. Smoking habits of Canadians, 1965-1979. Health Promotion Directorate. 1980 December. T[08350712

Health and Welfare Canada. Smoking habits of Canadians, 1981. Health Pro=orion Directorate. 1983. Public Health Service, US Department of Health and Human Services. The health consequences of smoking: The changing cigarette: A report of the Surgeon General. Rockville, Hd.: Office on Smoking and Health, 1981. Collishaw NE. Disability attributable to s~oking. Chronic Diseases in Canada 1982; 3: 61. Canada, 1978-79. Horm JW, Asire AJ. Changes in lung cancer incidence and mortality rates among A~ericans: 1969-78, J Nat Cancer Instit 1982; 69: 833-837. Litven W, Smith H. Respiratory disease mortality trends: 1971-1980. Chronic Diseases in Canada 1982; 3: 1-2. I0. Rosenberg L, Shapiro S, Kaufman DW, Slone D, Miettinen OS, Stolley PD. Cigarette smoking in relation to the risk of myocardial infarction in young women. Modifying influence of age and predisposing factors. Int J Epidemiol 1980; 9: 57-63. ii. 8ar~well PE, Stolley PD. Oral contraceptives and vascular disease. Epidemlol Rev 1982; 4: 95-109. 12. Olsen J, Rachootin P, Schmidt AV, Damsbo N. Tobacco use, alcohol consumption and infertility. Int J Epidemiol 1982; 12: 179-184. 13. Sloss EM, Frericks RR. Smoking and menstrual disorders. Int J Epidemiol 1983; 12: 107-109. 14. Willett W, Stampfer MJ, Bain C et al. Cigarette smoking, relative weight and menopause. Amer J Epidemlol 1983; 117: 651-658. 15. MacMahon B, Trichopoulos D, Cole P, Brown J. Cigarette smoking and urinary estrogens. New Eng J Med 1982; 307: 1062-1065. 16. Public Health Service, US Department of Health and Human Services. The health consequences of smoking: Cancer. A report of the Surgeon General. Rockville, Md.: Office on Smoking and Health, 1982. 17. Mclntosh ID. Smoking and pregnancy: Attributable risks and public health implications. Can J Public Health 1984; 75: 141-148. 18. Morrlson JB. Smoking habits of Winnipeg school students, 1960-80. Can Med Assoc J 1982; |26: 153-154. 19. Hanley JA, Robinson JC. Cigarette smoking and the youth: A national survey. Can Med Assoc J 1976; 114: 511-517. 20. Johnson MRD, Hurray M, Bewley BR, Cl~de~D~=_Banks_MH_,_Swa~V~_~qoclal claas~parents~-cS~'q-dre~nd smoking. Bull International Union Against Tuberculosis 1982; 57: 258-262. T[08350713

21. L~oto J. 3educlng the health consequences of smoking - a progress report. Publlc Health Reports 1983; 98: 34-39. 22. Tagliacozzo R, Vaughn $. Women's smoking trends and awareness of health rl,k. Prey Hed 1950; 9: 384-389. 23. Canadian Health Facta, Health and Welfare Canada. Perception o~ smoking behavlour by the Canadian public. Health Promotion Directorate. 1981 October; S-03. 24. Black D, Morris JN, Smith C, Townsend P. Inequalities in Health: The Black Report. (Edited and with an introduction by P. Townsend and N. Davidson}. Great Britain: Penguin Books, 1982. 25. Health and Welfare Canada. Smoking habits of Canadlans~ 1977. Promotion and P~eventlon Directorate, 1979 January. TI08350714

25 The Honourable Monique B6gin Minister of National Health & Welfere* Government of Canada Ottawa, O~tar~o I am pleased to have been invited to deliver one of the opening addresses at the Fifth World Conference on Smoking and Health, When I heard about the overall themes of the Conference, I was excited to know that smoking among women and children and in developing countries would receive special deliberation. The incidence of smoking-related diseases among women is continuing to rise to unprecedented levels. Indeed, the tobacco industry has responded to the growing women's movement over the last two decades by promoting smoking directly to women in their new and expanding roles and by developing new brands which will be more appealing to women. Trends in smoking among children, especially girls, have, until recently, been very discouraging as wet1. While cigarette advertising is not explic- itly targeted at young people, the attractive models and lifestyles cannot help but be appealing to them. Happily, in many countries we have begun to move away from smoking, to consider it less acceptable in public places, in meetings and at some social gatherings. This may not be as true in other countries, particularly in developing nations where I understand smoking is still increasing. Researchers throughout the world have provided us with biomedical and epidemiological evidence on the causal role of smoking in human disease. Certainly this century's distinctive patterns of lung cancer - by far the fastest growing of all cancer types, especially among women - are directly related to trends in smoking. As well, the differences in the rates of lung cancer and chronic respiratory diseases between income groups can be direct- ly related to different rates of smoking. The impact of smoking on the health and productiv{ty of our citizens is astounding. As a direct result of lung cancer, heart attacks and other known smoking related diseases, nearly 30,000 Canadians died prematurely in 1979 alone. As well, in 1979, some 37 million person-days of illness were due to smoking - this represents nearly four extra sick-days for every Canadian smoker. In all, there can be no doubt that smoking is today, the single most important preventable cause of {llness and death. * At the ti~e of the Conference. TI073350715

Nevertheless, as professionals in the health field, ~e can all be encouraged by the growing number of non-smokers. In the mid-1960's over half of Canadian men s~oked daily, but by 1981 little more than one-thlrd s=oked. The trend has been the same for teenage boys; over the last decade and a half the rate of smoking has dropped by over 34%. However, those men who quit tend to have been light to moderate smokers. Those who still smoke are likely to smoke more heavily and are at twice the risk of premature death compared to non-s~3kers. Among women the overall prevalence of smoking has remained virtually unchanged. While some women have quit, the rate of smoking among teenage girls increased by 35% during the 1960's and early 1970's. Smoking among all women is now beginning to decline. But for the first time ever, in 1981, the rate of smoking among women 15-24 years of age rose above that of young men. We should not assume from these trends that the battle against smoking is being won. On the contrary, we are only just now starting to see the results of years of activity by many organizations to educate the public about smoking. We still have much to accomplish. For example, we know from a recent survey conducted by my Department that young people are beginning to try smoking at even younger ages than did their parents - an average of four years younger. We also know that young people are still being attracted to smoking by smokers around them and by their own perceptions of the extent and accept- ability of smoking. When asked, in a recent survey, to estimate the propor- tion of smokers in Canada, young people and adults agreed that the majority or almost two-thirds of Canadians smoked. This is in direct contrast to the actual prevalence of smoking since only one-third of Canadians smoke. The people who most influence a child to start smoking are his or her friends and family. Parents, in particular, are very powerful role models. If even one parent smokes, his or her children are twice as likely to smoke than are children in non-smoking families. But most parents still deny that what they do has an effect on what their children do. As professionals working in the health field, it is our responsibility to correct these misperceptions and, in doing so, to create attitudes and an environment that support and reinforce non-smoking. I would like to talk about what we are trying to do in Canada, through policy and program development, to reduce the number of smokers and to accelerate the trend toward non-smoCking. Two important and perhaps unique features that have characterized the Canadia_n~experienc#_ip_r~ga~d tp__pp_l_i~y_~nd__p=o~development-,~ are~l~r~- ........ tary agreements between industry and the federal government concerning many aspects of tobacco production and ~rketing, and collaboration among govern- ments, health agencies and health professionals in planning and developing prevention and cessation programs. TI08350716

In the 1960's and 1970's, as the public berate more and =ore aware of the impact of s=oking on health, there was growing pressure on the industry to begin to regulate some of the activities of its members. In response, and as a result of ongoing discussions with =j Department, the Canadian Tobacco Manufacturers' Council voluntarily co~z~itted its =embers to do several things which included: a withdrawal of radio and television advertising; a limit on annual promotional expenditures; limited tar and nicotine yields per cigarette; and the placement of health warnings on cigarette packs. As recently as January of 1983, I asked the Council to interpret their advertising and promotion code even more strictly. Because I believe that consumers are not yet adequately warned about the hazards of tobacco use, in every cigarette advertisement and on all tobacco packaging, I have asked the manufacturers to ensure that: - the health warning as well as the tar and nicotine informa- tion appear on all print advertisements, including bill- boards; - no cigarette advertisement appear on public display within 500 metres of any primary or secondary school property; - the health warning, as well as the tar and nicotine information, appear on all cigarette packaging, including cigarette carton wrappers; the Canadian health warning and tar/nicotine information appear on all cigarettes of foreign manufacture which are imported and distributed in Canada by the Manufacturers' Council's members; and - the Canadian health warning also appear on all packages of cigars. The industry has now agreed, in principle, to the last two of these requests and I am optimistic that progress can be made towards agreement on the other three. I have also asked Canadian cigarette manufacturers to take certain steps to reduce the levels of some of the more hazardous constituents of cigarette smoke. By 1984, the level of carbon monoxide should not exceed that of tar for any brand of cigarette. As well, each company has been asked to reduce average tar yields to 12 mg and to ensure that the average nicotine yield does not exceed I mg for any brand. Given the encouraging results of our annual analyses of cigarette yields, I am again optimistic that the manufac- turers will be able to reach these targets by the end of next year. These industry initiatives are being complemented by new initiatives to reduce what is known as "involuntary smoking". Although the evidence is not yet complete, there is growing recognition that second-hand smoke is an ir-ritant and heelth-haz-ard-~c~-~on--~-mo~6r~. A number of Canadian municipalities have already taken the lead in protect- ing Canadians from smoking by passing bylaws to regulate smoking and non- TI08350717

smoking areas in indoor locations under their jurisdiction. I am pleased to tell you that Winnipeg is one of the nine cities that have such bylaws. To further protect Canadians, my Department is taking steps to restrict smoking in public areas under federal jurisdiction. To begin with, over a year ago guidelines on smoking and non-sm=klng in federal health buildings were introduced, on a voluntary basis. I am hopeful that, through discus- sions with some of my Cabinet colleagues, these guidelines will be adopted throughout the federal government. In the case of interprovincial transit vehicles, such as planes and buses, a number of measures have already been taken, voluntarily, by the carriers themselves. Together with Transport Canada, my Department is now working in consultation with the carriers to develop even more effective control of tobacco smoke. These policy initiatives are an important step toward promoting health by establishing smoke-free environments. In order to have a significant impact on the prevalence and resulting consequences of smoking, we must complement and reinforce these societal changes with public information and education which promote healthy indivi- dual decisions. In this regard, we are very fortunate in Canada to have a we11-developed network of health agencies, which are concerned about the impact of smoking on health and which are working actively to develop programs to prevent smoking among young people and to help those who do smoke to quit. The contribution of the Canadian voluntary organizations, which are respon- sible for organizing this Conference, has been outstanding. Over the last several decades, the Canadian Cancer Society has made a major contribution to public education about smoking, particularly through their school-based programs. The Lung Association has also been very active in all areas and is well-known for its program to help smokers to quit and then remain non- smokers. The Canadian Heart Foundation, for its part, has made significant contributions to smoking research and education. Together, in the mid 1970's, these three health organizations decided to form a national body whose focus would be smoking. This body, called the Canadian Council on Smoking and Health, has made the issue of smoking a visible public concern through its annual National Education Week on Smoking. It has also funded research, sponsored conferences, produced educational materials and advocated legislative action. There are, of course, many other national, provincial and local agencies who have resolved to work, in any way they can, to reduce smoking in Canada. It was in recognition of this wealth of interest and experience in the field of smoking, that my Department decided to expand its own activities to ...... ~edu~e-sm~k-lng--in--C~nad~ Before telling you a bit about my Department's new smoking prevention program, I would like just brlefly, to highlight the main features of a Tl08350718

smoking cessation program recently developed jointly by =j Depart=ent and the Canadian Cancer Society. The program called "Ti=e to Quit", combines a self-help approach to quitting, ~otivation and reinforce=ent through a three part television series, and strong community-based support and follow-up. The program, which will reach two-thirds of Canadians in communities across the country next year, was pilot tested in this city, Winnipeg, a few months ago. Many of the energetic and hardworking health agency staff and volun- teers, who undertook what was really a major community-wide and, in fact, almost province-wide program will be presenting to you the results of their pilot study later in the week. This is just one very good example of how cooperation between agencies and community support can be effectively com- bined to produce a successful program. "Generation of Non-Smokers" is the name of a major smoking prevent{on program which my Department is now developing in collaboration with other interested governments, health organizations and health professionals. The major thrust of the program is to build a social environment and a system of supports that reinforce non-smoking from infancy to adulthood. In planning the 'Generation of Non-Smokers" program, we recognized that one of the major barriers to overcome was the apparent social acceptabilty of smoking, which still exists in spite of the known health risks and the recent steps to control smoking in public places. To have an impact in todayts society, we felt that there were some very important principles which must be basic to the program design. These principles involve: comprehensiveness, in terms of '~ho" the program reaches and "how" they are reached; speclally-designed resources and programs; continuity, over a period of many years; and most importantly, cooperation among all organiza- tions in the field. The "Generation of Non-Smokers" program plans have incorporated all of these elements. In considering at what stage to intervene to promote non-smoking, it became clear that all stages during childhood and adolescence are impor- tant, in terms of developing values, attitudes and decisions about smoking. Programs that only look at the critical stage, when young adolescents are trying smoking, are ignoring the real and powerful influences throughout society which lead young people to believe smoking is all right and even natural. Parents have a particularly important role in encouraging their children to adopt healthy lifestyles. Since values and attitudes in favour of non-smoking must be developed from an early age and then consistently reinforced, "Generation of Non-Smokers" is being planned as a long term, possibly 15 year, program. My Department has already developed plans for its contribution over the first five years of the program. Our initial activities will involve developing and promoting projects aimed at developing values, attitudes and decision-making skills a=ong very young children and adolescents. "Generation" program name. My officials have, in fact, already discussed the program with representatives of provincial governments and voluntary T108350719

health organizations and I am very pleased to say that a number of these groups are ve~] interested and supportive of the Generation program. Smoking is a very serious health and social problem that requires concerted action on the part of government and other health organizations in three major areas. Firstly, effective action is needed to reduce tobacco consump- tion. Secondly, protection of the health of non-smokers by reducing involuncary exposure to second-hand smoke is a pressing priority and one that requires action at the national, provincial and community levels. Thirdly, and most important, our goal must be to prevent the initiation of smoking during childhood and adolescence. To do this, we all need to recog- nize smoking prevention as a priority and then allocate adequate resources on a long-term basis to coordinated and comprehensive programs. Let me assure you that I am committed to the pursuit of these goals at the federal level in Canada. But governments cannot achieve these goals acting alone. I know that most of the world's professional, voluntary and govern- ment organizations concerned with smoking and health are represented at this Conference; #our help will be a vital part of the development of effective public policies concerning smoking and health. Let us all work together and do our part to reduce the prevalence of smoking throughout the world and ensure that our children have the opportunity to grow up in a healthy, smoke-free environment. T[08350720

31 LEGISLATION ARD POLITICAL ACTIVITY Ejell Bjartveit, M.D. National Council on Smoking and Health Postboks 8025 Dep Oslo 1 Norway Legislation and political activity were a main theme throughout the Fifth World Conference. Such measures were emphasized as essential to combat the world smoking epidemic, and this paper was intended to give a general intro- duction to the sessions which dealt in more detail with various facets of legislative and political action. What kinds of measures can be used in legislative action? Professor Ruth Roemer has given a brilliant and thorough presentation (i); and here I shall just try to summarize briefly the restrictive measures which have been implemented or proposed in an attempt to influence smoking behavlour (2-6). They can be grouped as follows: Restriction of influences encouraging smoking (a) Reduction of explicit influence, for example, a ban on promotion. (b) Reduction of implicit influence, for example, a ban on smoking on tele- vision. Restrictive measures to discourage smoking (a) Health warnings on tobacco-packets, including packets for export. (b) Health warnings in such tobacco advertisements as are permitted in the absence of a ban. (c) Declaration on packets and in advertisements of emission levels of harmful substances. (d) Mandated health education, including mandated funding. Sales restrictions (a) Limitation of sales outlets, i.e. number of shops permitted to sell tobacco, and of vending machines. (b) Limitation of hours of sale, for example, only during ordinary opening hours for shops. (c) Age limitations, i.e. prohibition of sale to minors. (d) End to sales in health premises. (e) End to duty-free sales. TI08350721

32 BJA~T~Z~T Product restr~ctlona (a) Upper li=it of tobacco content per cigarette (concerns also goods for export). (b) Upper limi~ for e=isslons of defined harmful substances (concerns also goods for Taxation General tax increase on tobacco products. This is undoubtedly one of the mos~ effective measures available. To achieve maximum impact and Co preven~ a vaning of the effec= over ~ime, the exercise needs to be repea=ed a~ more or less regular intervals. Selec=ive tax ~ncrease, i.e. a graded taxation according to emissions defined harmful substances. Restrictions on a~oklng which establish non-smoking as a norm, and limit smoking to defined zones and/or times. (a) Res~rlc~ions on smoking in public places. Restrictions on smoking at places of work. This applies partlcularly to occupations where industrial pollution, e.g. asbestos, causes a synergistic risk increase for the smoker (probably also for the passive smoker). A DILENNA This list, which is by no means complete and whose arrangement may be ques- tioned, presents an arsenal of powerful weapons. A signal of their impor- tance is the fec~ that, with few exceptions, these measures have been fought vigorously by =he tobacco indus=ry. On the other hand, it is also clear that the llst of restrictions may cause irritation in the general population, and a feeling of being under the guardianship of the authorities. Boomerang effects may occur if restric- tions are not introduced with caution. Therefore, it is important not =o implemen= restrictions which will not be understood and respected. This s~tuation demonstrates our dilenma: if restrictions are not utilized, we may lose opportunities to influence a major health problem. If utilized too fast, and without inte111gence, they may turn out to be useless and even counter-productlve. To find a point of balance, it is necessary to observe public opinion in t~i~--fi-~Id--closely?-~urpri~-i-nglyT--,hr~v~r~en--publ~c opinion--i~ me~-~u~ed, it may reveal that people agree to restrictions more often than anticipated. TI03350722

LZGISLATI0.~ ~ P~LITICAL AL'TIVITY 33 ATTITUDES T~ lEGISLATION In s 1982 survey (7) of a representative sample of the entire Norwegian populatlon aged 16-74~ • substantial proportion wanted ~ore or less strong restrictions on smoking in publlc places (Table 1): TABLE 1, ATTITUDES TO RESTRICTIONS ON SMOKING IN PUBLIC PLACES, REPRESENTATIVE SAMPLE OF THE NORWEGIAN POPULATION AGED 16-74, 19B2. N " 2597. Question: ~t has been discussed whether or not specific rules should be introduced £or smoking in public places and work premises. Which of these arrangements would you prefer? Percentages Waiting ~ooms for travellers Waiting rooms in hospitals at work Offices and smaller work premises Restau- rants and cafes Total Separate Smoke- Sum Request prohi- rooms for free restrict- to bltion smokers zoues ions smokers not to smoke 16 49 21 86 8 53 36 6 95 3 17 38 24 79 II 37 23 9 69 22 8 13 28 49 ii Un- limited smoking 2 I0 40 Source: TI08350723

~JARTVEIT In a 1982 survey (8) involving a representative sa=ple of the population across Canada, aged 15 and above,* 90% wanted non-smoking areas in restaurants which are large enough to have them. 92% wanted non-s~oking areas on buses, trains and air- planes. In a 1973 survey (9), before the Norwegian Tobacco Act entered into force, 81Z of the adult population aged 16-74 were in favour of the advertising ban on tobacco and of the compulsory health labelling.** In 1982, 46% of Canadians wanted cigarette advertising to be elimlnated al- together. Only 7% would have a11owed it to increase (Table 2). In a 1979 Norwegian survey (I0), 56% of the adult population aged 16-74 supported an increase in cigarette prices as a measure to influence the damaging habit.*** TABLE 2. ATTITUDES TOWARD CIGARETTE ADVERTISING. REPRESENTATIVE SAMPLE OF THE CANADIAN POPULATION AGED 15+, 1982. N z 2340 Eliminated altogether Reduced to lower level Restricted to current level Allowed to increase Percentages 46 19 28 7 Don't know answers disregarded (5%) Source: (8) L~ISLATION: O~ PART OF A COMP~XH~NSIV~ P~0CRAMlq~ The positive attitude towards the restrictions referred to here probably reflects results of earller information and education activities. Here we touch upon an indispensible prerequisite for implementing legislative action: it n~st be integrated in a total, comprehensive, we11-balanced smok- ing and health programme, which includes both information and education, as * N = 2340. ** N = 2313. *** N = 1511. Don't know answers disregarded (1-2 %) Don't know answers disregarded (18.0%) Don't know answers disregarded (5.5%) T106350724

35 well as cessation activities. In such a progra~.r~e, the restrictions will function as a catalyst to the other elements. One advantage of legislation is that, in terms of money, it costs very llttle. However, so=e governments might think that legislation is a cheap alibi for not investing in more expensive education and cessation programzes - people will still have the impression that something is being done. To introduce legislation in a vacuu~ of other activities, however, would probably have no effect at all. The same applies Co partial or inadequate legislatlon. ~UEII~G EFfeCT OF IW~CISLATION What effect has anti-smoking legislation? Let us take an advertising ban as an example. It is remarkable how the interest in this question has increased steadily, along with a demand for data. Enthusiasts ask, obvious- ly because they want proof that can convince their legislators about the necessity of an advertising ban. The tobacco industry asks, obviously because they want proof that can convince their legislators that an adver- tising ban does not work. Bureaucrats ask, because they want a solid basis for action, or, a few of them may want excuses for postponing the trouble of preparing drafts for their legislators. Politicians ask, because they want evidence than can convince their parliamentary colleagues in either direc- tion. I feel it is necessary to pour some cold water on the confidence in the measurements used for estimating an effect of an advertising ban (ii), and these are my reasons: (1) I think that it is quite impossible, in the strict scientific sense, to quantify the effect of an advertising ban. After all, we are not deal- ing with a controlled trial, where we have two isolated communities, identical in demographic, social and cultural structure, in smoking habits as well as in smoking and health programmes, one with and one without an advertising ban. (2) Legislation is part of a comprehensive programme and we want it Co be so. It is imposslble, however, to select one element from such a programme and estimate its isolated value. (3) Legislation may have several phases of effectiveness. First a short- term effect due mainly to the publicity generated when the Act is announced, discussed and implemented. In this connection, we should also welcome resistance to legislation, so long as it fails, as it may focus people's interest even more on the health consequences of smok- ing. Then there is the long term effect, particularly on children and young people, who will grow up in an environment free of an adver- tising pressure which glorifies the smoking habit as a key to success, self-confidence and adulthood. This long term effect is the important one, but this implies that it will take decades before new trends in s=oking habits of young generations will substantially_i~flue~ce the t~al~pef-capita-co~shmpt~on. TI08350725

(4) We should never accept the challenge from the tobacco industry, and from politicians and bureaucrats, that we have to prove an effect of an advertising ban before this question can be decided. In =r.~ view, the burden of proof lies with the tobacco industry. It is up to them to prove that a world-wide investment of 2 billion Canadian dollars a year* in tobacco advertising and promotion has no inpact on consumption levels of youth. In rmny countries today, politicians and bureaucrats sit on the fence, say- ing: Let us wait for results from countries which have already implemented legislation. From what I have said, they will have to wait for many years before so-called proof of effect can be given, and hence, many health benefits may be lost. MOTIVES FOR LEGISLATION Politicians must have the courage to enact legislation without advance proof of its effect. Their decisions have to be motivated for other reasons, and I shall put forward two of them: (1) If it is true that smoking is the cause of the greatest epidemic of modern times, then it is unethical to permit sales promotion of these deadly products. We should not make it more complicated than that. A child, a teenager, will argue along these lines. They will question the double set of morals of the government, and ask very logically: if you try to convince me that smoking is dangerous for my health, why don't you stop advertising? (2) The industrialized countries should be aware of the gigantic adver- tising campaign which the tobacco industry has launched in third world countries, in order to compensate for the market they are losing in the rich countries. This campaign is the most cynical and reprehen- sible marketing activity I know of, because the tobacco industry knows very well what the health authorities have predicted as a result of an increasing tobacco consumption: these areas, where smoklng-related diseases are, as yet, relatively seldom, will come to experience them in only a few years. It would be impossible for the industry to intro- duce their products to the third world countries at the same speed if they were unable to utilize their refined and skilled advertising tech- niques. The health authorities in these countries often look to the industrialized part of the world for signals to follow. If we do not ban advertising, it is not likely that they will do so. This forces us to review our attitude towards an advertising ban. We are dealing with a pandemic, and our responsibilities go beyond our own borders. These two arguments touch upon one important aspect of legislation, namely underlining how gravely the government looks upon the smoking and health problem, thus re~nforcing and increasing the effect of information * Figure presented in 1979 (12). T108350726

LEGISLATION AND POLITICAL ACTIVITY 37 work. The intention of legislation, together with all other anti-s~oking activities, is to establish non-smoking as the norm. NORWEGIAN EXPERIENCES Although difficulties are involved in presenting so-called proof, some indi- cations, but nothing more, of an early effect can he given from countries which have introduced legislation. In my own country, a Tobacco Act was enforced in 1975 (13,14,15) including, inter alia, a total ban on advertising (also indirect advertising) and a health warning on packages.* In addition, from 1980 to 1982, we have had three price increases, due to taxation, of 29, 22 and 10% respectively. Because a government programme is expected primarily to have an effect upon young people, trends in their smoking races are of particular interest. Since 1957, nationwide surveys of smoking rates among students in the basic .school have been conducted four times (Figure I). Increasing rates were registered up to 1975, and smoking among girls in particular showed a drama- tic and alarming increase, with rates in 1975 equal to or above those of the boys at all age-levels in the upper grades. In 1980, the rates were on the decline for both sexes, most pronounced for the girls, who at all age-levels were back again to lower smoking rates than the boys. Because these surveys have been carried out at long and irregular intervals, it is uncertain whether or not 1975 represents the peak year. The rates may have been even higher but without doubt the top was reached during the 1970s. The decline in 1980 is most promising. Sales figures also support a new trend after the Parliament decided to introduce the government programme (Figure 2). There was an increase in per capita tobacco consumption until 1970, the year when the Parliament discuss- ed the issue and endorsed a government programme on smoking and health, including legislation. Since then, the per capita consumption has levelled out, and during the last years shows a tendency to drop. Although some of this decline may be due to increased purchases abroad, there seems little doubt that a decrease in consumption has taken place. * It may be noted that the extent of tobacco promotion in Norway, measured by expenditures, was moderate compared, for example, with the UK and the US. In 1974, the year before the enforcement of the advertising ban, the equivalent of about US$2.75 million (1974 value) were spent in Norway on newspaper and magazine advertisements, movies, trade papers and outdoor posters for tobacco promotion (9). The figures for roughly corresponding categories of promotion were US$55.89 million for the UK (I0) amd US$241.I million for the US (11), the US figure referring to the 20 best selling brands with about 90% market share. Calculated expenditures per inhabitant: US$0.69_fo~_Norway~ l..00_for the_UK_~_~n~A~om_the_qU_S_(_h0~4 vaiue~). It should also be remembered that large sums were, and are, used in the UK and US for other promotional expenditures which did not exist in Norway in 1974 (gift coupons, sports sponsorship, etc.). TI05350727

38 FIGURE I. PER CENT DAILY SMOKERS, SCEOOL STUDENTS, BY AGE A~-D SF~X, NORWAY, 1957, 1963, 1975 AND 1980. Source: 1957 and 1963: Nilsen (16,17). These surveys were carried out in samples of Norwegian schools. Sample sizes within each age/sex- group ranging from 812 to 1245 in 1957, and from 440 to 605 in 1963. 1975 and 1980: National Council on Smoking and Health, Oslo. On 4 November 1975 and 1980, all students in Norway in grades 7, 8 and 9 (aged 13, 14 and 15) were asked to fill in a questionnaire on smoking habits. Representative samples of these questionnaires were collected for analysis, samples sizes within each age/sex-group ranging from 790 to 957. The most important feature, however, is the extension of the regression line for the consumption from 1950 to 1970 (Figure 2). If the upward trend for the 1950s and 1960s had continued in the 1970s and 1980s, we would have had today a per capita consumption which would have been about 30% higher than it is. In my opinion, the shaded area illustrates what has been gained in recent years. Some may say that the Norwegian consumption reached a maturity level in the 1970s. However, the Norwegian consumption figures are far below the level for Britain (Figure 3), and Norwegians do not differ that ~,ch from the figures the United Kingdom experienced ~re than 30 years ago. Our figures are also below those of Britain in other respects (Figure 4): a=ong males aged 60-69, the British lung cancer death rate is more than three times the Norwegian. Here again our figures are about 30 years behind the British. T108350728

LEGISLATION ARD POLITICAL KC'EIVIT~ 39 FIGURE 2. CONSUMPTION PER ADULT (AGED 15+) OF K~2~FACTUP~!D CIGA- RETTE + SMOKING TOBACCO, NORWAY, 1950/51 - 1982/83. Note: The dotted line is an extension of the regression line for the years 1950152 to 1969170. The arrows indicate points of time for parliamentary endorsement of the governmental control programme, for enforcement of the Tobacco Act and for recent price increases due to taxation. Source: Sales figures: Reports from the Directorate of Customs and Excise, Oslo. Population figures: Reports from the Central Bureau of Statistics, Oslo. This implies that, since the increasing trend in tobacco consumption has now been stopped, thus avoiding a rise to the level experienced in other nations with a history of longer and heavier smoking, a considerable amount of human suffering has been avoided. TI08350729

4O FIGURE 3. CONSUMPTION YERADULT (AGED 15+) OF VAh'OFACTURED CIGARETTES + SMOKING TOE%CCO. UNITED KINGDOM A~ NORWAY, 1931 (1934/35) - 1978 (1979/80). 1930 -40 -50 -60 -70 -BO YEAR Source: D'K: (18,19). Norway: As for Figure 2. FIGURE 4. Source: LUNG CANCER DEATH RATES, MALES AGED 60-69, ENGLAND & WALES AND NORWAY 1931 (MEAN 1931-35) - 1979. ICD CODE (STH REVISION): 162~ AND COERESFONDING CODES FOR PREVIOUS REVISIONS. YEAR England & Wales: (20,21). Norway: Annual Reports from the Central Bureau of Statistics, Oslo. T[08350730

LEGISLATION ~ POLITICAL AL'TIVITY &l This new development in the smoking epidemic is still not enough. It is time to call a spade a spade, and announce our final goal: the eradication of the problem. In 1981, the Norwegian Medical Association passed the following resolution: "The Representative Body of the Norwegian Medical Association urges the Government to work towards making Norway a smoke- free society by the year 2000... Phasing out the consumption of tobacco is an important step towards improving the health of the nation." Thls resolution has received extensive publicity. The doctors ask the Covernment not only to turn its attention to this avoidable health problem, but to rid the country of it within a reasonable time. -So~e will find this goal utopian and unrealistic, and think that more time is needed. This may very well be so. The ~min point, however, is that eradication has been set up as an attainable goal, and that this goal should be reached within the forseeable future. This ought to be possible. With few exceptions, cigarettes started to invade the industrialized countries at the beginning of this century. It should be possible to get them out before we have gone too far into the next. POLITICAL VICTIMS Now the question arises whether or not the politicians are willing to take the necessary steps to reach this goal. Such steps could cost them their political career. We have already seen a couple of victims. At the Fourth World Conference in Stockholm, the US Secretary of Health, Education and Welfare, Mr. Joseph Califano, gave a plenary address, where he heavily attacked the tobacco industry (22): "We can expect that the tobacco industry will do everything in ~ts power to counteract our public-health efforts. We should, however, view such determined opposition not only as an obstacle, but also as a challenge to our creativity and skill." One month later Secretary Califano was fired by President Carter. Politi- cians from tobacco-producing states cheered: "...that'll get a million votes alone", one of them said (23). Another outstanding politician also spoke in Stockholm: Sir George Young, the OK Junior Minister of Health, said that (24): "..the solution to many of today's medical problems will not be found ~a th~-re~e~rch--l-aho-r-a~-drles ~Z-'--du~--hosp~is, but ~n our Parliaments. For the prospective patient, the answer may not be cure by incision at the operating table, but prevention by decision at the Cabinet Table." T103350731

~J~IT In 1980 in Oslo, at a World Health Day Conference, Sir George was given ~he topic: "Srmking or health - a choice for the politicians". He then said, referring to what happened with Secretary Callfano; "The words might, I think, bear more than one meaning...Smoking is in every sense a political issue, and those politicians who concern themselves with it find themselves unexpectedly promot- ed or demoted." Did Sir George have a presentiment? One year later, Prime Minister Thatcher transferred him to another Ministry, where he would be less dangerous to the tobacco industry. Press comments underlined the connection between this event and Sir George's commitment to anti-smoking legislation: '~epresenta- tions by the tobacco industry against the Government's anti-smoking campaign are believed at Westminster to have played a part in persuading the Prime Minister to shift Ministers .... ". "I never knew the tobacco industry was so powerful," said a top civil servant (25). STATEMENTS ~ lqOR~F_,GIAN POLITICIANS In January 1983, the Norwegian Minister of Health, Dr. Leif Arne Hel6e, stated in the Parliament: "The systematic work of the Government to influence public opinion has the same long term objective as requested by the Norwegian Medical Association. A smoke-free society is also the aim of the proposals in a planned White Paper." I also have the honour to draw y2ur attention to a special message which the Norwegian Prime Minister, Mr. Kate Willoch, sent to the Fifth World Confer- ence on Smoking and Health. This message is published in "Tobakken og Vi" (26), edited by Allan Aarflot, and was available to all delegates. Mr. Willoch says, among others: "The Norwegian Government will continue its efforts to reduce the use of tobacco in Norway. Two main conditions are indispensable for a successful result: firstly, a ban on tobacco advertising and promotion, and secondly, an active, informed opinion and attitude against smoking, emphasizing non-smoking as the normal social behavlour". MOBILIZIBG IwrEI~ATIOIIAL ORGABIZATIOHS Some people may think, and rightly so, that such a statement is easy to make in a country which is economically independent of tobacco production. In many third world countries, tobacco production, manufacturing and trade count for a substantial fraction of the gross national product, and the-d~iiy llv-fng-for people who have no other al~ernative than s~arvatlon and hardship. The tobacco industry has very cleverly utilized this situation. TI03350732

A~D POLITICAL /~'IVIEY 43 lith this background it sects at first glance hopeless to stem the epidem- .¢o What we need, therefore, is a worldwide political strategy for reaching :the final goal. This means an active involve=ent by international organlza- ~ns, for exa=ple the EEC, which in 1982 spent Can.$550 million on subsi- lies to tobacco production and trade, and where the corresponding budget for [~ is Can.$830 million (27). In particular, it is a challenge to members the UN family, such as FA0. This organization has previously provided Ichnical assistance for tobacco cultivation and marketing. In 1978 the ~ statement came from FAO (5): "However, since the resolution on smoking and health passed by the ~wenty-ninth World Health Assembly in May 1976, FAO has not promoted any activities leading to project execution." gh this is a good start, we are not satisfied with this policy. We :nt the international organizations to give priority to a comprehensive, lobal plan for development of substitute crops and industries, and render • possible-~assistance to achieve this goal. cannot expect the tobacco industry to support such a plan. They have had chance, and have failed to show genuine concern about the serious health consequences of their products. MOBILIZII~ POLITICIANS then, are the prospects? The simple answer is that our goal will not be obtained unless politicians and the general public all over the world are lized on our side. Mike Daube has a favourite sentence: "If you wish to do something about the smoking and health problem, you are in politics". We have to realize that this is the fact. Our task is to confront the politicians with the enormous • magnitude of the problem, to make them see that it is the greatest epidemic of modern times. They must pass from a stage of only pretending serious concern, into s stage of active involvement and determination. Let us ask "" them the pertinent question: Do you really want to do something about the .~ problem? Or is your involvement only a question of lip-service? Some of us nmy think that our job is merely to account for the scientific "~ - evidence, and that the medical journals are the only media acceptable as s conmmnication channel. Involvement in political pressure is below one's for many professional people. In my opinion, such an attitude is : out of touch with real life. We should be aware that our opponents, the tobacco industry, are experts in lobby activities and creation of political pressure. Who is going to create a counter-pressure, and tell the decision-makers the other side of the story, if not us? This does not mean that we have to become politicians. But we should realize ~ that we all are polltlcal human beings, and utilize all possible channels to make the politicians stand up and take responsibility. One thing is certain, without active political involvement,~we shall_u~r~_~eacb__~h~fi~l goaL_ ~fore, let us act, and let us act no~ T108350733

I0. ii. 12. 13. 14. BJARTFEIT Roemer R. Legislative action to combat the world smoking epidemic. Geneva, Switzerland: World Health Organization, 1982. Bjartveit K, Christie N, Holbaek-Hanssen L, et al. Influencing smoking behaviour. Geneva: International Union Against Cancer, 1969. UICC Technical Report Series, Vol. 3. Bjartvelt K. P~virkning av r~ykevaner ved restriktive tiltak. Social- medicinsk tidsskrift 1971; 48:123-127 (in Norwegian). World Health Organization. Legislative action to combat smoking around the world; a survey of existing legislation. Geneva, Switzerland: World Health Organization, 1976. WHO Expert Committee on Smoking Control. Controlling the smoking epidemic. Geneva, Switzerland: World Health Organization, "1979. Technical Report Series, No. 636. Gray, N, Daube M, eds. Guidelines for smoking control. Geneva: International Union Against Cancer, 1980. 2nd ed. UICC Technical Report Series Vol. 52. Central Bureau of Statistics. Report prepared for the National Council on Smoking and Health. Unpublished data, Oslo, 1982. Canadian Gallup Poll Ltd. Report prepared for the Canadian Cancer Society. Courtesy of David Nostbakken. Toronto, 1982. Central Bureau of Statistics. Survey on smoking habits, 4th quarter 1973. Report from the Division for Interview Surveys. Oslo, 1974. Aar~ LE, Brekke T. Health education in Norway 1979/80. Description of projects and methods. Mimeographed report. Dept. of Social Psychology, University of Bergen, 1981. Bjartveit K. How to measure effects of a governmental programme. In: Ramstr~m LM, ed. The smoking epidemic, a matter of worldwide concern. Proceedings of the Fourth World Conference on Smoking and Health. Stockholm: Almqvist & Wiksell International, 1980: 155-157. Wickstr~m B. Cigarette marketing in the third world. In: Ramstr6h LM, ed. The smoking epidemic, a matter of worldwide concern. Proceedings of the Fourth World Conference on Smoking and Health. Stockholm: Almqvist & Wiksell International, 1980: 98-105. National Council on Smoking and Health. The act relating to restrictive measuKes [qr the_mayke.t_ing_e[_~oha~co_producr.s_etc~_(Norway),--N~--14 of March 9, 1973. Oslo: National Council on Smoking and Health, 1975. Bjartveit K. The Norwegian Tobacco Act. 1977; 36: 2-9. Health Education Journal TI08350734

45 15. Bjartveit K, L~chsen PM, Aar5 LE. Controlling the epidemic: legisla- tion and restrictive measures. Can J Public Health 1981; 72: 406-412. 16. Nilsen E. S~oklng habits among children in Norway. Med 1959; 13: 5-13. Brit J Prey Soc Nilsen E. RCykevaner og opplysnlngsarbeld. Society, 1963 (in Norwegian). Oslo: Norwegian Cancer 18. Lee PN, ed. Statistics of smoking in the United Kingdom. Tobacco Research Council, 1969. London: Action on Smoking and Health. London (personal communication). General Register Office. The Reglstrar-general's statistical review of England and Wales. London: 1921-1973. Office of population censuses and surveys. Mortality statistics, cause. Reglstrar-general on deaths by cause, sex and age in England and Wales. 1974-. London: 1977-. Califano JA. Remarks to the Fourth World Conference on Smoking and Health. In: RamstrSm LM, ed. The smoking epidemic, a matter of world- wide concern. Proceedings of the Fourth World Conference on Smoking and Health. Stockholm: Almqvist & Wiksell International, 1980: 118- 122. International Herald Tribune, 1979 July 20. Young G. The politics of smoking. In: Ramstr~m LM, ed. The smoking epidemic~ a matter of worldwide concern. Proceedings of the Fourth World Conference on Smoking and Health. Stockholm: Almqvist & Wiksell International, 1980: 123-127. Raphael A. Nov 16. Tobacco barons and health reshuffle. The Observer, 1981 Willoch K. Greeting to the Fifth World Conference on Smoking and Health. The smoking epidemic can be conquered. Tobakken og Vi (Oslo) 1983; 2: 102. Commission of the European Communities. The agricultural situation in the community. 1982 Report. Brussels, lg83. Also: Personnal communication, Royal Norwegian Embassy, Brussels, 1983. TI08350735

47 PUBLIC INFORMATION PEOGRAMMES Mike Daube Senior Lecturer in Education Department of Community Medicine University of Edinburgh Edinburgh, U.K. During the Conference I had a conversation with one of the five representatives from the British-American Tobacco Company, who tried to express to me the views: (a) that there are in the tobacco industry people who are seriously concern- ed about the smoking and health problem, (b) that the tobacco industry spends much money on research, (c) that we should put greater emphas~s on persuading continuing smokers to choose lower tar brands, and (d) that although he believes the case on smoking and disease to be "statis- tically incontrovertible", he is not prepared to state publicly that smoking kills. Indeed, although when we met a few years ago, he accept- ed that smoking kills, since going to work for a tobacco company he tells me that his position has "modified". I do not think that it misrepresents my response to say that he concluded that "the gulf between us is wider than I thought". He is not alone. There are here at least ten people identifiably from the tobacco industry and their hired guns, and I would estimate at least another ten to f~fteen hiding behind home addresses, if not false moustaches. I used to think it r~ght to allow free entry to these meetings - but I am beginning to wonder if we are not a little crazy: how often do the cigarette manufacturers invite us to their planning sessions? Hard experience shows that there is no place for serious dialogue between those genuinely concerned for public health and the tobacco industry. We are adversaries, and should not be conned ~nto concession by smooth public relations men or the apparently civilised scientists employed to add a Veneer of respectability. What they say and admit privately is immaterial: it is what they do publicly that matters - and publicly they spare no effort to ensure the highest possible levels of sales for their products. Add~e2~--for--corre~pondence-:--Mr~l.-e~D-a~be~Ex~¢utlve Director, Health Promotion and Education Services, Health Depart=ent of Western Australia, 60 Beaufort Street, Perth, Western Australia 6000, Australia. TI08350736

I had the privilege of su~ning up the Fourth World Conference on S=oking and Health in Stockholm four years ago, and I said there, '%%ere do we go from here? I believe we go to war. We recognise quite clearly that this is war with a determined enemy. The tobacco industry has demonstrated in every continent that it has forfeited the right to be regarded as anything other than the opposition. The Merchants of death are the manufacturers, and we must confront them on every battle ground, whether it be health, political, social, environmental, economic, or any o~her." One of the most encouraging developments since Stockholm is that more of us have gone to war in countries ranging from Hong Kong and Austria, to China, Kuwait and Australia. Not only are we at war, but we should recall the dictum of the great military strategist, yon Clausewitz, "in such serious matters as war, those errors which spring from a spirit of benevolence are the worst". While we recognise, with WHO, that "the control of cigarette smoking could do more to improve health and prolong life in developed countries than any single action in the whole field of preventive medicine", the tobacco industry not only promotes cigarettes in developed and developing countries alike, but also denies that smoking is harmful to health, and funds people to promote doubts about the evidence. A major part of any Public Information Programme is to counter the activities of the tobacco industry~ partly for direct reduction of their impact, but also because every time we criticise the industry in public we also draw attention to the dangers of smoking. We must become involved in what Mark Worden has called "unpopular preven- tion" or, in the words of Lord Milner "If we believe a thing to be bad and we have a right to prevent it, it is our duty to try to prevent it and to damn the consequences". I shall discuss Public Information Programmes in four sections: i. Their place in the smoking control programme 2. History 3. Technique 4. Broad context A public information programme, as set out by Nigel Gray in the UICC Manual, "Guidelines for Smoking Control", is essentially the process of transmitting information on smoking to the public at large, whereas a public education programme is more specific and target oriented. There is obviously overlap, but the two re-inforce each other and one is likely to be much less effec- tive in the absence of the other. This paper deals with the media, and the ~ore ~eneral i~ues-. Public information work is just as important a function of health education as the public education programme, although traditionallyo and in some cases disastrously, health educators have received little formal training in T1077350737

pE~LIC INFOE~.ATI(]N PRDG~S 49 infor=ation activities. Two sets of categories can help to place public information progra~es within the context of health education. i. Peter Draper has set out three types of health education, to which I have added another two: Type A - Traditional health education - didactic, classsroom-oriented, =ainly biological. Type B - Health service education - often provided for the convenience of health professionals. Type C - Political/ecological health education. Type D Pseudo health education (for example, advertising of foodstuffs or pharmaceutical products purporting to be motivated solely by considerations of health). Type E - Anti-health education - such as tobacco advertising. A public information programme on smoking clearly fits into Type C - but should complement types A and B, while setting types D and E in context. The American writer, Mark Worden, has described preventive medicine as falling into two categories, "popular prevention", and "unpopular prevention". Popular prevention is familiar to us all. It involves low key education programmes, posters, leaflets, and all the cosmetic activ- ty thai'provides an easy way out for politicians frightened of taking serious action. These activities take place in isolation, not as part of the larger programme, but as something governments can point to as a kind of political fig leaf, and deliberately use as an alternative to action that might be effective. "Popular" prevention, because it causes little trouble, is popular with governments, the tobacco and allied industries, and even the public who are given the impression that some- thing is being done hut are in no way inconvenienced. Unpopular prevention, however, involves hard decisions, recognising the magnitude of the problem and the action necessary to bring it under control. It entails legislation, taking on major industry, fighting political battles. It actually has some impact on the problem, but it is unpopular and, as Worden writes~ "unfortunately, one may safely assume that un- popular prevention will remain difficult and unpopular, and popular prevention will remain attractive, enjoyable, popular~ cosmetic, and inconsequential .... " Anyone seriously concerned to reduce smoking must be active in unpopular as well as popular prevention. More than this, however, it is, perhaps, above all the role of the public information programme to affect public knowledge and attitudes to the desirability of smoking control so that ultimately-- the .... measures ~r~ty~t~h~ht of as U~popuiar become popular. If there is one factor that differentiates smoking control activity now from that of a decade ago, it is that there is broad agreement on the ~easures T108350738

required. Of course, not all the answers are known, but now there is good evidence from around the world that a combination of =easures and activities, united in a comprehensive smoking control progra~e, can succeed in reducing the scale of the problem. The seminal Norwegian report, "Influencing Smoking Behaviour", recommended as long ago as 1967 that the full programme should consist of three parts: Information/Education, Legislation, and Cessation Activities, all prosecuted simultaneously~ and with equal vigour. The responsibility for this programme rests with governments. On the basis of international experience, one can distinguish ten main areas in which they should act towards implementation of the programme. I. A ban on tobacco advertising and promotion 2. Adequately funded health education 3. Strong changing health warnings on cigarette packets and such advertisements as remain 4. a) Regular annual tax increases b) Differential taxation (higher tar cigarettes to be more expensive) o Increased provision for non-smokers: a) public places b) health premises c) other places of work 6. Ban on sale of cigarettes to children and young people Adequate information on smoking patterns and consequences 8. Establishment and progressive lowering of upper limits for emission products. 9. Assistance for those who require help in giving up smoking 10. Evaluation. A comprehensive programme including these measures has been recommended by WHO Expert Committee reports, the World Health Assembly, the UICC~ the IUAT, and many other national and international bodies, reports, and committees. Of c~urse there must be slight variations depending on local circumstances and political pragmatism, but two points are absolutely vital. First, that we do not waste time re-inventing the wheel. For several years, there has been a concensus among smoking control experts on the measures required, and this has been a major strength of the smoking and health campaign inter- nationally as distinct from, for example, those concerned with problems of alcohol abuse. The question is no longer '~hat needs to be done?" but "how -and-~when -do we--g~ ~--ddffe.~P.~Sb---f~ose new t0 s~king control policy issues, or the politically inexperienced, should think twice before seeking to change the policy objectives. Indeed, p~rhaps they should think four times: first, because there are few enough of us, and our energies should be devoted to implementation rather than repeating old debates; second, TI0~350739

PUBLIC INFORHATIOH PROGEAN/~S 51 because it is i=portant that we all sing the sa=e song, our opponents will be quick enough to exploit any apparent disunity; third, because if a battle is, as in ~any countries now, being fought and won, nothing is more foolish than to reduce and limit one's objectives: it is axiomatic a=ong chess-players that more games are lost through changing strategy mid-way than through a fault in the original strategy; and fourth because it is just possible that experts, expert committees, and international experience have actually come up with the right answers. The role of the public information programme is to communicate directly with smokers about smoking but also to promote action in all the areas in which it has been recommended. It spans all aspects of the overall programme, justifying some, furthering others, and at times depending on yet other parts for its own legitimation (for example, a dynamic exercise such as the Australian BUGA UP Campaign would not be possible without the more or less conventional school and other education programmes taking place there). The public information programme links with and overlaps the public education programme particularly, but llke public education, it is very much more than a matter of simply presenting information on mortality and morbidity and hoping that public attitudes will automatically change. A public information programme must be run creatively, recognising that it carries responsibility for implementation of other parts of the programme, and that its targets include not only the general public but specific key groups such as the medical profession, journalists, politicians, and the tobacco industry. This means, as I have said before, "If you want to do something about the smoking problem, you're in politics". The whole point of a public informa- tion programme is not simply that it presents information, but that it generates activity. A programme with no political awareness, targets, and activity is no public information programme. And this, in turn, means that the skills to be developed are in areas such as dealing with politicians and the media. The next stage is to recognise that although media and political activities should be well planned, they include by their nature much ad hoc activity and in-fighting. In his excellent book, "The Lung Goodbye - a manual of tactics for counter- acting the tobacco industry", Simon Chapman writes: "The litmus test of whether ~ smoking control intervention is worthwhile is the industry response. If they try to oppose an action, you're on the right track. If there is a deafening silence you need hardly bother. The industry makes very loud noises indeed when legislation is being proposed, but seldom rustles when education is being discussed. Indeed, the indus- try has often advocated health education as the proper sort of ,! government response... Health education must, therefore, be turned into a threat. As David Player would say, "Let's ~ke it famous". Political progress, whether through taxation, advertising bans, protection for non-smokers, or properly funded health education, is what the industry fears; and there are good grounds for believing that, in some. countries, this is already proving so effecti,:e that so~e of the longer-term education progra~es ~ay become redundant. TI03350740

52 We need ~re act~vlsts, people who know how to walk tightropes, ~o are prepared to get their hands dirty in the hard, tough business of politlcal in-fighting, because it is, above a11, through political change that the decllne in s=oklng will come_ about. Let =e stress that I do not in any way play down the importance of education progra~=:es. We need them: i) because of their present and future impact on the problem and because they legitimate our political activities. But no-one should believe that education alone is enough. HISTORY We sometimes forget how recently smoking was recognised as a major public health problem, and how much has happened since. It is only twenty-one years since the first Report of the Royal College of Physicians, and nine- teen since the US Surgeon General's series started. This is, after all, only the Fifth World Conference on Smoking and Health since the American Cancer Society invented the idea in 1967. So, just as we cannot afford to be complacent, let us not be too modest about the changes that have occurred. Despite the most formidable opposition, smoking is now recognised as a major international public health problem. There are few countries where progress has not been made, and in some it has been dramatic. In terms of legisla- tion, Norway, Finland, and Sweden have shown the way, but at least 16 coun- tries have passed leglslation to ban cigarette advertising, while 35 have health warnings. Education programmes are being given in schools. Public information programmes have been developed in every continent and in many, even most countries, with a considerable degree of ingenuity. Much progress is being made in other areas also, from tar to protection of passive smokers. But above all, the increase in sales is being halted, and in some countries there has been a notable decline. In the United Kingdom, for example, cigarette sales peaked in 1973 at 137,400 million. Each year since, there has been a fall, and the 1982 f~gure is 102,000 m~111on: that is a drop of more than 25% in a decade. Between 1972 and 1982 prevalence of smoking fell among men from 52% to 37%, and among women from 42% to 33%, wh~le another survey shows that in the two decades from 1961 to 1982, smoking in men was halved - from 72% to 36%. We have about 9 milllon ex-smokers. As in other countries, continuing smokers recognise that we want to help, not hound them, and 70% want to give up. These encouraging figures are given not to be complacent, for there are still problems and failures enough, but to emphaslse that we should take encouragement from some_ trends, that we can be successful even in a country where the opposition is formidable, and that times have changed, along with ~uch conve~ional_wisdom~_as_~ell_as__~h.e_va~idity- In a different climate. The place of smoking in society has altered fundamentally, as have smokers, and not only because more than a third of those who ever smoked have now given up. At any time, between 5-10% of smokers are giving up, and up to 20% are seriously considering doing so. It TI08350741

is no longer true that, as conventional wisdom and some research once had it, a child that smoked two cigarettes was all but certain to he a smoker for the rest of his/her life. The change in atmosphere is reflected in children, increasing numbers of whom, it appears, now take up smoking only to give it up again. There are success stories to tell. A 25% fall in ten years is dramatic indeed, virtually unbelievable for any other product. Of course, if smoking is a short-term problem, if the expectation was that all smokers would give up overnight, we have failed; but for a long-term public health problem the portents are encouraging. Public information programmes on smoking have had similar histories in most countries. The first publicity on the dangers of smoking appears almost by happenstance, and anything that follows depends on the efforts of a few isolated public health or passive smoking campaigners. Then, after a major national or international report, the government or a health agency takes the initiative in launching a publicity campaign. Expectations of success are inflated, and there is some disappointment that everyone does not give up smoking overnight. Longer-term programmes are developed, with sooner or later the recognition that this is an adversarlal problem, and must be treated as such. We have all, or most of us, experienced any number of press and poster campaigns, television and radio commercials, blockbuster shock-horror programmes, news programmes, documentaries, scientific programmes, programmes and series to help smokers give up and so on. We have all, or most of us, produced our own advertisements, posters, leaflets, badges, display materials, non-smoking days or weeks, books, articles, and the rest. Understandably, we all think that ours are the best. We all, equally understandably, at meetings like this tend to speak, as it were, from press releases - showing material produced without discussing how and whether it was delivered; presenting a campaign launch without saying how long it last- ed or whether it was later re-inforced; where there is evaluation, giving results in terms of appreciation of mater~al rather than effectiveness or efficiency of the campaign. The purpose of painting this picture is not to dampen any enthusiasm, but to emphasise: That there is now such a vast range of international experience that we cannot afford to continue repeating mistakes that have been made else- where. We have so many weapons at our disposal that we should not be dis- appointed if any single poster, programme, or leaflet seems to achieve little on its own in the short-term. One final historical point. We are here because smoking is a health problem. Work by doctors has identified the problem, and it was doctors who generated the first real action on smoking. But this does not make doctors --~h~ automat~c~choices to i~lement smoking control program~e~-, any more than physicians are the right people to carry out operations that they deem necessary. This is not an antl-med~cal view, but an assertion that imple- mentation of the program~-~ may require differing skills. TI08350742

For public information progran=nes, as for neuro-surgery or professional football, technique is vital for continued success. We must be as professional in our approach to public information progran==es as to any other aspect of our work. Lack of professionalism in both running public information programmes in general, and dealing with the ~edla or politicians in particular, should be avoided at all costs. This entails specific points, such as ensuring accuracy, never exaggerating, learnlng about different media and politicians, and how they work, pre-testing publicity material wherever possible, and so on, but also a more general point. On this sensitive issue it is the responsibility of any who are likely to deal with or be contacted by the media to ensure that they are as professional in this as in any other aspect of their work. To take an obvious example, the doctor who publishes a scientific paper on smoking should consider whether it is likely to attract media interest, and if so how this should be handled. It is simply irresponsible to publish without any thought as to the wider impact. And those who seek or are likely to gain publicity should make every effort to ensure that they are in liaison with, and complementary to, those responsible for smoking control programmes on a day-to-day basis. The same applles, of course, to contact with politicians and decision- makers. They are professionals and we should be equally professional in our approaches to them, if only for the thoroughly pragmatic reason that this will make us more effective. Apar~ from the need for professionalism, there are a number of points that can be made: No public information programme can expect to he effective in the long term unless it is carefully planned, with an eye to objectives, target groups, and, wherever possible, evaluation. We must be persistent: it is naive to imagine that we shall persuade all smokers to give up (or the industry to diversify) overnight, and equally naive to cry "failure" simply because we have not achieved all our objectives at speed. We want the tobacco industry to diversify, but there is a danger that we ourselves may diversify too much, and dissipate our effort. Organi- satlons dealing with smoking, whether health ministries or cancer socletiesj have other problems to tackle. Therefore there is the danger that, because of the number of other problems (or simply because another programme seem~ easier and meets fewer obstacles), smoking may be downgraded. We must not forget that it is in developed - and will soon be in developing - countries the largest avoidable cause of death and disease. Health education agencies, whether governmental or cancer and heart societies, for whom smoking is not the major priority, with the_lion's shar9 of resources,_ are unlikely~tp~have their_priorlties right. T10~350743

pUBLIC INFOPR!~TIO~ P~RA~ES 55 i0. There are success stories to celebrate, whether they are s{ngle-issue wins or encouraging trends. These should be publicised. We can ill afford to compete amongst ourselves. There is no need for extremism, whether in attibuting to smoking diseases it does not cause, or in giving the impression that we are against smokers, when in fact we seek to help them, and are only against smoking and those who promote it. For every argument in defense of smoking, whether at the individual level or in defense of the tobacco industry and its activities, there are counter-arguments. These are in the major reports, and are easily found and learned. Exchange of experience and information, on failures as well as successes, at international meetings is important. Running public information programmes should be fun. There is no need to be constantly downcast at the scale of the problem or of the opposi- tion. Publicity campaigns can be exciting; the more enjoyment we get, the less likely we are to appear as killjoys. Even the multi-natlonals can be amazingly inept. For example, a British company started to market two cigarette brands named "Rapier" and "Stiletto". I wrote a suitably publicised letter to the company chairman congratulating him on his choice of names for the brands and the company, embarrassed, withdrew the brands from the market. There is no reason why we should not enjoy ourselves .... There is no single perfect poster, leaflet, commercial, or programme that should be replicated universally, but it is worth looking at some of the varied approaches adopted in different cultures, with different target groups. In this section I want to discuss the size of the problem as it relates to Public Information Programmes in four areas: Size of the health problem We all know just how serious a problem smoking is, yet it is so vast that we tend to shy away from it. The British Government still tends to use ~ figure of 50,000 avoidable deaths caused annually by smoking, although even conservative estimates put the figure at around 90,000. The public knows that smoking is a health problem, but is unaware of its size: that it causes ten times more deaths annually in Britain than road accidents, more than five times as many days lost from work as s~r~ke~_that_ it_kills_.wo~em ~--~e-l~ as men. Perhaps we still do not trumpet enough just what this means; that each year more than one million people around the world die needlessly early because of smoking. The problem can be illustrated locally TI08350744

as well as nationally. Scottish ASH produced a highly successful report, "The Scottish Epidemic", ~hich broke the problem down by region, health board and parliamentary constituency so that Members of Parliament could see the consequences of smoking amongst their o%m voters, and its pre-eminence as a preventable cause of death. Smoking is not a health problem but a public health catastrophe and we should not hesitate to keep repeating this. Size of the opposition The smoking problem would not exist on this scale but for the international tobacco industry, which is dedicated to maintaining its sales at the high- est possible level, regardless of the health consequences. They are for smoking: we are against it, It is an adversarial problem; anyone who believes otherwise is naive or influenced by the industry. The industry is adept at gathering support, for example from politicians such as Ronald Reagan, who forty years after he featured in a cigarette advertisement, wrote to a tobacco farmer, "I can assure you that members of my Cabinet will be far too busy dealing with substantive issues to waste their time proselytizing against the dangers of cigarette smoking"; they gain support from the media, from journalists, from sponsored arts and sports .,. the list could be endless. Experience internationally has shown that, for the tobacco industry,"co- operation" is simply a means of mitigating the impact and potential of ant~- smoking campaigns. This means that those who accept funding from the industry in any form should be seen as part of its public relations programme, and that its defence arguments should be treated with the utmost scepticism. So we must develop and use political skills, for our opponents do. We must be tough, for our opponents are. We must be professional. We must lobby, use our political systems properly, identify the opposition, make sure that our supporters are properly mobillsed. The magnitude of the opposition would be daunting if we did not know that it can be defeated. But this will happen only if we are prepared to put our heads above the parapet, and to put their supporters under pressure. One of the great figures of pressure politics was the late Saul Alinsky, an American campaigner, who wrote "Rules for Radicals". I want to commend to you the "Rules for Power Tactics" which he drew up. It is remarkable how relevant they are to our work. Rules for Power Tactics I. Power is not only what you have but what the enemy thinks you have. 2. Never go outside the experience of your people. 3. Wherever possible go outside the experience of the enemy. 4. Make the enemy live up to their own book of rules. 5.~id~cule is-a man~s most potent weapon. 6. A good tactic is one that your people enjoy. 7. A tactic that drags on too long becomes a drag. 8. Keep the pressure on. 9. The threat is usually =ore terrifying than the thing ~tself. TI08350745

The r.ajor premise for tactics is the development of operations that will =aintain a constant pressure upon the opposition. II. If you push a negative bard and deep enough it will break through into its counterside. 12. The price of a successful attack is a constructive alternative. 13. Pick the target, freeze it, personalise it, and polarise it. Perhaps above all the final rule: "Pick the target" - governments and the tobacco industry; "freeze it" - easily done with the overwhelming evidence we have; "personalise it" - which we have not done enough, because we are simply too nice to identify publicly the presidents, chairmen, directors, and senior executives of companies such as British American Tobacco, Philip Morris, the Imperial Group, RJ Reynolds, Rothmans International, and others, as those personally responsible for hundreds of thousands of deaths and for promoting a lethal habit to children; and "polarise it" - recognise that it is an adversarial problem. There are no easy options in smoking control. iii) Size of resources required For far too long we have operated on mlniscule budgets. Governments pay lip service to the scale of the smoking problem, and then allocate a derisory level of resources to those trying to bring it under control, resources far, far less than are available to those treating the consequences. Clearly, it is not and should not be a matter of "Prevention" versus "Treatment" or "Cure". The activities are complementary. But we must stress that we are being asked to achieve miracles on resources that would be regarded as ludi- crous by those in the commercial arena, with some of whom we are in direct competition, and even when we do have useful ideas or progra~mes we cannot afford the kind of repetition and re-inforcement that are necessary. In Britain we spend about £2 million p.a. on anti-smoking publicity, compared with about £150 million p.a. spent on tobacco advertising and promotion. Given the scale of the problem, the opposition, and expenditure on other parts of the health service, we should be demanding budgets of a different order of magnitude. iv) Potential size of success The very scale of the problem brings with it opportunities. If cigarette sales in Britain have fallen by more than 25% in a decade~ a vast amount of death and diseases has been prevented. I believe, as I have written else- where~ that we should now set ourselves the targets of reducing prevalence to less than 25% in both males and females within a decade, and sales by at least a third over the same period. Once we have achieved these realistic objectives smoking will still be a major public health problem, but it will be a problem under control and demonstrably on the way out. If only to assist our public information programmes, we should be producing models to show what it means in terms of mortality and morbidity if we prevent the problem from taking off in developing countries and if we are indeed able to reduce it dramatically in developed countries. To sun=narise, public information work as part of the smoking control pro- gramme entails highly professional publicity and political activity. But all this takes place against a background of overt and covert opposition TI08350746

from the tobacco industry. Curbing their activities is the first step to creating an atmosphere in which true education can flourish. So where do we go from here? I believe that once again we go to war. Inspired by Sir George Godber and Kjell Bjartveit, we see the opposition not as an i~movable barrier, but as a challenge, and one that we ~ust meet, for if we do not, nobody else will. The main developments at previous World Conferences have been the establ~shment first of education and publicity programmes, then of pressure campaigns. At this conference we heard of campaigners taking on the industry in direct confrontation. I hope that the major development by the time of the next conference is that governments and major health agencies will recognise, take over, and claim that they invented this approach. Then, perhaps, time may indeed start running out for the tobacco industry~ and the toll of damage caused by its products. T108350747

59 SMOKIN~ IN DEVELOPING OOUNTRIZS D. Femi-Pearse, M.B, B.S. University of Lagos College of Medicine PMB 12003 Lagos, Nigeria Tobacco was introduced into Egypt and Asia by the Turks in the 14th century. Cultivation of tobacco in East, West and South Africa is attributable to the explorers of the 16th century who, on graduating from the School of Prince Henry the Navigator, sailed to all parts of the world. The earliest date of tobacco cultivation in the rest of Africa, outside Egypt, is about 1560. Taha and Ball (I), quoting from the literature, state that Crowley was the first to report on smoking in Africa in the year 1607. Calling at Sierra Leone on his way to the West Indies, he saw "tobacco growing in small patches and natives smoking it". Cigarettes reached developing countries by importation (e.g. Ethiopia and Sudan) or by local cultivation of tobacco and manufacture of cigarettes (e.g. Zimbabwe, formerly Rhodesia, and Nigeria). The cultivation of tobacco in Africa, Asia and other developing areas has been encouraged in recent years by multinational companies such as the British American Tobacco Company, Philip Morris and Rothman. This has been necessitated by the desire to avoid import duty on raw materials and to conserve scarce foreign exchange. In Nigeria, 83 and 3 acres of land were cultivated for tobacco in Ogbomosho and Zaria, respectively, in 1934. In the same year a pilot cigarette factory was established in Oshogbo. Modern factories later sprang up in Ibadan in 1937; Port Harcourt in 1956 and Zarla in 1959, producing the destructive little "cylinder" by the millions. A Green Leaf Thresher was installed in Zaria in |982. The latter is capable of handling 6000 kg per hour of green leaf and in peak season will employ 480 persons on two shifts. In Bangladesh over I00~000 acres of land is devoted to tobacco cultivation, while in Nigeria 60,000 farmers are committed to an acreage of 120,000 acres The effects of cultivating tobacco are as follows: I. Competition with cultivation of staple food crops such as rice, millet, cassava, guinea-corn, etc. 2. Displacement of necessary cash crops e.g. cotton. 37~Loss o~ tfmbdr ~hroug~, trdd-~-~-~g-~n~bu-sh--f~re~ due =o ignited~c£ga- retie stubs and promotion of erosion and sahelian migration in areas with already sparse vegetation. Tl08350748

60 In the Sokoto region of Nigeria, tobacco thrives in the flood-plains where rice would normally be expected to grow. Because tobacco provides ready cash, rice is a second choice for cultivation. The net result of such dis- placement of staple food-crops is that rice is now imported into Nigeria. In a choice between cultivation of rice and tobacco, any development economist would prefer to cultivate the former. Loss of forest reserve has resulted from clearance of bush to promote cultivation of tobacco and use of wood fuel in flue-curing of tobacco. The ecological consequences, especially in areas bordering on the desert, are favourable to desert encroachment. The advantages to the grower of tobacco can be considered: Cash returns for cultivating tobacco are better than for food crops. In the Sokoto area of Nigeria, crop value in 1977/78 was N4,270,682 or $6,406,023, which provided an annual income per head of N153 or $230. If other crops are cultivated, the ordinary farmer can make more income. 2. Tobacco farmers have generally higher status than food growers. Because tobacco growers are relatively prosperous, they tend to stay on during periods of drought, whereas other food growers tend to migrate to the urban areas. Transfer of modern skills is associated with growing tobacco. The multlnational tobacco companies take pains to teach local farmers modern methods of land preparation, ploughing, harrowing, germination of seeds, planting, use of fertiliser, etc. These skills are transferred through extension services and it is hoped a spin-off will be evident in the improved cultivation of food-crops. Good tobacco growers are best at food-growlng and the converse is also correct. 5. Scholarships are provided as incentives for the children of tobacco growers. 6. The traditional business sector is exposed to new marketing strategies and international business practice. 7. Tobacco companies provide long and short-term credits for development of farms. Despite the enlightenment generated by tobacco companies as enumerated above, the subservient role of women, particularly in the farming villages of the Third World, has not been influenced. The tragedy is that women are still the intelligent beasts of burden through whom the tobacco companies and male farmers become prosperous. This background resum6 of the advantages and disadvantages of tobacco culti- vation clarlfies the attitudes and positions that are necessary to promote or--discourage tob~cco--c~rl~a-~tlon. It--i-s--ob-ri-o~s--th~-t. in- ~h~ countries of the Third World, governments must employ the strategies of tobacco companies to sustain the growth of food crops and they must do much more than the tobacco companies in preparation of land, ploughing, and distribution of germinated seeds and fertillser. Governments must also TI08350749

~CMOKING IN I).=VELOPING COITN~I'RIES 61 offer stable economic prices for food-crops through co-operative societies, and credit facilities should be available to bona fide farmers. The fight against cultivation of tobacco cannot be won by rhetorics but by planned action. Recently, tobacco companies have introduced positive pro- grammes among tobacco growers such as the idea of block farms. Farmers are now encouraged to grow not only tobacco but also other crops, especially those related to food, in small land holdings. Several acres of such land are demarcated for growing a food crop e.g. cassava, and are cleared and prepared to accommodate new species of the food crops. For example, in 1978, the Nigerian Tobacco Company set up a block farm in Ekwotso, Bendel State for the cultivation of a new high-yielding and disease-resistant cas- sava obtained through the International Institute of Tropical Agriculture (IITA) in Ibadan. Farmers were encouraged to plant the cassava in rotation with their tobacco. This cassava matures in |l to 12 months instead of the normal 15 to 18 months and yields up to 25 to 30 tons of tubers per hectare, compared to the traditional 5 to 7 tons. 'When the harvest was in, net earnings on this block farm averaged NI000 or $1500 per hectare from tobacco and NI000 or $1500 per hectare for cassava. Since each farmer has 3 hectares - one each for tobacco and cassava and one for rotation, a farmer and his family were able to net N2000 or $3000 per annum, after providing food for themselves from their cassava." "With farmers' earnings reaching such dimensions, the drift from rural to urban areas may not only be halted eventually, but may even be reversed." (2) The tobacco industry in Nigeria, particularly in the past five years, has begun active re-afforestation programmes, since 31% of wood-fuel consump- tion is devoted to flue-curing of tobacco. Trees commonly grown are eucalyptus, teak and gmelina. It is hoped that re-afforestation will prevent erosion and desert encroachment, and provide wood-pulp for the paper industry. Governments in the Third World have been tardy about arresting the tobacco- smoking habit because of large government revenues derivable from sales and manufacture of cigarettes. In Nigeria in 1965, revenue of about £9 million sterling was collected. The figures for 1980 are quoted hereunder from the Nigerian Tobacco Company Limited Annual Report and Accounts (3). "During the Company's financial year ended 30th September 1980, out of a gross annual revenue of N95 million (or $143 million) about N34.36 million or $51.5 million was due to Government in cigarette excise, import duties on materials and in company tax. During the same period, Nb.0 million ($7.5 million) (5.3%) is being distributed to stockholders, with N2.8 million (2.9%) or ($4.2 million) retained in the business." The consumption of cigarettes is underestimated in government or commercial statistics because smuggling accounts for 25% of total cigarette consump- tion. In Nigeria, loss to Government in revenue is about N40 million or $60 Government economics would seem to be shallow. ~en the balance-sheet of income frem tobacco (excise duty, import duty, profit tax, dividend tax) is prepared aBainst the cost of importing food-crops (e.g. rice) displaced by TI0,3350750

tobacco-growing and costs of treating clgarette-lnduced diseases, pro=orion of tobacco sales ceases to he credit-worthy. Incidences of regular smoking among school children in secondary or middle schools are: Lagos, Nigeria (boys 17.5%, girls 2.7%); Accra, Ghana (boys 10%, girls 8%); Lusaka, Zambia (boys 40%, girls 4%); Gondar City, Ethiopia (boys 25%, girls 0.7%); Beijlng, China (boys 19.7%, girls 0.4%) (Table I). In Dakar, Senegal, cigarette-smoking among school-children has shown an exponential growth with 71% of boys and 52% of girls in a secondary school declaring that they have smoked or are current smokers. This population of smoker children is about the same as in a developed country such as France. Adolescent children who are non-scholars but do either menial jobs or are urchins have an even higher incidence of smoking (7). The definition of a smoker in the Dakar study is "anybody who has smoked at least one packet per week during two months and still smokes now." The high incidence of smoking among Dakar school children would therefore seem to be real. While adult smokers are the main consumers of tobacco, the habit nevertheless begins in secondary schools between the ages of I0 and 17 years (5). This is a period when young people are impressionistic, recklessly adventurous and subject to peer influence. Smoking by children assumes enormous importance with the realisation that the risk of lung cancer is greater among those who start the habit early. It may be argued that students smoke because of ignorance of the dangers of smoking. Senah (8), analysing data from Ghana, reported that only 8.5% of 12,516 survey respondents had knowledge of the product as a drug of depend- ence and addiction. Perhaps greater awareness of the dangers will reduce the number of regular smokers. In this study, only 16.3% of 705 school teachers indicated cigarettes as a drug of dependence and addiction in a list that included alcoholic beverage, marijuana, dexamphetamine, black coffee, opium, hashish, heroine and cola. The score for marijuana was high- est at 28.4%. It would seem that teachers ought to know a lot more of the dangers of cigarettes and other drugs. The curriculum of all teacher train- ing colleges should therefore have "Drugs of dependence and addiction" as a compulsory course. In Senah's study, he also asked teachers to name two drugs which they used regularly; 140 out of 450 (31.1%) admitted to being cigarette smokers. Among 331 parents, only 12.1% were aware of cigarettes as a drug of depend- ence and addiction. Forty percent (99) of 233 parents were regular smokers. Both parents and teachers recommended the following measures to combat adop- tion of the smoking habit. I. Intensification of "Drug Education" for students, teachers, parents and community at large. 2. Banning of smoking in public places. 3. Dismissal of student and teacher drug abusers. 4. Control of cigarette sales. Why do our students smoke? Studies show that students smoke for the follow- ing reasons: TI08350751

TABLE I. INCIDENCE OF SMOKING AMONG STUDENTS IN DIFFERENT COUNTRIES 1970 Kampa University 1976 Lagos Secondary Medical St 1979 Gondv Secondary la, Uganda: Students (4) BOYS Total Regular Experi- No. Smokers mental Smokers C~) C~) (33.4) , Nigeria: 1026 180 227 School Children; (17.5) (22.2) udents (5) 196 42 i00 (21.4) (51.0) r City, Ethiopia: School Students (6) 1980 Dakaz,, Senegal: School Children and Street Bo~s (7) 255 64 6 (25.1) (2.3) 260 185 (71.0) GIRLS Total Regular No. Smokers (7.0) 947 26 (2.7) 36 1 (2.8) Experi- mental Smokers (2) (5.7) 7 (19.4) 1980 Accr , Ghana: Students n Secondary School an, Teacher- Training olleges (8) 1981 Lusaka, Zambia: Unlvers ity, Training Colleges, iNursing Schools, SeconaaryiSchools (9) 1982 Belj~ng, China: Middle School Children (Ordinary Schools) (I0) 3119 311 (9.98) 1234 494 (40.0) 1396 275 (19.7) 153 (5.5) (boys & girls) 283 2 6 (0.7) (2.1) 130 68 (52~) 1097 90 (8.18) 602 24 (4.0) [394 5 (0.4)

I. To induce relaxation 2. To assist concentration in study (positive effect of smoking) 3. Because friends smoke 4. To satisfy curiosity 5. To reduce feelings of despair and hopelessness (negative effect). Teachers, on the other hand, add other reasons for students smoking, such as parentage, influence of bad friends, and easy acquisition of cigarettes. Elegbeleye and Femi-Pearse (5) showed that in Nigeria parental objection to school children smoking is quite high (75% of fathers and 95% of mothers for boys and 85% of fathers and 97% of mothers for girls). Most parents of medical students also objected. Why then is there defiance? While the number of experimental smokers may not diminish appreciably, it is possible through education on drugs to teach young people to make the right choices. Three strong influences on students smoking are parents, friends and socie- ty. Elegbeleye and Femi-Pearse showed that the smoking habits of secondary school children and university students were related to the smoking habits of their parents and friends. Studies of adolescent smoking have demon- strated that, in cases where both parents smoked, the children were more likely to smoke (11,12). Among ordinary middle school children in Beijing, China, 13.6% of students who smoked had parents who did not smoke, while the percentage of smoking students whose parents did smoke was 25.4% (i0). In Norway, the smoking rate of 15 year-old boys whose parents smoked was 67% and that of the girls 78% whereas the smoking rates of boys and girls whose parents were non- smokers were 9% and 11% repectlvely (13). In the Dakar study (7), 55% and 28% of school children were offered the first cigarette by friends and parents respectively. When asked "why did you start to smoke?" the responses were: To look llke an actor or sportsman 45% To be like a friend or parents 18% For curiosity 23% Cigarette-smoking influences the usage of other drugs of addiction. Haworth (9) in Zambia found that smoking tobacco and cannabis were related, in that among 15% of males and 37% of females who had never touched alcohol or smoked tobacco, only 2% had used cannabis, whereas 35% of those who had ever smoked tobacco, 13% of those who had ever taken alcohol and 55% of those who had both smoked tobacco and taken alcohol, used cannabis. Among_adu!_ts~he__~lJ.o~ing__incide_nc_es__hayg__been__reported for ciEarette- smoking: 31.5% among University students in Uganda (4), 51.5% among 35 to 54-year old Africans in a Guyana village (14) and 41% of Nigerian men in Lagos who were over 20 years old (15). In 1974, Sofowora (16) in Nigeria surveyed smoking among males aged 15 years and above in rural and urban TI08350753

communities. Of 1544 rural parsons questioned, 2.2% smoked, -,~ile 7.4% of 659 urban men smoked in the city of Ibadan. In Asia, statistics from Bangladesh, China, India, Pakistan, the Philippines and Sri Lanka show that smoking is mainly a preoccupation of men, with very few women adopting the habit. The same is true of Africa and probably the Carribean and South America. However, Nepal has a different pattern in that, in the remote and very poor rural hilly areas, the prevalence of smok- ing is up to 84.7% in men and 71.5% in women of 21 years and above. Arya and Bennett (4) highlighted increased incidence of smoking among Indian university students in Uganda compared with Calcutta students. Reasons adduced include better financial resources of the Ugandan Indian and probably increased permissiveness. Cohen (17) described the situation in Bangladesh. In I0 to 15 years, ciga- rette consumption has more than doubled. Cancer of the lung is already the third commonest cancer among males, whereas in Nigeria it occupies the thirteenth position. The Ibadan Cancer Registry for 1960 to 1966 reveals among males: lymphosarcoma 131, hepatoma 105, leukemia 57, gastric carci- noma 55, Hodgkln~s disease 39~ cancer of prostate 38~ cancers of connective tissue 24, salivary gland 17, bladder 16, bone 15, pancreas 14, nose and sinuses 14, kidney 12, bronchus 12. An astonishing relevation is that lung cancer due to cigarette-smoking is now the commonest cancer in males in Pakistan and the Philippines. Femi-Pearse et al. in 1973 (15) showed that among Nigerians the prevalence of morning phlegm was significantly higher in smokers than non-smokers (P<0.01) while no significant difference could be found with other respira- tory symptoms such as morning cough, persistent cough and persistent phlegm. Cookson and Mataka (18), in a questionnaire survey among 9768 subjects aged 5 years and above in Rhodesia (now Zimbabwe), reported the overall prevalence of chronic bronchitis as 1.12%. These findings relate to the fact that most Africans are, as yet, light smokers. All the pathologic consequences of cigarette-smoking occurring in Europe and North America occur in the Third World, for example lung and oesophageal cancer, chronic bronchitis, emphysema and coronary artery disease. They will not be discussed further here. The incidences of these diseases gener- ally follow the cigarette consumption rate. There are, however, no studies on high and low yield cigarettes as well as passive smoking or "side-stream" smoke. Mention will now be made on some scanty and isolated reports from developing countries. Ghawabi (19), ~iting on workers exposed simultaneously to jute and hemp, showed increased airway obstruction after a work-shift. Also smokers and those with chronic bronchitis had greater reductions in F.E.V. values at the end of the work-shift. With regard to duodenal ulcer, Lewis and Aderoju (20) found dietary agents such as chillies and cassava grains to be major precipitants, but they TI0~350754

FErMI-PEARSE considered aspirin, alcohol and cigarette s:~king contributed to the worsen- ing of the condition. Of interest is the black-fat tobacco, inhalation of which leads to lipoid pneumonia and pul=onary fibrosis in Guyana. This is tobacco to which petroleum jelly (e.g. Vaseline) has been added for flavouring and hu=ectant purposes. Inhalation of this tobacco smoke results in droplets of oil reach- ing the lungs, thus causing a diffuse form of lipoid pneumonia (21,22). Cohen (17) points out that the effects of cigarette expenditure on nutri- tional status. Households with marginal incomes are likely to divert money from nutritional foods. He calculates that smoking only 5 cigarettes a day in a poor household in Bangladesh might lead to a monthly dietary deficit of 8000 calories (33.5 l~). Enhanced metabolism of antipyrine among cigarette smokers has been reported by Uppal et al. (23). It is conjectured that other drugs may have their metabolism altered by cigarette-smoking. Gupta and Pindbors (24) studied the incidence rate of leukoplakia in Kerala State. The annual incidence rate per i000 adults was 2.1 for males and 1.5 for females. The rate was highest in the mixed tobacco habits group. The rate of malignant transformation was also highest among leukoplaklas asso- ciated with tobacco-chewing habits. In most African and Asian countries at least a third of practising doctors smoke, thus reducing the credibility of anti-smoking efforts. In summary, a series of actions that have been taken with success, can be repeated in the Third World. By the individual: Stopping of smoking by parents, health professionals and workers, and other opinion molders - "example is better than precept". By so.cietal ~roups: Stopping of smoking at meetings and social gatherings. All schools must teach health education to students and teachers alike, including the effects of smoking on health. By government: Warning state=ent on packets and on individual cigarette cylinders. Ban on smoking in public utilities, buses, trains, cinemas, airports, stadia. Ban on direct advertising in media and on bill-boards. Ban on indirect advertising e.g. sponsorship of sports, T.V. quizes, game_s, etc. TI0~350755

Printing of tar and nicotine contents on packet. Imposition of limits on tar and nicotine contents. Ban on cigarette sales on 'smokeless' days (e.g. Dry days in India). Finally all nations of the Third World must examine critically the relevance and desirability of imported life-styles such as cigarette-smoking. The human, material and social costs of the pandemic of cigarette smoking cannot be justified. Cigarette-induced diseases will further overburden the fragile health services and destroy them in the Third World. In our bid to halt the habit of cigarette-smoking, role modelling assumes prime importance. All health educators such as nurses, doctors, teachers, civil servants and all educated people must provide an example in healthy living. In order to facilitate roles of different members of the community in positive health education, the knowledge base of all must be improved such that young people~ patients and susceptible individuals can be motivat- ed and reinforced to stop and control the habit of cigarette-smoking. To this end, there must be total mobilisation and commitment of the mass media. The role conflicts of teachers and nurses who smoke must be high- lighted. The exponential growth of the habit of clgarette-smoking can he controlled only when government, tobacco companies and the populace subscribe to a moratorium to reduce the scourge of tobacco smoke. i. Taha A, Ball K. Smoking and Africa: the coming epidemic. B Med J 1980; 2: 991-993. 2. Nigerian Tobacco Company Ltd. Annual Report and Accounts. 1979: I. 3. Nigerian Tobacco Company Ltd. Annual Report and Accounts. 1980. 4. Arya OP, Bennett FJ. Smoking among university students in Uganda. Aft Med J 1969; 47; 18-28. East 5. Elegbeleye O0, Femi-Pearse D. Incidence and variables contributing to onset of cigarette smoking among secondary school children and medical students in Lagos, Nigeria. Br J Prey Soc Med 1976; 30: 66-70. 6. Ahmed Z, Abuhay M. The prevalence of cigarette smoking among secondary school children in Gondar City, Ethiopia. Ethiop Med J 1979; 17: 41-47. Wone I, Koate P, de Lauture H. optique de sante de co==~unaut6. 573-579. La lutte con=re le tabagisme dans une Medicine d'Afrique Noire 1980; 27: T[08350756

Senah AK. A study of problems associated with the use of drugs and drug education in Ghana. U.N.E.S.C.O. Distribution Limited, 1980. 9. Haworth A. Study of smoking among students in Zambia. Paper presented to a national seminar on cardiovascular diseases, Lusaka 1981 Oct 16-17. I0. Gong-shao Y, Wang-sheng L. Cigarette smoking among Beijing high- scholars. Chin Med J 1982; 95(2): 95-100. II. Cartwright A, Thompson JG. Young smokers: an attitude study among school-children touching also on parental influence. Br J Prey Soc Med 1960; 14: 28-34. 12. Salber EJ, MacMahon B. Cigarette smoking among high school students related to social class and parental smoking habits. Am J Pub Health 1961; 51: 1780-1789. 13. Berglund EL et el. Smoking habits of Norwegian school children. In: Directory of on-going research in smoking and health, 1978. Maryland: Dept. of Health, Education & Welfere, 1978: 94. 14. Miller GJ, Ashcroft MT. A community survey of respiratory disease among East Indian and African adults in Guyana. Thorax 1971; 26: 331-338. 15. Femi-Pearse D, Adeniyi-Jones A, Oke AB. Respiratory symptoms and their relationship to cigarette smoking, dusty occupations and domestic air polution: Studies in a random sample of an urban African population. West Aft Med J 1973; 22: 57-63. 16. Sofowora EO. Personal Communication 1974. 17. Cohen N. Smoking, health and survival: Prospects in Bangladesh. Lancet 1981; i: 1090-1093. 18. Cookson JB, Mataka G. Prevalence of chronic bronchitis in Rhodesian Africans. Thorax 1978; 33: 328-334. 19. El Ghawabi SH. Respiratory function and symptoms in workers exposed simultaneously to jute and hemp. Br J Ind Med 1978; 35: 16-20. 20. Lewis EA, Aderoju EA. Factors in the aetlology of chronic duodenal ulcer in Ibadan. Top Geogr Med 1978; 30: 75-79. 21. Miller GJ, Beadnell HMSG, Ashcroft MT. Diffuse pulmonary fibrosis and black-fat tobacco smoking in Guyana. Lancet 1968; 2, 259-260. 22. Miller GJ, Ashcroft MT, Beadnell HMSG, Wagner JC, Pepys J. The lipold pneumonia of black-fat tobacco smokers in Guyana. Quarterly J Med N.S. 1971; 40: 457-470. Uppal R, Garg SK, metabolism in cigarette smokers in an Indian population. Int J Clin Pharmacol Ther Toxicol 1980; 18(6): 269-271. Tl08350757

SNOKI-NG I1q DEVELOPL~G ~0~i'/'RIES 69 24. Gupta PC, Pindhors JJ. Chewing and sm=king habits ~n relation to precancer and oral cancer. J Cancer Res Clin Onto] 1981; 99(1-2): 35-39. TI0335075~,

71 PRESENTATION TO ~HE I~lV~H ~0RLD COltl~RENCE O~ b-~KING Ah~ HEALTH Sir George Godber 21 Almoner's Avenue Cambridge CBI 4NE The World Conferences began in New York at the instance of the American Cancer Society 16 years ago, and have continued at four yearly intervals since. I have been privileged to attend them all and honoured above my deserts by speaking at plenary sessions of each. The conferences have been unusual in their provenance~ for none has been promoted by a government or a national ~edical organisation. We have come together from many different ba=kgrounds and with various supports, because we know what harm smoking does and want to see that harm reduced. In 1967, some of those attending may have thought there was still a case to prove, though I recall that even the television technicians had put out their cigarettes before Robert Kennedy was half way through his remarkable speech at the opening session. I do not imagine there is anyone here - even the odd agent from the industry - who doubts still that smoking is the largest single avoidable cause of ill-health and premature death in the industrialised world today. Our problem is not whether, but how, we should persuade smokers to stop and others to refrain from starting. It is the active efforts of the promoters of smoking that has made it so difficult for the promoters of health. Yet 1967 is only half way back to the ti~e when we had the first unequivocal proof of the causal relationship of smoking to lung cancer. I remember that Dr. 0chsner, who had suggested there might be such a link thirty years earlier still, was at that first conference. Richard Doll and Austin Bradford Hill in Britain and Wynder and Graham in the U.S. first gave us proof in 1950. By 1967 Doll and Hill had shown not only that smoking causes far more illness and death from other diseases than from cancer, but also that stopping smoking reverses the effect. Hammond and Horn had reported on a far larger study in the US. Hill, Hammond and Horn were all at that first conference, which was chaired by Luther Terry and included many others, such as Fletcher and Evang, who have contributed so ~uch to the base of our cam- paign. The epidemic of lung cancer, in Britain alone, has cost some three-quarters of a million lives since we have known its origin and how to stop it, and that epidemic was only half way to its present peak 30 years ago. Future generations will be aghast that so little had been done to stop it in the first decade or indeed by the time of the first conference and what will they think of the last 17 years. The most that health ministries were doing in the 1950s was some occasiona! propaganda which had little and transitory effect on most people, though the medical profession itself responded. The indus_~y was ~I/_~oo_al~_~_~o_~he_thr~a~_t.o_i~_s_ms.rket_and_ra~ increased TI08350759

72 GOD~ER its sales promotion. We were content to tell the public on a take it or leave it basis. Indeed, that is just the line so~a politicians take now, because they maintain that people must be free to choose - and so they =ust, provided it is an informed choice. That choice is hopelessly prejudiced if it has to be made against the constant pressure of intensive sales promo- tion, always presenting smoking as a sociable, attractive activity of normal men and women in pleasant circumstances. By now we have come to realise that smoking is essentially a form of addiction, which is cleverly rein- forced by sales promotion. Two events of the 1960s changed public attitudes. In 1962 the Royal College of Physicians of London published their report on Smoking and Health, the result of a three-year study by a committee chaired by Robert Platt and with Charles Fletcher, one of the ablest communicators in British medicine and still President of ASH, as its secretary. That report had a tremendous impact and it sold more copies in North America even than in Britain. Like the British Medical Research Council, an earlier Surgeon General had already stated in 1957 that the relation of smoking to lung cancer was causal, but in 1963 Luther Terry appointed a committee which made an even more extensive appraisal, with wide consultation, and completed its report in little over a year. After those two reports there could no longer be any real doubt of the enormity of smoking as a man-made threat to health. Since then the literature providing information on every detail of the smoking menace has multiplied till it fills a modest book from the Technical Information Center of the U.S. Office on Smoking and Health every two months. The full report produced by the Office before the Stockholm conference contained over 1200 pages, and by last October the Center had listed 35,000 items of evidence. There have been yearly reports from the Surgeon General since 1964, two more reports from the Royal College and reports in most industrialised countries from government or other committees all presenting broadly the same conclu- sions. Smoking is the largest, single, avoidable threat to health in the industrialised world today and the cigarette the most lethal instrument devised by man for peace time use. Many books have been written, from Harold Diehl's semi-popular "The Smoking Disease" 20 years ago to Bobbie Jacobson's 'The Ladykillers' last year, and sober analyses within the last year such as Ashton and Stepney's 'Smoking: Psychology and Pharmacology' or the review of 'Legislative Action to Combat the World Smoking Epidemic' done by Ruth Roemer for WHO. Biomedical scien- tists, psychologists, sociologists and statisticians have built up a moun- tain of evidence, against which there has been no serious counter. We are left with the problem, not so much of deciding on the right action but rather how to ensure it is taken. Looking back over 33 years one can recognise three phases in the cam- paign, each lasting roughly a decade. First there was the period of prov- ing the case, ending with the first major reports of 1962 and 1964. Second was the period of seeking ways of convincing the public and governments, while __tb~e__indus_t_t.v w.as__tJty~ng__to~_~o~_e~__i~s ~a~k~t ~ de~e~op_ing~a_less dangerous product. Third came the period of more intensive study of the factors which make escape from the habit so difficult, leading to the re- cognition of the nature of nicotine addiction, while the the industry seemed finally to accept that smoking cannot be made safe and the promoters of TI08350760

pK~'TA~I(L~I 73 sr_oklng ~ast simply brazen it out regardless of the harm that results. After Stockholm, an industry observer wrote "the social acceptability will be the central battleground on which in the long run our case will he lost or won". Apparently the industry's case is that the harm to health can be made to appear socially acceptable in return for the gratification of an addiction. Is that different, in kind, from the pushing of some other addictive drug - except that the law forbids promotion of other addictive drugs? Why should society all~w the pushers to continue? It is probable that, in the first phase, the tobacco interest simply treated the health campaign as a 'scare' which would die down. In the second phase, they took action which they hoped would minimlse the risk, and partly because of the ant~-smoking campaign, partly because present smoking mate- rials may be less carcinogenic, the damage now occurring may be a little less than it would otherwise have been, as Doll himself has pointed out. However, it is still enormous and growing. In the third ~hase, it has become clear that the only possible long term objective for those concerned for health must be an essentially non-smoking society, a world where smoking continues only among consenting adults in private. We need a world-wide reaction against a world-wide threat, for smokin~ promotion is now going on in its least inhibited form in less developed countries. W~O came onto the scene at the end of 1969 when first the European and then the American Regional Committees passed resolutions in favour of action by the Organization. The Executive Board took up the cause and asked the Director General to report to the Assembly in 1970. Dr. Candau, whose recent death saddened us all, invited Charles Fletcher and Dan Horn to write the report, which is still a model statement of the primary case. In May, the Assembly passed a resolution proposed by Uruguay, Uganda and the United Kingdom calling for action by member states. The second world conference took place in London a year later and in 1974 WHO's first Expert Committee reported in time for the third world conference in New York in 1975, propos- ing national comprehensive campaigns on a basis that the conference endorsed and commended to all Health Ministers. Four years later, a second Expert Committee reviewed and extended the measures proposed, in a report on 'Controlling the Smoking Epidemic' and expressed great concern about the uninhibited way in which smoking was being promoted in less developed countries. The third Expert Committee has been specially concerned with action on this problem, involving as it does economic and agricultural changes which will not be easy. There can be no doubt about the need for an international approach. From a very early stage, work in one country has been invaluable to others, as they have taken up the cause. Ruth Roemer's report, nonetheless, reveals the un- readiness of most countries to r~ove politically as far as the requirements of health promotion dictate. It is governments not just health ministries that ~*~st move. The smoking problem is not the only example of this kind of failure. Other international agencies have given some help - UNCTAD and FAO were represented at the Expert Committees though they have found more defin- itive action difficult. The UICC came rather late onto the scene, but has been very active in recent years, sponsoring regional meetings and national Workshops, as well as producing two excellent reports. A new generation of ~cr~ve health p~d~_~t~r~ arlsen ~Jn~d--t~--c~-l---~a'v~nce--~-i-nce-S-~ck~>!~-~' in improved understanding of the complexities of persuading people to chan~e T108350761

their life-style by giving up a behaviour that has become important to them and is reinforced by the addictive effects of nicotine. Hany countries have had their own independent anti-smoking organlsatlons for a decade - ASH in Britain and the USA, the Statens Tobakkskaderad in Norway, the Swedish National Smoking and Health Association, the Canadian Council and the Anti- Cancer Council of Victoria for example. Some Ministers have been prepared to speak with real force, as did Joe Califano, George Young and Hedda LindahI at Stockholm, ~hough, having done so, they do ~ot always se~ to be left long in their posts which gave the,. the opportunity. Halfdan Hahler, as Director General of WHO, has shown his usual forth-rightness. WHO has set up its own International Clearinghouse of Information. The movement against smoking has become truly international and must in the long run prove irresistible - our trouble is that the run has already been too long. The annual toll of premature deaths is of the order of 400,000 in the USA and UK alone. In the world, it must run into millions and many of those deaths occur in relatively young people so that the years of productive life lost each year are many millions. In a way, those premature deaths are of rather less concern than the years of ill-health that precede them. Some may remember a television film showing a cowboy who had to ride with an oxygen cylinder on his saddle because of emphysema due to smoking. Does anyone ever ask how that film came to be suppresssed? The short answer is that it was too commercially damaging to be allowed to survive. It was too true. Sixteen years ago I wondered just how useful a world conference could be. Now with the fifth conference, I have no doubt that the series has helped to give the campaign against smoking an impetus it would not other- .wise have had. But how many more is the world condemned to need? We have made some real gains in reducing mortality in younger men, but women are now exposed much more to the damaging effects of smoking than they were 30 years ago and the result in rising cancer deaths is all too apparent. More- over the danger to the fetus from smoking by pregnant women is now known and all too little has been done to reduce it. In the 16 years since the first conference, the people of our countries must have lost many millions of years of potential working lives because we have not succeeded to the extent we should have done. In recent years there has been a mounting threat to the less developed, less healthy and less affluent countries of the world, promoted by just those multi-national conglomerates whose advance we are slowly containing at home. It is this last development which exposes to the full the ~in support of the continued prevalence of the smoking disease and explains the remit given to ~O's third Expert Committee. The commercial interests show no scruple about promoting a habit with the devastating consequences we know all too well and of which they cannot be ignorant. The ~eveloped countries have begun to assert some control over promotion of tobacco- limited and ambivalent as governmental action has been - and the industry has used every endeavor to circumvent control, even in countries like Norway where forceful laws have been enacted. Politicians ~n some countries, like my own, have been so misguided as to accept inept and futile voluntary agreements about direct promotion which will never be et[ect~v~e so ~ong as ~e indlrec~ and supposedly [.u=cent ~= ~=~ ~ pr~tlon of the Arts and, ~st ironical of all, S~rts are left o~n. ~ose agree~nts would not ~ concluded if the co~rcial interests really believed that the result ~uld ~ the end of their coerce. ~e ulti~te cynicism in Britain was a Trust Fu~ for hea1~h pro~ion not to be used T108350762

against smoking, the m~jor danger. Do ~overnments have a secret reservation of their own that they can seem to fight for our cause, so long as they do not actually win? I ~e~ a Health ~inister ~ce ~o said ~ha~ we could ho~ £or an e~ to s~king. Bu~ surely all og us here do s~king as a soclally acceptable activity, ~a~ever addicted s~kers have do in private. ~ ~hat is not our objective ~ are simply no~ £aclng up ~he realities. For 30 years people such as Hammond, Born, Bradford-Hill, Doll, Auerbach, Fletcher, Diehl, McKennell and Russell have worked to build up an over- whelming case - naming only some of those I have known personally. I remem- ber many of the delegates to the 23rd World ~ealth Assembly who voted to bring WRO into the fight, and the first two Expert Committees. It is a new generation including such as Gray, Bjartveit, Daube, Ram, from, Crofton, Loransky, Wake, Forbes and Best, that now enlarges it and will, in the end, ensure effective action which will remove the need for conferences like this. In the last year four books have presented the strategy most clearly to me: Ashton and Stepney's from Britain, Doll and Peto's "Causes of Cancer", four chapters in the 1982 Annual Review of Public Heal~h and the report of the Task Force of the Ontario Council of ~ealth, 'A Need for Balance'. At the end of the day governments must make this campaign effective and it maybe WIiO and this conference that will shame them into doing it. It will be no new thing in the health field that, where Scandinavia has led, the rest of us follow. Row many more times must we re-state the formula? It has been set out by the last two conferences and the WHO Expert Committees and now again by the Ontario Task Force, each time with a wealth of added detail. But the essen- tials remain the same. Broadly they are; stop commercial promotion of tobacco products; limit smoking in public places; increase the cost of smok- ing by progressive taxation on tobacco products; reduce the harmfulness of smoking materials and above all improve education for health, especially of children. Let us stop deceiving ourselves by half measures and make a serious attempt to realise the target Sweden once set itself of a non- smoking generation, and make it soon. Let no government hide behind the excuses that workers in the tobacco industry - producers, processors or sales force - will lose their incomes, or that taxes will have to be levied in different ways. There are other crops and other industries that might be far more beneficial to those people. We do not ask for laws a~ainst smokers, but a~ainst the promoters and reinforcers of tobacco addictiou. If we in the industrialised world are beginning to win - and make no doubt we are - then it is the manifest duty of governments and international agencies to complete that victory within years, not decades, for all countries. Let that be the clear message to them. T108350763

71 Higel Gray, A.M., M.B., B.S., F.R.A.C.P., F.A.C.M.A. Director, Anti-Cancer Council of Victoria 90 Jolimont Streeet gsst l~Ibour~e Victoria 3002 Australia This paper will discuss the social and economic implications of tobacco use. ~he second half of the paper will discuss the strategic and policy implications arising from the problem and look at the solution and things that -,,st be done in order to change the present situation. SOCIAL IMPLICATIONS What are the social implications of tobacco use? SOCIAL COSTS I. It is a national drug addiction 2. It kills people 3. It makes people sick 4. It increases absenteeism 5. It reduces productivity 6. Sick or dead breadwinners create disruption of family life 7. It costs money; personal, national health, national insurance 8. It Pollutes, Polarlses, Odorlses and Offends (PPO0) SOCIAL BENEFITS I. Smoking relieves withdrawal symptoms in addicts. Of course, most of the social implications are for smokers. The PPO0 effects are on non-smokers who provide a real strategic resource for the reduction of smoking rates through personal pressure. BGON~IlC IMPL ICATIORS We must now examine the economic implications of tobacco use. The claims of the tobacco industry and those working against smoking might be somewhat at variance. For convenience Australian figures will be used. T108350764

78 ~ In Australia, the industry might claim that tobacco use creates "benefits" shown in Table 1 (t). TABLE 1. "BENEFITS" OF HAVII~G A TOBACCO INDUSTRY IN AUSTRALIA IN 1980 the Tax ~come Wages Exports Advertisir~ Dividends $703,240,000 $173,940,000 $ 5,850,000 $ 15,000,000 $ 17,260,000 Total 'Benefit" $915,290,000 Employment 36,945 people Source: Egger (I) Whereas I could reasonably claim that tobacco use leads to the costs shown in Table 2. TABLE 2. COSTS OF TOBACCO USE AUSTRALIA 1980 High estimate Productivity loss - illness & death $532,350,000 Medical & Pension costs $143,970,000 Non-medical costs $213,230,000 Low estimate $557,420,000 $101,970,000 $206,090,000 TOTAL $989,550,000 $865,480,000 Source: Egger (i) This sort of sum can be produced in any developed country. As usual, the "costs" are not very different from the "benefits" if we consider only money. Tobacco clearly costs a lot in reel money. It also ~enerates a lot of so called "benefits", some of which are ~T real money or real benefits - they are transfer payments from one part of the community to another. The "benefits" do include employment; which is balanced by death, d~sease, dis- ability and lost productivity. They do include substantial tax i~come and profits; which is balanced by the medical costs. The cost side of the equa- transfer of mo~ey, taxes aad employment. No resources are created. Bo use- ful product produced. T108350765

Ne m~st be wary of being drawn too far into this detailed financial argument. The reason for reducing tobacco usage is disease ar~l misery. It is uot money. There is no case in favour of promoting a product which does only harm and no good. The fact that people are employed and paid for this is no m~re a justification for continued promotion of smoking than is the employment and payment which arises from production and m~rketing of heroin. Tobacco differs from heroin only in that its use was widespread and ~r~d~tiot~al in society lor~ before it was knowu to be harmful. It is a legitimate product only because of history, and in most developed countries it is illegitimate, i.e. illegal, to sell tobacco to children. In sun~ary therefore, there is too much tobacco use. We need a global program for a reduction ~n national and internat~onai tobacco use. In con- sidering this problem we should recognize that there are tw~ different parts of the world and they have two different types of problem. The developed world has one problem. In the developing world there are two problems. The situation is as follows: RICH COUNTRIES I. The manufactured cigarette DEVELOPING COUNTRIES I. The manufactured cigarette 2. Indigenous smoking/chewing tobacco products DEVIg~ PKD COUNTRIES The problem in developed countries is the manufactured cigarette. In reality, there is not much difference in tobacco usage patterns between the affluent countries. Some of them have taken stronger legislative steps and some of them have mounted larger educational programs for longer. Some of them have cultural backgrounds in which tobacco use has been more deeply entrenched than others. By and large, however, only in Scandinavia and, more recently, in France have we seen comprehensive programs directed towards tobacco use (2). Despite the dishonest interpretations, promoted internationally by the tobacco industry, of the results of these programs (3), it is possible to see that comprehensive programs are successful and gradually they are whittling away at the problem of cigarette smoking. A well established example is Norway (4). Table 3 for the total adult population, and Table 4 for the vulnerable younger population, show a real decline in smoking rates as measured over a decade. T108:350766

80 ~ 1973 I974 Tobacco Act 1975 I976 1977 1978 1979 1980 1981 1982 Source: PER CER~f DAILY SMOKERS - I~Oi~AY Aged 16-74 Hales Females 51 32 53 32 48 33 49 32 44 30 45 31 43 33 42 30 40 31 40 34 Central Bureau of Statistics, 12/82 TABLE 4. 1973 1974 Tobacco Act 1975 1976 1977 1978 ]979 1980 1981 Source: PER CENT DAILY SMOKERS - NORWAY Aged 16-24 Males Females 44 42 47 43 38 39 44 ~+0 36 37 39 36 33 37 35 36 34 33 Central Bureau of Statistics, 12/82 Of course, the tobacco industry claims the Norwegian program has not worked. This is probably not an accidental mlsconception~ b~ a deliberate commercial deception. It is also nonsense. We should not allow the commer- cial perceptions of the tobacco industry to cancel one of the conclusions which will become progressively more obvious during this conference. Where active programs are pursued, smoking rates are fall~ng, T108350767

SOCIAL ~D ~X~K~4IC II~LICATI~S 81 TABLE 5. IlqTERNATIOHAL S~40KII~ RA~'ES Male Female Count ry Smokers Smokers Year % % America 37 28 1979 UK 38 33 1983 Australia 38 31 1983 Japan 73 15 1979 Canada 39 30 1979 Sweden 31 30 1982 Table 5 shows rates published in the world literature. Ten to fifteen years ago in the US, UK, Australia and Canada, male smoking rates were 10% - 15% higher (5-9). Rates for Austria and France have also fallen significantly (10), POLICY ISSUES The things that need to be done in the western world can be summarised under three headings. I. To persuade/help smokers to give up 2. To lower tar, nicotine, etc. 3. To reduce recruitment Smoking is a class phenomenon, on which is overlaid a cultural component. To design cessation programs, more detailed analysis is needed of smoking rates in the many subpopulatlons in our societies. To illustrate this, there is a study (Table 6) done in Australia comparing a middle class suburb, Hawthorn, with a working class suburb, Richmond (II). Although it is an old study from 1974, the results are probably still true. TABLE 6. SMOKING RATES. AUST~RALIA - HAWTHORN/RIChMOND SURVEY, 1974 Male Female Smokers Smokers Hawthorn 43 30 Richmond 54 33 Source: Survey by mass X-ray service, Victoria Table 7 shows the percentage of smokers by ethnic background in the two combined suburban populations. TI08350768

82 glAY TABLE 7. ~I1~ RATES BY COUNTRY BY BIRTH. RICHMOHD/HANTHORH SUKVEY, 1974 Country Male Female S moke r s Smoke r s % % Australia 46.1 33.3 New Zealand 45.7 45.8 United Kingdom 51.8 38,4 Greece 53.1 14.2 Italy 55.9 16.1 Yugoslavia 53,1 23.1 Egypt 61.2 40.00 Source: Survey by Mass X-ray Service, Victoria However, if the results for the two suburbs are separated, the differences based on class, income and ethnic background are even more impressive (Table 8). TABLE 8. SMOKING ~ATES BY COUNTRY BY BIRTH. RICHMOND/HAWTHORN SURVEY, 1974 Country Hales Females Hawthorn Richmond Hawthorn Richmond Smokers Smokers Smokers Smokers % % % % Australia 41.1 52.0 30.1 37.5 New Zealand 46.4 44.4 45.1 48.0 ~ 45.3 59.7 34.4 44.6 Greece 50.4 54.0 I!,! !5.5 Italy 53.8 57.1 16,4 16.0 Yugoslavia 53.3 53.0 22.0 23.3 Egypt 54.3 62.7 53.1 37.3 Source: Mass X-ray service, Victoria The only inconsistency in this table relates to the migrant women. The percentage of non-smokers here may be exaggerated by the language problem. A random sample taken in 1980 shows the effect of income (Table 9) and occupation (Table I0) on a national basis (7). These are male rates only. ~he female rates are a little less striking. T108350769

TABLE 9. MALE SMOKING RATES BY II~CO~. AIISTRALIA, 1980 $ Income Smokers Ex-Smokers ]~ever Smoked % % % Under 8,000 39.4 28.0 30.7 8,000-12,000 44.5 18.3 3~.6 12, O00-20,000 39. i 21.5 36.2 20,000 & over 31.7 24.2 40.0 Source: Hill and Gray (7) Apart from the very poor people, the rates decrease with affluence. TABLE I0. ~LE SMOKING RATES BY OCCUPATION. AUSTRALIA, 1980 Occupation Smokers Ex-Smokers Never Smoked % % % Lower Blue 47.3 22.0 28.4 Upper Blue 45.4 18.4 33.9 Lower White 36.2 22.7 38.9 Upper White 30.2 25.7 39.5 Source: Hill and Gray (7) This picture can be observed in virtually any developed country. Hence the tarset populations in any of the rich countries can now be set out in some detail. They are: poor people, blue collar workers, and migrant groups. There is nothing surprising in this. Every affluent or semi-affluent coun- try has something similar to show. The detailed strategy for a rich country is shown in Figure i. FIGURE I. RICH COUNTRIES STRATEGIES FOR A MATURE M_~.RKET I. Ban promotion 2. Education programs 3. Cessation programs 4. Remove grower incentives 5. Taxat~q~ - remove the image - targeted and measured - small groups for hard core - simple techniques for mass use - reduce production surplus - especially high tar brands 6. Reduce tar, nicotine, etc. by regulation ar~ taxes 7. Keep measuring T108350770

The fact that anti-smoking progress is so good in the west has one very important economic implication, not for the rich, but for the poorer coun- tries. The tobacco industry no~ has only two options; to diversify and sell less tobacco or to look for uew markets in the developing world. Unfortunately, they have chosen the latter. DEVELO~IR~ In the developing countries there are two problems. The first problem is the same as in rich countries, the manufactured cigarette, although it is not yet as great a problem in the developinE countries. The tobacco industry and its mendicant friend, the advertising industry, are attempting to transfer the cigarette smoking problem to the developing world. That is, they are attempting to transfer the machine-made cigarette itself, with its attendant promotion and marketing. The result will be the recruitment of established smokers, and a new generation of adolescent smokers, away from their established habit to the habit of tobacco usage in the easy and casual western fashion. This is likely to lead to an increase in total tobacco consumption. Our first objective, therefore, is to resist the transfer of western smoking patterns to the developing world. The strategies to do this are well developed and much the same in India as they are in America. The order of priority is, however, very different and is summarised in Figure 2. FIGURE 2. DEVELOPING COUNTRIES - POLICY OBJECTIVES i. Ban all forms of promotion of tobacco 2. Tax cigarettes heavily 3. Regulate to lower tar content 4. Prohibit sales to and by children 5. Provide health warnings and tar content on packets 6. Begi~ medla i~formation program~ 7. Begin school education programs 8. Research the details of the cigarette problem The first four objectives are the top priority preventive measures. If these things can be done, then we may" prevent an overall increase in tobacco usage which will almost certainly occur if we graft onto indigenous smoking habits the habit of smoking m~nufactured cigarettes. The strategies that will resist the marketing of cigarettes in the West should resist the enlargement of ~he cigarette market in the East. The same strategies will resist the change of culturally entrenched smoking habits towards more ~se of manufactured cigarettes we know exactly what to do. It ~s true, however, that we do not know exactly how to overcome the political obstacles w~h~ch sta~d in the way of our doing it. T108350771

85 The second problem in the developing world is that we do not know what to do about the indigenous smoking habits so widely entrenched around the world and producing such a galaxy of neoplastic disease. Figure 3 sh~s so~ of ~he i~igenous fo~s of tobacco use. They ace representative of ~ny varia- tions in tobacco use. FIGURE 3. COMMON FORMS OF TOBACCO USE Bidi Chutta Chillum Bookah Snuff Chewing Niswar Biri Brus In India the first four types of smoking occur - bidi, ehillum, chuttah and hookah. They also use snuff and chew tobacco, usually betel leaf mixed with llme and various flavouring agents (12). Nepal, Pakistan and Bangladesh share similar smoking habits (13,14,15). Niswar is another type of smoking found in Pakistan. In Bangladesh the hookah used to be smoked. This is a long pipe with a bowl at the base in which tobacco is placed. This is seasoned with molasses and sometimes with narcotics and topped with burning charcoal. It does not permit much mobility. Now the hookah is being replaced by the manufactured cigarette and the biri - the home-rolled equi- valent. Brus is a cigarette made of partially cured tobacco treated with molasses which is smoked in Papua New Guinea (16). There are differences in the way in which these indigenous products and the manufactured cigarette are smoked (Figure 4). FIGURE 4. INDIGENOUS SMOKING PRODUCTS i, Smoked for a longer time Not inhaled or inhaled less deeply 3. Smoke two or more products 4. ~igh tar a~,d nicotine levels 5. Smoke fewer 'pieces' daily 6. Extremely cheap Items 1 through 5 mean that it is very difficult to measure the actual tar exposure for people smoking indigenous products. For example, in Papua New Guinea, inhalation is reported 5y only 18% of men who smoke brus but by 56% of cO, teat in brus in m=ch higher than in the manufactured cis~ret~e (16). To compound the problem, a person who smokes brus might smoke only two or three a day, while a person who smokes ~anufactured cigarettes might smoke ten to twenty a day (17). T108350772

Item 6 is an i~portant reason for stopping the introduction of the ,~nufac- tured cigarette. In India the bidi is extremely cheap, at $I for 400-500 cigarettes, while m~nufactured cigarettes are about $i for 40-50 cigarettes (12). Finally, the patterns of indigenous smoking are usually quite well-deflned. Men usually smoke much more than women. In Asia, between 40% and 70% of men smoke but less than 30% of women (18). In some countries, smoking is considered socially unacceptable for women and consequently the smoking rates are very low. In Lagos, 42% of men but only 2% of women smoke (19). CAIqCER L~i DETELOPING COI~NTRIES The pattern of cancer associated with the various smoking products of devel- oping countries differs sharply from that associated with western ciga- rettes. Although the sophisticated studies done in the West, which impli- cate cigarettes in cancer of the lung, have not been duplicated in the East, there is really no doubt that indigenous tobacco is a major cause of cancer in developing countries. Higginson estimated that tobacco is implicated in 40% of cancer in Bombay (20). It is, of course, very difficult to know what are the non-cancer health effects of indigenous smoking and chewing products. Smoking bldi, chutta, brus, etc. exposes the individual to much the same spectrum of compounds that are present in western cigarettes. But does bidi smoking cause heart disease? This will probably not be known with certainty for a long time. Heart diseaae attributable to smoking in, for example, Canada is also related to over-nutrltion, under-exercise and hypertension. Bidi certainly produces tar, nicotine and carbon monoxide but does it cause heart disease when allied with undernutrition, especially low fat diets? If so, how much? 0nly time will answer this. This comment leaves us with a crucial point. Cancer is caused by bidi, chutta, cheroots and chewing. It is caused in sufficiently high numbers to warrant public health action. We do not need to know whether or how bidi smoking causes heart disease to justify a public health program to reduce it. The fact that it causes one disease is enough to justify public health A good picture of the type of cancer problem arising from indigenous smoking habits ~s given by Jussawalla's case control study in Bombay (21). This study looke~ at the risk of developing oral, pharyngeal, laryngeal, and esophageal cancers among Indian smokers and chewers. Such studies allow assessment of the relative risk of cancer between tobacco users and non-users. Jussawalla found an overall relative risk for cancer of the head and neck of 5.6 for Indian smokers; 4.1 for Indian chewers; and 15.7 for people who chew and smoke. In other words, this ~eans that peop|e who chew and smoke tobacco have 15 times the risk of people who do not. This degree of risk is not very different from that associated with the persistent smoking of a fine old American Virginia cigarette. T108350773

Although smoking and chewing together in India are potentially as lethal as smoking American cigarettes, there are many competing causes of death in India and the general mortality from cancer in Bombay, for example, is relatively low if compared with other causes of death. Juss~walla gives a crude male mortality rate from cancer of all sites among men, of 41.8 (per 100,000) which compares with 123.5 (per 100,000) deaths from TB, and with similar numbers for p~eumo~ia and heart disease (Table II). TABLE 11. CAUSES OF DEATH (MALES) Metropolitan Bombay Cause Rate per i00,000 All cancer 41.8 Tuberculosis 123.5 Pneumonia 120.3 Heart diseases 112.9 Source: Jussawalla (21) So smoking and chewing causes cancer in Bombay but how much cancer does it cause by world standards? The answer is, more than many people think, Table 12 is extracted from 'Cancer Incidence in 5 Continents' comparing the incidence of smoking/chewing associated cancers, excluding oesophagus and bladder, in Bombay, Connecticut and Birmingham (22). The rates are compa- rable because they have been age standardised to a standard world population. TABLE 12. AGE STANDA_RDISED CANCER INCIDENCE (Standard World Population) Bombay Connecticut Birmingham Lip 0.3 2.1 1.2 Tongue 12.6 2.8 1.0 Mouth .6.7 4.3 1.5 Oropharynx 5.6 2.! 0.6 Hypopharynx 7.7 1.5 0,8 Larynx 13.6 7.8 3.9 Lung 13.5 53.7 77.1 Total 60.0 74.3 86.1 Source: UICC (22) Ne know that non-western smoking and chewing habits abound in Asia and Africa. T~ere may be hundreds of different tobaccos which are smoked and chewed in hundreds of different ways. But is the related cancer appearing ~- ~r is it ~>uried ~y in~ecEio~s ~isease an~ ot~er pro~!emsT ~e answer ~s that the cancer is there and is consuming substantial resources. Formal and comparable data are hard to find. However, a recent OICC conference on cancer in developlng countries was organlsed by Hirayama in Japan and T108350774

revealed quite clearly that tobacco cancers are common in developing countries. Incidence a~ mortality are often not measured so we must rely on samples from various sources which do carry quit~ substantial bias. Table 13 provides some examples (23). TABLE 13. PROPORTION OF SOME CANCERS Country Lung Larynx Head & Neck Source Cancer Cancer Cancer China 8.5 Mortality Statistics Korea 9.5 Cancer Hospital Thailand 8.8 4.2 18.0 Hospital Indonesia 11.8 Hospital Bangladesh 17.0 13.7 18.8 Hospital Sri Lanka 1.0 2.2 38.5 Hospital (M & F) Pakistan 10.6 5.6 14.7 Multi-centre study Cancer of the lung causes 8.5% of all Chinese cancer deaths; 9.5% of cancer in the Korean Cancer Research Hospital and 8.8% of deaths from cancer in a Thai hospital. Cancer of the head and neck causes 18% of cancer in males in a Thai hospital, 38.5% of cancer in the cancer hospital in Srl Lanka (males and females), and 14.7% in Pakistani males (the most common tumour; the second most common is lung cancer at 10.6% for males). In Indonesia, cancer of the lung is the second most common (after liver) in males causing 11.8% of admissions to the Dr. Sutomo Hospital where 5.63% of admissions are caused by cancer of the nasopharynx, and 3.96% for cancers of the larynx. The data presented at the Japanese conference were difficult to compare with western data because of the different types of bias. However it leaves one with very little doubt that tobacco associated cancer is a problem. S,o what sort of policies should the developing countries have for their established indigenous problems of smoking and chewing? We probably know enough to make some suggestions and to raise some questions. Figure 5 indicates some possibilities. FIGURE 5. POLICIES FOR T~E INDIGENOUS SMOKING PROBLEM i. Recognise the problem 3. !n~orm the public 4. Taxation? 5. Schools? T108350775

Recognition of the problem is a polltlcal and practical need. To obtain proper recognition it may be necessary to research it - the usage not the pharmacology. This is ~ot to pour cold water on admirable people like the Tara 14emorial Institute in Bombay; it is emphasising merely that usage information is enough for political and public health recognition. Ways to inform the public u~st be found. Almost every village in every country has a transistor radio. Tobacco use should carry a high priority on radio. The East can profit £rom one of the lessons of the West; large numbers of people give up smoking because they are informed and they believe that smoking causes disease. Next there are two questions. Is taxation a useable and powerful weapon? It certainly works with cigarettes; will it work with bidi? Can chewing mixtures be effectively taxed? Where should the tax be placed? Should the tax be placed on the raw tobacco or at the point of sale? Will taxing bidi make cigarettes more attractive? Should both be taxed and the taxes linked? Some experiments are needed to answer these questions. Can the schools be used to inform the next generation that their parents' smoking habits are unhealthy? This is a question as well as a suggestion which needs research. Probably the place to start is in the teacher training institutions. I do not think we can predict what is going to happen to developing coun- tries over the next decade. I hope we can modify the smoking habits of history and keep the western cigarette at arms length. Time alone will tell. The UICC Program on Smoking and Cancer has now run smoking workshops in twenty countries. We have learnt one important lesson. In every country, however complex the problem, there have been obvious, and neglected, public health opportunities. Most countries need to have a meeting, collect an activist group, establish some clear priority targets and start work on the things that can be done. Finally, there is a group of activities which remains largely neglected by antl-smoklng activists. These activities relate to the international behavlour p.atter~s of governments and the various cigarette ~tanufacturers. Some highly undesirable things are happening and it is time the health lobby did more about them. They include: I. E~rt of t~m~¢=o auxins. America is a well known exporter of tobac- co; sometimes it is sold, sometimes it is given away. There is, how- ever, considerable pressure within America to reduce the number of people dependent on tobacco manufacture and consequently to reduce exports. America is not the only offender. The United Kingdom, India, Brazil and Turkey are major exporters o~ tobacco or cigarettes (24). T~e em~ort o£ hip ta~ cigarettes. Table 14 shows a comparison o~ cigarettes sold in Singapore in 1980 compared with the same brands sold in Australia in that year. T108350776

TABLE 14. TAR DELIVERIES Country Brand Australia Singapore of origin Benson & Hedges 14 31 Singapore C~el I$ 27 USA Consulate 14 28 Singapore Du~hill 16 30 Singapore Kent 13 19 USA Lucky Strike 15 23 USA Marlboro 14 21 USA Peter Stuyvesent 16 33 Singapore Rothmans 14 32 Singapore Winston 16 23 USA It is obviously unnecessary for people in Singapore to be smoking clga- reties with a tar level over 20 m~. No doubt we will soon see something done about this particular problem. The people who are importing, obviously have the most power and we should counsel both developed and developing countries who import cigarettes to look at the tar content of the rubbish which is being dumped upon them. 3. Export of improperly labelled cigarettes. This is a universal practice connived at by the airlines and governments. It is a traditional practice and difficult to change. Nobody has made it their job to attack the problem. It is one thing to persuade developing countries to put s health warning on their own cigarettes. It is equally important that American and British companies are stopped, by their own countrles~ from exporting improperly labelled cigarettes. It is hard to interest governments in this sort of thing as it does not look like a high priority target. However, it is important and it is immoral for such things to occur. Health warnings are on packets because the product is unhealthy. There can be no justification for the sale of cigarettes without a health warning anywhere in the world. International a~ency ftmd~ of tobacco. The UN fami!y has considered this miter and has spoke~ some nice words on the eubject. Nevertheless I do not believe the problem is yet beaten sad we have to continue to oppose sponsorship of tobacco industries in developing countries by the international funding agencies. !q!e problem of a world increase in tobacco g~owing stimulated by presenti~ the ~obacco industry ~r~y line, ~ich is very irresponsible. The world ~eds food; it d~s no~ ~ tobacco. "Before I grew tobacco I was very poor and scratching a no bicycle aml ~o radio and ~o hope of a better future. ~ I have ~ot ~at I never drea~d would ~ ~ - ~ ~ ~rac¢or. I ~ also able to educate ~ children and look after ~ wives ~tter." Source: Advertisement in Ghana. T108350777

We need to address the question of w~o should work in the international ££eld. The World Health Organization has said the right things and has a good policy lint is emasculated by its inability to do anything m~re than advise; it cannot lobby. Other international cancer and international heart agencies are amall and lack funds, although they are well coordinated £n policy. I think the time has come for those agencies to ask their member bodies to 4~ m~re interuatioaal work in the national arena. As an example, this means the International Union Against Cancer should ask the member societies to join with the other health agencies in lobbying governments to reduce tobacco exports and the export of high tar and unlabelled cigarettes. My final point is, therefore, that the international health groups have a responsibility to create national lobbying to counter the tobacco industry's international immorality. I. Egger G. Estimated costs of the smoking habit in Australia 1971/72-1979/80. A report to the Anti-Cancer Council of Victoria. 1982 March. 2. Roemer R. Legislative action to control the world smoking epidemic. Geneva, Switzerland: World Health Organization, 1982. 3. Where cigarette smoking rises without advertising. World Tobacco 1981; 72: 71-75. 4. Bjartveit K, L6chsen PM, Aar@ LE. Controlling the epidemic: legislation and restrictive measures. Paper presented at the conference~ 'Smoking or Health in the '80s', Toronto, Ontario, 1981 May. 5. US Dept. of Health and Human Services. The health consequences of smoking for women. Report of the Surgeon General. Rockville, Md.: DHHS, 1980. 6. Britain - General Household Survey 1980. OPCS Monitor, June 1981. 7. Hill DJ, Gray NJ. Pattern of tobacco smoking in Australia. Med J Aust 1982; I: 23-25. 8. Hayashi T. Recent trends in smoki~ in Japan. World Smoking and Health 19.~0; 5(2): 40-42. 9. Health and Welfare Canada. Smoking habits of Canadians, 1963-1979. Ottawa, Ontario: Health and Welfare Canada, Health Protection Branch, 1980. (Technical Report Series No. 9). an epidemiologic study. ~rev l~d 1981; 10(3): 301-315. II. Richmond and ~awthorn chest X-ray survey, Victoria, Australia. 1974. T108350778

92 ~ 12. Sanghvl LD, Jayant K, Pakhale SS. Tobacco use and cancer in India. World S~oking and Bealrh 1980; 5(4): 4-10. 13. Pandey MR, Shrestha l~K, Opadhyaya AB, Neupone RP. Prevalence of smoking in a rural community of Nepal. World Smoking and Health 1981; 6(1): 14-18. 14. Smoking and health in Asia. NIK) Chronicle 1981; 36(4): 156-159. 15. Cohen N. Smoking health and survival prospects in Bangladesh. Lancet 19'81; I(8229): I090-1093. 16, Brott K. Tobacco smoking in Papua New Guinea. World Smoking and Health 1981; 6(3): 33-37. 17. Anderson HR. Smoking habits and their relationship to chronic lung disease in a tropical environment in Papua New Guinea. Bulletin de Physio-Pathologle Respiratolre 1974; 10: 619-633. 18. Benjamin B. 94-97, Tobacco smoking in the world. WI~O Chronicle 1979; 33: 19. Taha A, Ball K. Smoking and Africa: the coming epidemic. Arab J Med 1981; I(I): 24-27. 20. Higginson J. Personnal communication. 21. Jussawalla DJ. An assessment of cancer mortality rates in Bombay and future problems by examining current tobacco smoking and chewing habits and case control study on tobacco smoking and lung cancer. In: The UICC Smoking Control Workshop, Nagoya, 1981. 22. International Union Against Cancer. continents. Geneva, Switzerland: UICC. Cancer incidence in five 23. Cancer prevention in developlng countries. Proceedings of the First UICC Conference on Cancer Prevention in Developing Countries, 195l. 24. Record Setting volume in '81 will not be nmtched by '82 shipments. Tobacco Reporter 1983; i: 26-29. T108:350779

The Honourable Samuel Hynd Minister of Health Ministry of ~ealth P.O. Box 5, Hbabane Swaziland The very title of this conference tells us of its nature and content. The matters discussed reach out from the sophisticated centres of the so-called 'Developed World' to the farthest village, in the forests, plains and deserts of the 'Developing World' I believe there is symbolic significance in the fact of our meeting on the soil of one of the American continents, for if my history books tell me correctly, it was within these continental shores that all these troubles began. It also recognises a major advance in attitudes and mood that we can risk returning to this continent to ask what can be done about the leaf that has come to mean death and destruction to the lives of so many people and to such a degree that one of our guest speakers went so far as to describe it, in Stockholm, as a 'Holocaust'. It is equally significant that someone like myself, coming from the conti- nent of Africa and from a developing country, would be invited to chair this session. I feel honoured to have been invited to do so and willingly accepted, for I consider that I represent the millions out there who are to become the major target of what is one of the most organised, wealthy, agriculturally based industries in the world today. What is more, all the indications are that we, in Africa, are now to have all the attention of the big guns of the tobacco industry turned on us because we are less advanced in preparing for the onslaught of knowledge, education, legislation and m~ny other things, than other continents. I have personal proof of this, for it was just over a month ago that my country had the honour of being visited by a h~gh ranking marketing agent of a major international tobacco organisation for the first time. In addition, word has leaked out that we are currently preparing legislation for presentation to Par!iam~nt. One of my first appointments on returning home is to meet top level officials of two major tobacco companies to discuss the contents of the proposed Bill to control the smoking of tobacco. It is a horrifying thought that, w~en we are still struggling hard to over- come all the health hazards of communicable, diarrhoeal, respiratory, nutri- Address for correspondence: Dr. Samuel W. Hynd, National Council on Smoking, Alcohol and Drug Dependence, P.O. Box 384, Manzlni, Swaziland. T108350780

tional ar~i parasitic diseases that plague our communities, we are ~ow to contend with the export of diseases from the developed industrialized nations. •here are all s~rts of protest meetings being held all over the ~orld on such things as nuclear warfare a~d ~he endangered environment, yet there is so little protest on what is becoming the most dangerous and most widespread preventable disease ~o inflict humanity. The problem is that it is a slow, sinister, suicidal drug that mkes us die quietly and with only a ~nlmper. I ask the question, how do you expect us to reach the W~O goal of 'Health for s11 by the Year 2000' when we carry the heavy burden of poverty, ignorance and disease, only to be exploited by those affluent, educated countries who have the expertise to stop all this nonsense of self destruction? Why must we wait until the crisis in our developing countries reaches catastrophic proportions before we do something about it? To date, we have treaded too softly, but we must break out and make our protest loud, long and clear. We have accumulated sufficient scientifically based, tested and tried knowledge; we have enough experience of all the forces of vested interest to proclaim that we have had enough. In my welcoming speech when Swaziland hosted the First African International Conference on Smoking and Health in April 1982, I went so far as to shock my hearers by stating that the time had come to declare War on Tobacco. When we reach the year 2000, I hope we can sign the Peace Treaty, lay down the terms on the most preventable cause of death in the world and declare that the only way to stop it is to stop smoking. The forces at work are heavily against us. If you do not think it possible to win, remember that the same was said of smallpox, but by mobilizing the international community and by showing some evidence of tenacity and courage, the battle was won. We are not doing too badly on the breast milk substitutes issue and so let us take heart and fire some shots, small as they may be. Let us not fool ourselves on tobacco - the forces are stronger, better organlsed, have so many tied to their strings and have so much money, that it is going to take everything we have got to beat them. I have read a story of two armies ~ha~ faced each other m~ny centuries ago. One army had a giant come out in front to challense so~eone from the opposln$ army for a fight. All were petrified and demorallsed because there was no one with the courage to accept the challenge. A small herd boy came to check on them. He was shattered to find not one person willing to challenge the giant. He had no weaponry to equal the challenger, but went to the stream that separated them, picked up five pebbles, put one in his sling and fired. T~e ~iant fell flat on his face and the day was won. We face a similar David and .Goliath situation between what ~he Conference seeks to do and what that huge giant has with which to challenge us. Our task is to find the courage to pick the right stones, and to make it the be~innin~ of the end qf ~ ~attle. T108350781

95 H. Mahler, M.D. Director-General World Health Organization 1211 Geneva 27, Switzerland The goal of 'Health for ali by the Year 2000' is about people. Whenever peoples' habits conflict with their health, it is the duty of those who understand the danger to warn those who may not, in order to motivate them to modify their habits and thus protect their health, in their own interest and in the interest of their families. However, to get people to modify their habits in order to improve their health is no easy matter. Difficult as it might have been to get people to work together to combat health enemies that descend from without, that has proved infinitely easier than getting them to combat health enemies that descend from within. For that is what smoking does - the urge to smoke descends from the mind, and the carcinogenic smoke descends into the lungs. Before attempting to indicate any new directions that we might usefully follow, I will sum up the present world situation concerning the smoking epidemic, and in particular the changes that have occurred since the Fourth World Conference took place. There are both encouraging and disturbing signs as we look at present day trends and the published reports concerning the smokin~ epidemic. Generally speaking, smokers now form a minority in most highly industrialized countries, with some exceptions. For example, in most of the European Common Market countries the percentage of adult smokers is decreasing. Another encouraging note is that in other countries, there has been a marked decrease also in the number of young smokers. The most alarming trend today is towards major increases in tobacco consump- tion in developing countries. In some of these countries smokers form the majority, or are close to it. This is not purely fortuitous, since we now know that the major tobacco companies in the world have set their sights on this population to offset declining sales in industrialized countries. How- ever, even in the developing countries there are some encouraging signs. In 1976, only 20 countries in the world had some kind of smoking control legls- lationo Today the number has risen to 60. At the time of the Stockholm conference, only a handful of countries, mostly the highly industrialized ones,, had any smokln~ and health associations° Today, al~st all countries, and this includes m~ny of the developing ones, have a committee or other body that deals specifically with smoking and health issues. The first reaction in many parts of the world in dealing with the smoking problem is legislative action and this is perfectly understandable; it shows T108350782

ever, ~e have to recognise that legislation can only be effective if it has public support, and that this support will only be forthcoming if people understand and accept the need to avoid smoking, or ro discard the habit if it has already taken root. How can ~e generate this public support? I have already referred to the disease descending fro~ the mind. It is at that level that ~e ~st exert our influence; when the smoke has descended into the lungs it is too late. We ~st recognise that, if we want to influence people to change their ways, we cannot restrict such a change to smoking a|o~e~ for s~klmg is c~ly ~e sy~to~ of a bro~der syredro~e of ~d~ealthy living. In addition to not smoking at all, I am referring to eating wisely, drinking moderately, driving carefully, taking enough exercise, learning to llve under the stress of city life, and helping one another to do so. This is individual and community involvement in health to the extreme - that is: radical changes in life styles. If we want to influence people to change their corporate life style and individual behaviour we will have to display realism. I am convinced that we will also have to be much more inmglnative in combining social, psychol- ogical and economic research in order to arrive at any reasonable solu- tions. One thing is clear, the "thou shall not" approach has never succeed- ed. As part of our realism of replacing negative commandments by positive injunctions, we shall have to propose positive social alternatives to nega- tive social phenomena. And here we enter the realms of value judgement. Some will advocate the promotion of sport and the provision of sports- grounds for youngsters. Others w~ll claim that what is needed is a restora- tion of family life to what it was in days gone by when families gathered round the dining table and food and drink were holy rituals. The dining table has given way to the television set. In theory, this could be an excellent medium for health education. However, television is all too often used for mass advertising to induce people to buy what they do not need, to drive cars whose performance tempts the unwary to speed, to eat and drink to excess and to smoke tobacco. The educational system has been blamed for its indifference to society's problems, and yet it could have a powerful influ- ence'on people of all ages if only the right messages were got across in the right way. If only we knew what the "right ~essages" really are and how to get them across in the right way! For some people the preventing answer might lie in music, art or literature; for others in meditation. But how can they be identified, and who can do this? These are the challenges we have to face. We must get people to act as agents of change. ~or example, we ~st get people in developing countries to raise their voices against tobacco neocolonialism so that they do not become the dumping grounds for highly toxic cigarettes, with no health warn- ing, that the developed countries no longer a11ow for sale at home. The opiumwar of the mid-nineteenth century could act as a provocative historical reminder. We ~st do more to influence the members of the health professions to take an even more active role in fighting the smoking epidemic. It is not enough to provide medical care to others. Their personal example and their rela- apeutic act they perform. I can say this with an easy conscience, because in W~IO we have put an end to the sale of cigarettes in the building and have o~tlawed smoking during meetings. Moreover, we started by introducing a few T108350783

non-smoking tables in our staff cafeteria iu Geneva and now more than two- thirds of the cafeteria is out of bounds for smoking. Ne must look for imaginative ways of influencing young people to assume responsibility for their own health. Horal£sin~ will certalnly have no influence on them, particularly if it comes from their elders. W~O has been contemplating launching a series of activities called 'Winners don't smoke". This would eutail e~liati~ the aupFort of well-kuo~a al~rtsme~, both world-famous ones and those who are heroes in their own co~mm=nities. It would entail invol~ing eminent public figures and film and theatre stars. We need to find people who would be models for youth: to assemble a galaxy of stars that will outshine the false glamour of professional models who promote smoking. We would associate success in all walks of llfe with non-smoking and a healthier life-style. WHO cannot do all this alone. The initiatives, and public lobbying by others are crucial for the success of this drive for health. T1083,50784

R. l~sironi, Ph.D. Coordinator, ~0 1~rogramme on Smoking & Health Wo~ld HeaI~h OTganization Geneva, Switzerleml The title of the Fourth World Conference on Smoking and Health, which was held in Stockholm in 1979, ~as: "The Smoking Epidemic - A Matter of World- wide Concern". In the same year, the World Health Organization issued the report of an Expert Committee under the title "Controlling the Smoking Epidemic". This co~ncldence of titles using the word "epidemic" underlines the consensus of public health circles that the menace to health that we are deal~ng with here, namely tobacco smoking, must be considered as a new kind of communicable disease. In the past, epidemics were only spread by bacteria and viruses, by insects and other animals. Today we are being confronted by a new phenomenon; that of a man-made epidemic which is spread by images, aided by every device of modern communication technology. This phenomenon reaches to every part of the world. Although not alone, smoking is certainly one of the most striking examples of this commercially motivated contamination. It is a global problem, and one which is really spreading like an epidemic from country to country and from continent to continent. Likewise, smoking control action must also be a global undertaking, as it cannot be successful if carried out by countries in isolation. Action must be comprehensive both in structure and in geography. International action is mutually reinforcing. Many governments, indeed, will not act until they know that other governments have already taken certain measures that have proven to be successful. For these reasons, the World Health Organization is ready to assist coun- tries in their quest for an efficient way of collaborating among themselves in health mt~ers. The question therefore arises: ~d~at is W~O's action in this f~eld and how ca~ countries, through W~.O~ optimize their national smok- ing control programmes a~d their international collab.oration~ The kind of action that hrdO can carry out is, first of all, to facilitate communication through technical meetings, through collection and dissemina- tion of data, and through the issuing of guidellnes that public health a~thorities could follow if they so wish. Among the activities that the World Realth Organiffiation has carried out since the Stockholm conference, the following could be mentioned in chronol- ogical order: T108350785

The selection of the theme "Smoking or Health - The Choice is Yours" for World Health Day on 7 April 1980. The establishment of an identifiable WR~O programme on smoking and health. A ser~es of national and international seminars on smokin~ and heal~h issues ~hich were held, or will be held shortly, in Lanka, Swaziland, Mongolia, India, Nepal, Thailand, and in other developing countries. The aims are not only to exchange ideas but also to help build national programmes on education informa- tion, legislation, and research. Following the co-sponsorship of the Fourth World Conference on Smoking and Health, the World Health Organization has co- sponsored the international conference on tobacco and youth, which was held in Venice in 1981, and is co-sponsoring the Fifth World Conference. In November 1982, an Expert Committee met in Geneva to discuss smoking control strategies in developing countries. The report will be discussed by the Expert Committee's chairman, Dr. A.R. AI-Awadi, at one of the sessions of the Fifth World Conference. This report is the third of its kind. In a more technical field, the Organization has published standardized ques- tlonnalres for the assessment of smoking habits in adults and young people. The questionnaires are available to investigators and public health author- ities that intend to carry out smoking habit surveys in their countries in a standardized, reproducible way. Indeed, most of the surveys that were car- ried out in the past, were of little use in determining time trends in smok- ing habits, both nationally and internationally, because these studies have usually been carried out in an uncoordinated, unstandardized way. A few days before thd Fifth World Conference a group of experts met in Winnipeg under the auspices of the American Cancer Society, the International Union against Cancer, and the World ~ealth Organi~atlon to work out a standard- ised questionnaire for the assessment of the smoking habits of medical personnel and other health professionals, including primary health care workers. Much of the efforts carried out by hea|th educators in tryin~ to diminish the intensity of smoking in populations would lose impact if health professionals themselves smoke. It is therefore essential to ascertain their smoking habits, try to modify them, and follow them over time. The Organization also promotes the standardization of methodologies and of analytical procedures for the determination of tar, nicotine and carbon monoxide in cigarettes, particularly those available for sale in 4~veloplng countries. It also assists these countries in the establishment of labora- tories for cigarette smoke analysis, and in the training of analysts. The ~yses are o~ones:~rc~ ~ounoa~l(m ~n Finally, the Organization collaborates with other I/~.4 agencies (e.g. UI~ESCO, FAO, UNCTAD) and non-governmental bodies, according to their various fields of expertise. T108350786

Is international action effective? Encouraging signs are available. Since the last World Conference on Smoking and Health, momentum in smoking control action has increased rapidly worldwide. World Health Day in 1980 which, as already mentioned, had as a theme '~moking or Health - l~ne Choice is Yours", was very successful throughout the world and its impetus is still being felt. Many countries produced special events and even postal stamps on that o~assi~ a~i a~ couotries are still producing such stamps uow. Anti-smoklng education campaigns have been mushrooming around the world. The number of national smoking and health associations has also been increasing rapidly. A few years ago, only a few countries, mostly indus- trialized ones, had smoking and health associations. Nowadays such bodies are present in most countries, both developed and developing ones. In 1976, a WHO survey of smoking control legislation revealed that only about twenty countries had some sort of legislation in this field. The latest WHO survey, published last year, revealed that about sixty countries now have such legislation. This is a clear indication of how the momentum of successful legislation in some countries can induce other countries to follow the example. From a social point of view, smoking is becoming less acceptable. The evidence of decreasing social acceptability is still mostly circumstantial, but appropriate studies could certainly quantify this evidence, for instance through surveys of people's attitudes towards smoking. As an obvious result of these actions, the percentage of smokers is decreas- ing, particularly among the middle-aged educated males in industrialized countries. But there are also decreasing trends among women and children. Parallel to these decreasing trends in smoking, although not necessarily and totally related to it, decreasing trends in smoking related diseases e.g., in cardiovascular mortality, can also be observed in many industrialized countries. In some of them, even lung cancer rates are going down. If steady educational and legislative pressure is maintained and further expanded, it is reasonable to expect that the habit of tobacco smoking could be progressively phased out of the social mainstream. The measures adopted in developed countries are mainly applicable to the urban centres of developin~ countries, ~fnere only a s~ll fraction of the population lives. For the vast majority of people living in rural areas of developing countries, antl-smokin~ camp.ai~ns should be incorporated into health education programmes through the primary health care system, through the schools, the religious leaders and other appropriate channels, as part of comprehensive health awareness campaigns. The tobacco industry is a transnational enterprise, and this is why it is so powerful. Therefore t~e health enterprise must also be transnatlonal. Ideas spread easily nowadays. Fashion and artificial needs such as the use of cigarettes are created and are spread across the world. The very same brands of cigarettes that are so popular in a given country are often also popular in other countries on the other side o[ the earth, b~ving completely different socio-economic and political systems. In spite of all the differ- ences that may exist between countries, the cigarette is a common factor, unfortunately a factor of addiction and disease. It is high time that T108350787

102 ~overaments s~itch their emphasis from the perpetuation of s harmful habit to the elimination of it, so as to fulfill their responsibilities for the health of their peoples as spelled out in the constitution o£ the Norld l~ealth Organi=ation, which these very same ~>vernments have a~reed to. T108350788

---

103 DAY Inger Asmussen, M.D. Department of Cardiology B 2142 University Clinic of Cardiology Rigshospltalet University of Copenhagen Copenhagen, Denmark Smoking during pregnancy bears an increased risk to the unborn child. The risk is present all through pregnancy as well as after birth. Smoking induces spasms in the uterine arteries thus decreasing oxygen supply and nutrition to the uterus, placenta, and fetus. Measurements have been performed on fetal blood in animals showing the following in response to maternal smoking: hypoxia, increase in carbon monoxide, presence of nicotine, benzipyrenes, thiocyanates, etc. Thus the active substances found in the blood of smokers can be found also in the blood of the unborn child. When the mother smokes, the placenta is poorly vascularlzed, in part due to arterial spasms which occur in response to every smoked cigarette. In addition, the placenta is immature, with a poorly developed vascular tree. Thus in smokers the placenta is underperfused with oxygenated blood which may reach critical levels when inhaling a cigarette. The child, in response to the smoking of a cigarette may, as in the mother, show tachycardia, a sign of stimulation of the adrenergic system. However, bradycardia may occur as well and is regarded as a sign of serious trouble for the child. Smokers have a high incidence of premature birth, still-born children and term children with neonatal death. The most well-known complication of smoking during pregnancy is low birth weight - a reduction approximately of 10 to 15%. This reduction in birth weight can be explained as a response to the underperfused and immature placenta with a reduced number of capillaries. The placentas are small and fibrotic, a sign of tissue damage a~d repair, and the function of the placenta is reduced. The protein synthesis, essential for fetal growth, in placentas of smokers is signiflcantly lower than that of non-smokers. The children born to smokers are smaller with less subcutaneous fat and shorter. Children born to heavy smokers have a reduced number of neutro- philic white blood cells, which may explain the high incidence of infectious diseases, mainly upper respiratory tract, in these children. Another expl~nation of the frequency of bronchitis and p~eumonia in smokers' off- spring is immunosnppression. Children of smokers remain small and slim at least till teenage. The gross and fine motor development is decreased compared with that of non-smokers' children. Also verbal comprehension development is reduced. The children may suffer from minimal brain damage. At school, children of non-smokers do better in mathematics and form a and restless. T108350790

ATJ~O~)SIS ~ CA~C~ The main risks attributed to tobacco smoking are the development of athero- sclerosis and cancer. Epidemiological studies have shown the high incidence of atherosclerotic complications in smokers: stroke, intermittent claudica- tion, angina pectoris, myocardial infarction, ischemic heart disease, cardiovascular deaths, etc. The severity of the disease correlates with the number of cigarettes smoked. Studies have been performed on umbilical arteries of newborn children of smokers. The morphological studies have been carried out blindly and the morphology was compared with that of non-smokers. In umbilical arteries (these vessels form part of the fetal circulation) early arteriosclerotic changes w~re visualized: increased endothelial cell death with an increased turn-over rate of both endothelial cells and smooth muscle cells. Formation of pseudoendothelium by the media smooth muscle cells was seen. An increased amount of fibrous tissue and intense smooth muscle cell proliferation were observed, similar to that of the adult atherosclerotlc lesion: the fibro- muscular lesion. Also, the capillaries of the placenta (fetal vessels as well) suffer microangiopathy. Such vascular lesions - macroanglopathy and microangiopathy - could be demonstrated at birth in smokers' offspring, when the mother smoked more than ten cigarettes daily all through pregnancy, but also in a small group consisting of women who had been heavy smokers only during the first trimester and then ceased smoking. This tends to indicate that, with respect to smoking, the first trimester during pregnancy is the period during which most damage occurs to the unborn child (similar to infectious diseases). The increase in ischemic heart disease and myocardial infarctions among the young age groups (18 years and up) may perhaps be a late result of maternal smoking during pregnancy. Chromatin changes were observed in a recent study in smokers' offspring compared with non-smokers. Ultrastructural studies were performed on biop- sies available at birth from newborn children of heavy-smoking mothers (ten or more cigarettes per day) and ne6er-smokers. The biopsies were taken from the placentas and from the cord vessels. A total of 52 subjects were studied - 22 heavy smokers and 30 non-smokers. Fine granular spots within the cell nucleus of both endothelial cells and smooth muscle cells were demonstrated. These changes were observed in smokers (X2 = 8.28, p<0.005). Similar chromatin changes in man have, till now, only been reported in relation to tumors. In animals with experimente! atherosclerosls, as well as animals with medically induced diabetes mellitus, nuclear changes with alteration of the euchromatln / heterochromatin ratio have been reported in the vascular wall. The umbilical arEer~es from these heavy smokers showed signs of increased cellular turn-over rate compat[b!e with the findings in experimental atherosclerosis. Also, in tumors an increased cell turn-over is found. Thus the nu=lear changes found in healthy newborns delivered by smoking mothers could be taken as an indication of altered nuclear activity in the smokers' tissue. In experimentally "smoking" animals an increased frequency of cancer has been reported among the offspring. The finding of chromatin changes in cord vessels and placentas from newborns delivered by these childre,. T108350791

Clearly, smoking during pregnancy carries a risk for the newborn child as shown in numerous epidemlologlcsl studies. But recent follow-up studies and morphological studies, as mentioned above, may indicate a risk for the offspring which will show up many years later. Therefore, follow-up studies are needed of smokers' children to elucidate the long-term risk of smoking durin~ pregnancy. Also, efforts must be put into advising girls and women of childbearing age to stay non-smokers. Support was obtained from: The Egmont Foundation, The Danish Heart Founda- tlon, The P. Carl Petersen Foundation, The F.L. Smldth & Co. Jubilaeums Foundation, The Danish Medical Research Council grant No. 512-8262 and -21122, and Dronnlng Louises Bornehospitals Research Foundation. e BIBILIO~RAPIP/ US Dept of Health, Education and Welfare. Smoking and health. A report of the Surgeon General. Washington, D.C.: USDHEW, Office on Smoking and Health, 1979. (DHEW publication no. (PHS) 79-50066). US Dept of Health and Human Services. Directory of on-going research in smoking and health. Rockville, Md.: USDHHS, Office on Smoking and Health, 1982. Asmussen I, Kjeldsen K. Intimal ultrastructure of human umbilical arteries. Observations on arteries from newborn children of smoking and non-smoklng mothers. Circ Res 1975; 36: 579-589. Asmussen I. Ultrastructure of the human placenta at term. Observations on placentas from newborn children of smoking and non-smoking mothers. Acts Obst Gynecol Scand 1977; 56: 119-126. Asmussen I. Ultrastructure of human umbilical veins. Observations on veins from newborn children of smoking and non-smoking mothers. Acts 0bat Gynecol Scand 1978; 57: 253-255, Asmussen I. Ultrastructure of the villi and fetal capillaries in the placentas delivered by smoking and non-smoking mothers. Brit J Obst 1980; 87: 239-245. Asmussen I. Oltrastruct~re o~ human umbilical arteries fro= newborn children of smoking and non-smoklng mothers. Acts Pathol ~icrobiol Immunol Scand [A] 1982; 90: 375-383. Asmussen I. Ultrastructure of the umbilical artery from a newborn Pathol ~icrobiol l~a~nol Scand[A| I~82; 90: 397-404. 9. Asm~assen I. Chromatin changes of endothelial cells in umbilical arteries in smokers. Clin Cardiol 1982; 5: 653-656. T108350792

107 Neal L. Benowitz, M.D. Department of Medicine ~n Francisco ~eneral l~ospital University of Californ~a, San Francisco San Francisco, California, USA The importance of nicotine in maintaining cigarette smoking behavior and contributing to adverse effects of smoking is generally appreciated. Yet, until recently, relatively little was known about the time course of nico- tine in the body during cigarette smoking and even less about the relation- ship between nicotine concentration and effects throughout the day. In addition, despite the intense interest in the question of regulation of nicotine intake by smokers, no adequate methods for determining daily intake of nicotine have been described. In addressing these issues, this paper will consider four questions: 1. What is the time course of nicotine in the body during daily smoking? 2. What is the relationship between blood nicotine concentration and cardio- vascular effects? 3. How much nicotine is consumed during daily smoking? 4. To what extent is intake of nicotine determined by smoking machine yield? The first published studies of the time course of nicotine blood concentra- tions after smoking a cigarette indicated a brief half-life of 30 to 40 minutes (1,2). This supported the widespread idea that the effects of nico- tine were transient and that a smoker smoked on repeated occasions through- out the day because the nicotine was rapidly dissipated from his body. How- ever, these studies were based on blood concentration data followed for an hour or less after a given exposure. We have investigated the disposition kinetics of nicotine after intravenous administration (3). We followed blood concentration of nicotine for several hours after the end of infusion so as to obtain a better estimate of the terminal disposition characteristics (Figur~ I). The use of intravenous 4~siug also allowed ~s to know the dose exactly, so that the volume of distribution and clearance could he computed. Data from pharmacokinetic studies in 14 subjects receiving intravenous nicotine are summrized in Table I. T108350793

FIGURE 1. PLASMA RICOTIRE CO~CEWI'RATIORS (-+S.E.M.) IR FIFE S~B~ECTS DURING AND AFTER CONSTANT INFUSION FOR 30 MINUTES (3). NtCOTtNE 2 ~ug/kg/min 8 2 0 50 100 150 200 MINUTES I 250 TABLE 1. P~COKINETICS OF NICOTINE Half-life Volume of diatrib.tion Total clearance Renal clearance (acid urine) Honrenal clearance 120 min 180 liters 1500 ml/min 200 ml/min 1100 ~I/mln Nhat can we learn from these pharmacokinetic parameters that allows us to understand cigarette smoking behavior better? The large volume of distrlbu- tiou indicates that nicotine is extensivel~ distributed to body tissues ~d that relatively little is in the blood at any time. Alth~ ~t ~o~n in Table I, we have found that there is considerable ~ndlvldual variability in the volume of distribution of nicotine (3), so that there is not a constant relationship between blood concentrations a~d total amount of nicotine in T108350794

109 the 5~iy among different people. This ~skes it dlfficult to interpret blood concentrations follo~ir~ a single cigarette as an indicator of the dose of nicotine delivered by that cigarette. Whether different distribution characteristics mean that there is a different relationship between dose and effect as a function of those characteristics remains to be determined. Clearance is a mathematical term which describes the rate of elimination of a dr~ in relnti~, to the coacentrati~* of that dr~ iu t~ bl~. The clearance of nicotine can ~ considered as the sum o~ clearance by ~he renal and nonrenal routes. Renal clearance is kno~ to be influenced substantial- ly ~ urinary ~ and urinary flow rate. In our studies, ~ found that the clearance may range from 15 to ~50 ml/min ~n alkaline and acid urine condi- tions, respectively (4). But even at its ~x~mum~ zonal clearamce is a relatively s~ll fraction of the total clearance, which is primarily a result of ~tabolism by the l~ver. The fact that the nonrenal clearance approaches the rate of l~ver blood fl~ suggests that hepatic extraction h~gh and ~hat the ~tabol[sm of n~cot{ne is qu~te rapid and likely to vary w~th factors that influence liver blood flow. As expected, we observed ~rked varlab~lity ~n ~abolic clearance among ~ndiv~duals Knowing the clearance of nicotine is important because clearance determines the plateau level of nicotine which will be achieved in the body with the intake of any given dose. Thus, s person who rapidly metabolizes nicotine will require a higher daily dose of nicotine to obtain a given level in the body compared with a slow metabolizer. Assuming that both the rapid and slow metabolizera are seeking the same average level of nicotine in the body, the rapid metabolizer will have to smoke more cigarettes or m~re intensively, taking in also larger amounts of toxic constituents of tobacco smoke. Although its overall clearance is high, because of its extensive distribu- tion in body tissues, the terminal elimination half-life of nicotine is relatively long, that is, 120 minutes on average. The significance of half-life is that it predicts the ti~e course of accumulation of nicotine during constant dosing. Based on pharmacokinetic principles, it is known that a steady state is reached in 3-4 half-lives. Thus, one might predict that, with regular smoking, blood concentration of nicotine would plateau at 6-8 hours. This is, in fact, what we observed (Figure 2) (5). One would also predict that when one stops smoking, as overnight, concentrations of ni¢otlne will remain at substantial levels for many hours. This was also eo~firmed in studies of smokers (Figure 2) (5). Thus, nicotine does not behave like s drug whose presence in the body is brief, but rather like a drug with gradual accumulation and disposition kinetics such that the smoker is exposed to substantial levels of nicotine for 24 hours a day. We h~ve also studied physiologic effects of intravenous nicotine and ciga- rette smoking. An increase in heart rate is a sensitive m~asure of nicotine effect (3). During intravenous infusion of nicotine (Figure 3), heart rate increased shortly after the onset of infusion, and reached a plateau quick- ly, despite continsally rising concentrations of nicotine in the blood. As blood concentrations fell after the end of infusion, heart rate fell as decli~ ss c~ared with ~e ascent ~ase. ~is is a characteristic of ~evelo~t of ~olerance. T108350795

110 FIGURE 2. MEAN CIRCADIAN BLOO~ ~',oO)TIRE AND ~~TIOSS ~I~ S~I~ ~NI~I~ (0.4 HIG~NI~TIB~ (2.5 ~) ~SE~ CIGarS ~SU~ B~ OF CIG~S. Bars indicate S.E.M. (5) O O 2O c~ I0 O O = 0 12 10 6 2 CLOCK TIME TI08350796

111 FIGURE 3. CARDIO~ASC~LAE ~SPORSE TO NICOTIk% (}{=5) ~ SA.LII~E (R=5) II~FUSIONS. ~scerisks indicate P<.O5, comparing nicotine and saliue conditious by Near.an Keuls post-tess. BP, blood pressure (3). MINUTES Can information obtained during intravenous infusion predic~ ~es~onses to cigarette smoking during the day? One would .predict that hear~ rate would increase with the first few cigarettes of the day, and then remain elevated throughout ~he day so long as nicotine was present. This is indeed what was observed (l~i~re 4). M.eart ~e wh~le s~kins followed a circa4ian ~=Zern similar ~o ~ha£ seen in abstinence~ but was consistently higher. ~art rate remained elevated even overnight when the subjects were not s~klng, confirmin~ that in s~kers ~e ~y see effects of nicotine for 24 hours a day. Persistent effects of nicotine throu~hou~ the day ~y explain why cardiac e~en~s such as sudden ~eath and wyocard~al infarction, the risks of Zi~ of the day or might, no~ in direc~ ~e~oral rela~io= Zo ci~are~[e S~kin~. TIO8350797

(Mean ±S.E.H. for I0 subjects.) To estimate daily intake of nicotine, we have used a method similar to that employed in drug bioavailability studies. The metabolic clearance of nico- tine was determined after intravenous administration of the drug. Clearance data were then used in conjunction with blood and urinary nicotine concen- trations, measured during a 24 hour period of smoking, to determine daily intake of nicotine. Using this approach, we examined intake of nicotine in people smoking different brands of cigarettes and the regulation of nicotine intake when smokeTs were switched from high- to low-yield commercial ciEarettes. Volunteer subjects who were habitual cigarette smokers of at least one pack per day were hospitalized in the Clinical Studies Center at San Francisco General Hospital for the duration of each study. On the morning of the second hospital day, an intravenous infusion of nicotine was administered to determine metabolic clearance. Total clearance, that is, the sum of metabo- lic and kidney c]earance, can be com~uted from the ~ose of nicotine i~fuaed and the area under the blood nicotine concentration-time cuvve (ADC). Clearance of the kidney can be computed from the amount of nicotine excreted in the urine and ~C. Honrenal or metabolic clearance is the difference. desired. All cigarette butts ~ere collected. Urlr~e was collected each for measurement of excretion of nicotine. ~fter three or four days in a particular experi~ntal s~king condition, a circadian blood sa~llng study ~as perfor~d ~n order ~o es~[~te daily nicotine ~ntake. An ~nd~elI~ng T108350798

113 catheter was inserted into a forests vein and blood samples were collected every t~o hours. Ti~e of blood sampling was independent o£ when the subject saoked his last cigarette. In a group of 22 subjects, including 13 men and 9 vo~en, ages 22 to 55. we found that daily intake of nicotine averaged 37.6 ~g (± 17.7, S.D.). The range was broad - 10.5 to 78.6 mg - sad there was no difference between and women. The daily intake of nicotine correlated significantly with the v~mber of cigarettes smoked per day (R2 = 0,35) but not with United States Federal Trade Commission (FTC) smoking m~chlne yield of nicotine. Average intake of nicotine per cigarette was about 1.0 mg, similar to the average smoking machine yield, but ranged from 33 to 155Z of machine nico- tine yield. The estimated nicotine intake per cigarette was not signifi- cantly correlated with machine yield. That the number of cigarettes smoked per day accounts for only 35% of variance of nicotine intake is consistent with the idea that how cigarettes are smoked is much more important than how many are smoked. The lack of correlation between FTC smoking machine yield and intake of nicotine per cigaret.te can be understood because people smoke cigarettes very differently from machines. How well do people regulate their intake of nicotine when smoking different commercial brands of cigarettes, and if they do, how do they do it? That is, how much compensation occurs by smoking different numbers of cigarettes and how much by adjusting the intake of nicotine per cigarette? Eleven subjects were tested in a study comparing smoking their usual brand versus Camel~ 85 mm filtered (¥TC nicotine 1.2 ms), and versus True~ 85 mm filtered (FTC nicotine 0.4 W~). The latter two were smoked by different ~ubjects in a balanced o~er adjusted for sex. O~ average, subjects consumed 35 ~g nicotine per d*y while s~oking their o~n cigarettes and 26 mg per day while smoking True or Camel. There was no difference in nicotine intake betwee~ the latter two clgare~tes.Examining nicotine intake ~s a function of n~er of cigarettes smoked and nicotine intake ~r cigarette, we f~nd ci8arettes ~ile ~mki,g Tr~e, but the sam number ~hile ,moki~ us,al brs~d compared with Ca~l, and (2) the nicotine intake per cigarette was for the usu~1 brand (average 1.2 ~), next highest for Ca~l (average 0.9 ~) and l~est for Zrue (0.7 efficiently, perhaps related both to ~intaininz a s~t level of nlcoti~e in the body as ~11 as for the pleasure o~ smoking a cigarette ~nich they enjoy. Nl~en seitched to other ci~arettes~ they maintain a level at about T108350799

two-thirds of that of their o~n cigsrette, b~t =~intaln the level regardless of ~hat cigarette they are s~king. ~en ~i~ch~ ~o a cigarette yieldin~ c~parable or higher levels o~ nicotine, they ~ked the s~ n~er o~ cigarettes but s~ked less e~icien~ly. ~en switched to a I~ yield cigarette, ~hey ~ked ~re cigarettes and also ~k~ the cigaret~e~ ~re efficiently co~ared to the predicted s~k£~ ~chine yield. ]f individuals can compensate so well ~hen switching brands, does machine yield predict i~take in a large population of smokers? To expiore this question, we measured blood concentrations of cotinine, the primary metabolite o£ nicotine, in 272 persons entering a smoking cessation treatment program (6). Ne found that blood cotinine was significantly correlated with number of cigarettes smoked per day (R2=20%), but not with FTC smoking machine determined yields. Ne conclude that FTC m~chine yields of nicotine poorly predict the intake of nicotine by the cigarette smokers, and that a given smoker can obtain whatever level of nicotine he is seeking from commercial cigarettes irrespective of the nominal yield. In summary, the information gained in studies of the pharmacokinetics and pharmacodynamics of nicotine given by intravenous infusion have provided insight into the time course and effects of nicotine during daily cigarette smoking, and have provided a tool for estimating the intake of nicotine during daily cigarette smoking. Future studies of this type may elucidate reasons for indlvidual differences in susceptibility to adverse effects of cigarette smoking and in patterns of tobacco consumption. ACr~K~LED~ENT The work reported in this paper was supported in part by Grants No. DA02277, DA01696, CA32389 and HL29476 from the National Institutes of Health. '! !. I. Armltage AK, Dollery CT, George CF, Houseman Tl~, Lewis PJ, Turner DM. Absorption and metabolism of nlco~ne from cigarettes. Br Med J 1975; 4: 313-316. 2. Isaac PF, P~nd 14~. Cigarette smoking and plasma levels of nicotine. Nature 1972; 236: 308-310. Benow.its NL, Jacob P IIl, Jones ~T, Rosenb.erg J. Inter-i~ividual variability in the metabolism and cerdiovascular effects of nicotine man. J Pharmacol Exp Ther 1982; 221: 368-372. Rosenberg J, ~enowit~ HL, Jacob P II~, Nilson K~. Disposition kinetics ~ e~c~ ~f ~r~"~ ...... ~or~.~. Clln Pharmacol Ther 198,0~ ~; 517-522. TI08350800

5. ~eno~itx WI~, K~yt F, Jacob P III. Circadian blood nicotine concentra- tions during cigarette smoking. CI£n Pharmacol Ther 1982; 32: 75B-764. ~enowit: I~L~ Hall SM, l~erning RI, Jacob P Ill, Jones RT, Osm~n A-L. Smokers of low-yield cigarettes do not consume less nicotine, l~ew Eng J MAd 1983; 309: 139-142. T108350801

117 P~ AR~ P~S 0~' "I'OBACCO USE: SHIFTS A~) DIRECTIO~ Virginia Cresswell-Jones, M.Ed. Consultant, Georgian Bay Centre Addictio~ Research Fo~n~atio~ 13 Collier Streee ~arrie, Ontario Canada L~M IG5 IN~ROI~rloN In recent years, studies of the effects on pregnancy of drug use, including tobacco, alcohol, caffeine and over-the-counter items, have raised concerns about adverse effects for both the fetus and the mother (1,2,3). Many sources, including professionals, media and advertising, provide drug information and influence the drug use of pregnant women (4). They continue to be exposed to a variety of potential teratogens, the majority being non- prescription drugs, such as tobacco (5). Studies have indicated a decrease in drug use during pregnancy, but even diminished use continues to be of concern (4,6). Some have attributed this decrease to physiological changes that affect the desire for substances, as well as to concern for the welfare of the baby (7). More information is needed on what can support and maintain a change of habit toward low or non-usage of tobacco and other drugs, in conjunction with the natural phenomena of pregnancy. This report provides data on tobacco, alcohol, and other drug use in a group of pregnant Ontario women. It was expected that, generally, the patterns of tobacco and drug use would demonstrate a change to reduce intake during pregnancy, suggesting that it could be a natural intervention point for prevention education and for the motivation to sustain the change toward cessation. This pilot study conducted in 1980 was to document reported patterns of use, rather than actual amounts of use. This was the first study of drug use in pregnancy among women in Simcoe County, Ontario. The views expressed in this publication are those of the author, and do not necessarily reflect those of the Addiction Research Foundation. T108350802

118 An anonymous self-reporting questionnaire was completed by 132 volunteer pregnant women. They were interviewed in hospitals, malls, prenatal clinics (the largest sub-group) and at home. Respondents reported on their use of alcohol, analgesics, antacids, cannabis, caffeine, tobacco, tranquillizers and sedatives, in frequency, quantity and more than usual or "binge" use arourtd three time periods in pregnancy, namely "before" (6 months prior), "d~rinS" (pregnaacy) aud "aut~cipated-afcer" (6 months after birth). The "after" phase was a self-prediction of intended future use with the newborn at home. Additional data were obtained on demographics, on the reasons for change in tobacco use and knowledge of the effect on the baby of personal tobacco use. The total group (132) were of ages 18-35, of good educational level and from a wide range of occupations. Most were in the age group 22-25 (36%) and 26-30 (26%). This was the first pregnancy for 67%, while for 27% and 11% it represented the second and third respectively. Most women had some high school (70%) and 26% had some post-secondary education or had completed it. The highest frequencies of occupations were teacher (11%), secretary (i0%), assembly line (I0%), and homemaker (9%). Generally, the subjects reported low use of drugs. During pregnancy they indicated a reduction, on average~ in the frequency and amount of drug use. The one exception is a marked increase in the frequency of use of antacids. During the after pregnancy phase, the pattern reversed to increased drug use at slightly lower levels than in the before pregnancy phase. Using a code of I: never/rarely, 2:i-2 yearly, 3:3-4 yearly, 4:I-2 month- ly, 5: 3-4 monthly, 6: i-2 weekly, 7: 3-4 weekly, 8: I-2 daily, for the frequency of use, the phases of before, during and after show tobacco respectively at 4.57, 3.46 and 3.64, with alcohol at 4.28, 2.59 and 3.98, caffeine at 7.64, 6.70 and 7.48 and antacids at 2.05, 2.97 and 1.76. In amount of use, the phases show tobacco at 2.14, 1.73 and 1.80, alcohol at 2.87, 1.77 and 2.63, and caffeine at 2.46, 2.22 and 2.35. The appropriate code is I: none, 2: one, 3: two, indicating unit use e.g. pill. The pattern of drug use in "binges" was similar, for during pregnancy drug use dropped for all drugs except antacids, which increased somewhat. The hi,heat average dru~ use for tb~ pregnancy phases was of caffeine, alcohol and tobacco. Data on tobacco use in the total group indicate before pregnancy smokers at 49% and those that never/rarely use it at 51%. During pregnancy the smokers comprised 35% and never/rarely 64%. Smokers named two major reasons for cnan~e aur~ng prega~.ncy, that oz less aeslre, and concern for tr~ DaVy. Xn after pregnancy, "baby" drops out as a major change factor. The total group response to "In your opinion, can your baby be affected by your use of tobacco? was Yes 80%, No II~, Maybe 6~, and Don't Know 3%. The majority appeared to have some knowledge; however, the 20% who replied otherwise present a challenge for pregnancy and drug education. T108350803

119 Ybe results indicate a change in tobacco and drug use pattern~ in pregnant women. Yhe direction of this change is to less use or non-use with the exception of antacids. The responses were averaged to give a general p~ttern of use. Many women were already low or non-users of drugs. During pregnancy, alcohol dropped noticeably in frequency of use. Some women comnted on the disagreeable taste of alcohol or that they did not feel llke drinking. In the after phase, ~obacco use changed little from during pregnancy. It is suggested tha~ some drugs might be influenced by maternal bodily changes. Nearly three-quarters of the group were primigravida and their concerns for their first baby provided one of the major reported reasons for change in drug use pattern. This concern dropped markedly in the anticipated-after phase, along with an increase in desire for smoking and other drugs. The effects on the infant of continued use of tobacco within the home appeared not to be fully understood. It is suggested that pregnancy, especially the first, offers a potentially useful opportunity for prevention education. The duration of pregnancy is a natural intervention point for the woman and her family. Concern for the baby, alone or combined with physiologic changes, may be an important motivating factor. The role of community agencies in education, support and reinforcement of a healthy lifestyle in association with pregnancy should be studied further. The author wishes to thank Dr. M.J. Ashley, Professor, Faculty of Medicine, Department of Preventive Medicine and Biostatistics, University of Toronto for guidance and review. Clarren ~l~, Smith DW. The fetal alcohol syndrome. New Eng J Med 1978; 298(19): 10'63-1067. 2. Doll g. Hazards of the first nine months: an epidemio!ogist's night- mare. J Irish Med Assoc 1973; 66(5): 117-12.6. 3. Rantakallio P. The effect of maternal smoking on birth weight and the subsequent health of the child. Early Rum Dew 1978; 2(4): 371-382. 4. Stortz LJ. Unprescribed drug products and pregnancy. JOGN Nursing 1977; 6: 9-13. Hora JJ, ~ora AH, So~merville RJ, Hill exposure to potential teratogens. JAI4A 1967; 202(12): 91-95. Maternal "

6. Rook EB. Changes in tobacco s~oking and ingestion of alcohol and caffeinated beverages during early pregnancy: are these consequences, in part, of feto-protective ~echanis~s diminishing ~aternal exposure to embryotoxins? In: Kelly S, Rook EB~ Janerick DT, Porter IH, eds. Birth defects: risks and consequences. New Y0rk: Academic Press 1976; 173-184. association wi~h nausea and vom~E~ng during pregnancy. Acta Obstet Gynecol Scand 1979; 58: 15-17. T108350805

121 R.C. Frecker, M.D., Ph.D. Institute of Biomedical Engineering and Department of Pharmacology University of Toronto and Addiction Research Foundation Toronto, Ontario, Canada MSS IIITEODUCTIOH The importance of tobacco as a cause of morbidity and excess mortality has been thoroughly documented (1,2,3). The prevalence of tobacco use (especially as cigarette smoking), the continuing high incidence of use in the young, and the extreme recidivism among users who have attempted to quit - in light of the well known risk to health - argue persuasively for the addictive nature of the habit. Indeed, in its most recent Diagnostic and Statistical Manual of Mental Disorders (DSM-III), the American Psychiatric Association has included both tobacco dependence and tobacco withdrawal as diagnostic entities. The latter is categorized as a 'substance use disorder', and the former as an 'organic mental disorder'. These entities are characterized by a variety of observable phenomena (4). In order to examine systematically the role of pharmacological modulators in tobacco dependence, we have chosen to investigate the effects of various smoke constituents on the function of the oculomotor control system. To this end, there have been developed methods for generating and delivering a sub-micron nlcotlne-containing aerosol (5), and for measuring precisely the infrastructure of both saccadic eye movements and the miniature movements associated with fixation (6,7,8). There is a theoretical basis for suggesting that certain eye-movement parameters will be sensitive to drugs which alter arousal and, possibly, cognitive function (9,10). This paper discusses some of the issues related to measuring the effects of drugs on the human brain, and focuses particularly on the special problems which exist when attempts are made to study the effects of aerosolized nicotine. General reviews of drug effect measurement techniques are avail- able elsewhere (I]-|5). ~~KLI~Wtq,CS vs PE~RJ~CODYRAMICS Both these terms imply motion or action on the part of the drugs, but they enjoy a particular usage in pharmacology which should be defined. Phamaco- kinetics is the study of the movement of drugs into, through, and from an T108:35080~

organism or syst~a, including analysis of data related to the absorption, distribution, ~etabolis=, a~t excretion of drugs a~ ~a~li~es, a~ £o ~he ~i~ course of drug effects - [haE ~s, ~he s[udy of ~he effects of organis~ on drugs. Pharmacodynamics, on ~he o~her hand, ~chanis~ of drug ac~ions, and of khe responses of an organism or system ~he admini~ra~i~ of drugs, including analysis of ~he rela~ionship between drug concentration and effect - ~ha~ is, ~he s~udy of ~he effects of drugs ~ organ~s~ (~I). I~ shoald-be e~asi~ed ~ha~ ~hese ~ pharmacological approaches are used ~oge~her ~o explain ~he behaviour of drugs; and ~ha~ real life it is no~ possible ~o ~ease them apar~ as clearly as ~he defini- ~ion suggest. Clearly, ~o act, a drug needs ~o reach i~s si~e of action within ~he body. P~CODYIIAI~C ~'~ffR~I~NT SYSTEMS As it relates to drug-effect measurement, the man-machlne interface presents a number of ergonomic and other design challenges. Complex biological systems (such as man) contain many control systems which are highly integrated and usually autoregulatory. Developing a quantitative and predictive relationship for particular variables and the biological process which they represent is non-trivial, especially where the perturbation produced by the drug administration tends, homeostatieally, towards former values. Although this adaptive control process can be modelled, it is frequently non-linear; and input-output relations are incompletely under- stood at the present. Further, control parameters change over time in response to a multitude of internal and external sources of 'noise' Organisms, in fact, do not 'like' being probed, and interfaces which are established for this purpose are frequently unstable. Ergonomlc (humsn factors) considerations are often paramount, and ecological validity can be difficult to secure. Ideally, a pharmacodynamlc measurement technique should not pervert the response. Addltionally, sensing the dependent varlable should be without subject constraint or awareness; without attachment to the subject; and should be objective, automatic, rapid, accurate, and sufficiently precise to resolve in time the anticipated change(s) produced by the pharmacologic- ally-active ~ub~tance being studied. Further, the dependent variable should have low complexity for ease of extraction of single response elements; high s~nsitivity and selectivity for the drug being studied; low sensitivity to extraneous 'environmental' infIue,ces such as heat, light, sound, pressure and temperature; low sensitivity to changes in diet, exercise and sleep patterns; and low between- and within-day variability. If there is an infradian, circadian, or ultradian rhythm associated with the dependent variable, it must be characterized. The variable being measured should derive directly from the relevant behaviour, a~4 be presented in an 'ecologically valid' manner. Failing this, it should De an involuntary parameter requlrinK minimal (if any) subject participation. Of the above attributes, highly @pecific drug sensitivity coupled to high selectivity, and low sensitivity to extraneous influences, are the most difficult to achieve. Even when the above considerations have been taken into account, there remains the need to correlate observed changes under drug influence with

other known indicators of a particular drug's effect (at least for purposes of callbration). It should be possible to correlate observed changes with the 'effective' drug concentration at the suspected site of drug action. That this is probably the concentration of 'unbound' drug within the central nervous system (C~$) further complicates the matter. Corrections in timing mast also be made to accommodate pharmacokinetic phenomena if one is relat- ing observed changes in the dependent variable to drug concentration at other than the site of action. This is especially salient in cases where CNS effects are correlated with plasma drug levels, sometimes without regard to protein binding, distribution, and re-distribution phenomena. Output signals representing change in the dependent variable need to be readily amenable to available signal reduction and analysis techniques. This in turn will have implications for the computational hardware, and both systems and applications software. Data collection and pre-processing (if not analysis) should be on-line, and in real time. At the end of a given experiment raw data should be entered onto the storage medium from which subsequent data access will occur. Presentation of relevant numerical data should be possible within the time frame of the experiment, or very shortly thereafter (for those not required to be blind to the data). If possible, at least the summary statistics and preliminary graphlcal materials should be available shortly after the experiment is completed. TOBACCO P~COD~CS Oeneral considerations Notwithstanding the general applicability of what has been said above, there are other concerns which relate particularly to tobacco pharmacodynamics. These are rather specific in the case of cigarette smoking, which is said to be the most dependence-producing form of tobacco use (16). To study this effectively, it is necessary to mimic both the route and profile of administration. As constituent exposure profiles from cigarettes are diffi- cult to quantify and control reproduclbly, a high-output, sub-micron nicotine aerosol generator with precise monitoring and control capabilities was developed (5). With this device it is possible to deliver, in a double blind fashion, controlled individual and cumulative doses of nicotine, the constituent currently thought to be responslble for the addictive properties o5 tobacco (16,17). While such an aerosol permits the administration of nicotine alone, this may also be achieved through the use of intravenous nicotine or nicotine chewing gum. However, in such cases, the pharmaco- kinetic profiles will differ from those observed with cigarettes. One ~atter of particular importance in evaluating the addictive process (acquisitions maintenance, cessation) involved in smoking is resolving the question of whether or not the pulmonary route of administration is as critical as is suggested by the 'bolus hypothesis' (17). Another issue requiring consideration in studying tobacco pharmaodynamlcs is that of the ethical acceptability of using experimental designs in which naive or form, is unlikely to be a sidle phamacological phenomenon. There are fold behavioural and sociobehavioural aspects which are of great importance. TI08350808

Sl~Cif~c co~i~iers~io~s In addition to other pharmacokinetic complexitles mentioned above, there is significant difficulty associated with rendering dose a truly independent variable. There are many substances in tobacco, and particularly in tobacco smoke. As a result, the slmulation of smoking behavlour (even of nicotine and carbon monoxide alone) presents real technical challenges in aerosol generation, characterization, and delivery. To this one wast add the paucity of objective, sensitive and selective measures of nicotine's central (euphoriant?) effect. There is the further requirement for very 'rapid' dependent variables, capable of resolvlng within- and between-puff effects. This may prove to be of special importance during the distributional phase following inhalation, during which there is a rather rapid change in brain nicotine concentration over a very short interval (5-30 seconds). No comment will be made here on the stringent technical demands associated with performing and interpreting plasma and other assays for nicotine and its metabolites (18,19). Tobacco constituents of interest While nicotine is the 'prime suspect' as the principal pharmacological modulator of smoking behaviour, acetaldehyde is produced in significant quantities in cigarette smoke (20), and the interaction of this gaseous phase constituent with endogenous neurotransmitters (e.g., dopamine) has attracted the speculation of the scientific community (21). It is possible that condensation products of acetaldehyde may provide a link with the endorphin system. Carbon monoxide, while possessing no receptor-mediated central activity, does bind to hemoglobin, and at higher smoking rates may produce a relatively anoxic state, with perceptible central effects. It is not presently known if such effects are positively reinforcing. Carbon monoxide also binds to the cytochrome oxidase enzyme system which is primarily responslble for the conversion of nicotine to cotinine and nicotine-l-oxide (16,22,23). It is desirable to determine target organ exposure to particular constituents. This presents issues related to constituent availability (amount presented) which include the relevance of mouth-level and lung- level exposure m~asurements, and the interpretation of bound and un-bound peripheral plasm~ levels, which will not be discussed here. S~cron nicot£ne aerosol generator In an effort to solve problems associated with dose delivery, monitoring, and control, a 3-channel modified Collison nebulizer was designed and fabricated (24,5). This device incorporates an impact baffle system which precipitates larger aerosol particles, and is operated from pressurized medical grade air. The aerosol is generated continuously and, under micro- processor control, injected into the insplratory air stream during the appropriate respiratory phase. Some 70% of the particles in the 8 I/min output are below I.I microns in mean mass aerodynamic diameter (MMAD). The of nicotine b~se-equiva!ent per ml of air (or 56 ug per 35 ml ~puff'). The 'puff' volume is adjustable from 0 ml to 500 ml, with a 'puff' duration of from 1 to 5 or more seconds. T108350809

Nhy s~:udy ~ob~cco e££eel:a vith eye ~ove=~nts? Various eye-movement parameters have been shown to be sensitive to psycho- active drugs such as dlazepam, pentobsrbital, amphetamine, and alcohol, and dose-response relatio~ships were usually seen for such parameters as dura- tion, peak velocity, and magnitude (9,25-28). The author and co-workers have developed a high precision, non-contacting, high-speed, real time eye tracker for use in a clinlcal environment, which is suitable for measuring a wide range of eye movement characteristics (6-9). The approach used meets most of the criteria outlined above for general pharmacodynamlcs, and speci- fically addresses the needs for high spatial and temporal precision in studying the effects of inhaled nicotine. The current apparatus generates new positional coordinates in both the vertical and horizontal directions every millisecond, and has been designed to accommodate various forms of administration of tobacco and tobacco constituents. Properties of hu~n saccadic eye ~ovements Saccades are abrupt, high-velocity shifts in gaze which occur to re-fixate a visual target which has moved (29). They occur for target displacements greater than 0.3 degrees, and in response to target motion at rates higher than 30 degrees per second. Pursuit eye movement, on the other hand, serve to maintain alignment of the eyes on a visual target when it moves at slower velocities (29). Saccadic eye movement parameters are highly characteristic of individual subjects, and do not show excessive variability among groups of normal subjects (9). While parameters co-vary (e.g., velocity as a func- tion of magnitude), for given stimulus conditions these movements manifest low variability. Typical parameters of interest, with ranges shown in brackets, are (a) latency (100-200 msec), (b) duration (10-150 msec), (c) magnitude (0.3-90+ degrees), and peak velocity (I0-i000 degrees/set). Saccades may begin as either voluntary or reflex movements, but the trajec- tory is largely pre-programmed in response to the initiating stimulus (29). Thousands of saccades may be performed in rapid succession without eye muscle fatigue (9). ~iversity of Toronto eye tracker The current eye trucker was developed by the author and his co-workers at the Institute of Biomedical Engineering [IB:~E], University of Toronto (6-9;30). The subject is seated before the table-mounted device and views the stimulus through a mirror which reflects only infrared (IR) light (dlchroic mirror). The stimulus is created by a moving visible red laser beam which form~ a I mm dot on a reflectant surface placed 50 cm from the corneal surface. A collimated beam of IR light is reflected from the dichrolc mirror onto the cornea and a retro-corneal image of the IR light source is formed. In turn, this image is focused via a series of lenses and mirrors onto two discrete phototransistor arrays. When the eye moves (rotates), the retrocorneal image moves proportionately on the corresponding tight sensitive-arrays ~ormia~ the "x' a~ 'y' rectangular coordinates of the ~ve~nt vector. A u~ique s~nal processln~ technique extracts eye position 1,000 ~i~s ~r second, with a precision of 6 arc seconds of ~ular T108350810

126 is generated in the terminal 200 usec of each 1 msec sample period. Position information from the hard-wired signal processor is received on 16 digital lines by a DEC PDP 11/34 minicomputer which extracts the various eye ~ovement parameters as each eye motion is completed. Special attributes of the IK~K eye tracker The precision obtained with this technique is equal to or better than the best obtained with fitted contact lenses (31). New position and velocity data are obtained with a frequency which permits drug-induced velocity changes to be monitored in a virtually continuous basis. Complete eye- movement profiles can be 8enerated as often as the eye can respond to a new target position; and miniature movements which occur when the eye is nomi- nally stationary are easily detected and characterized. The saccadic system has an 'obligatory' latency of response in the order of 200 msec, and this physiological constraint is the only one which limits the number of stimuli which may be presented to the visual system with the expectation of a response being made. The instrument is ergonomically designed for human clinical trials (alignment, calibration, data acquisition, and data analysis are all performed automatically). A package of specially-developed graphic- al and statistical programs provides for ease of data manipulation. The level of infra-red light used is extremely low (some 300 uw/square cm) and this permits continuous monitoring with ocular comfort and safety. OVERVIEW OF IZOTORE KXPERIMEI~TA]~ PLANS A database of normal values for the various movement parameters will be generated using the apparatus described above. Initially, a group of healthy male smokers, and matched non-tobacco users will be investigated to characterize within- or between-day variability. Acute dose experiments in smokers will then be conducted to obtain optimum stimulus/response condi- tions for detecting dose-related effects of nicotine given ~n smoke and in plain aerosols. This will permit an analysis of the pulmonary pharmaco- kinetics of aerosolized nicotine and a comparison between the cigarette and ~he new aerosol generation system as dr~g-delivery systems. A group of smokers will then be studied under conditions in which their daily exposure to nicotine will be progressively reduced to a 'lower boundary' (32) at which withdrawal symptoms and signs may be expected to become manifest. During this phase resultant parameter changes will be cha'racterized. The stability of the lower boundary condition and any changes in tolerance (if manifest) will be assessed. Daily exposure will be reduced to the lowest acceptable level for given subjects, which in some cases will be zero. Within the bounds of ethical constraints, it is proposed to examine toler- ance and withdrawal phenomena at lowest levels of exposure, and to examine parameter changes in cases where subjects return to a h[~her level of nicotine exposure. In each of the experiments outlined above, other varia- bles such as heart rate, blood pressure, electroencephalogram, tremor, standin~ steadiness, priate i,tervals. T108350811

The above approach has been taken because of the ethical difficulties of chronically (or even acutely) exposing naive subjects to a drug (nicotine) which has a demonstrated and high addiction liability. The program outlined represents a number of clinical trials which will be completed over a period of years. In the meantime, instrumentation development is aimed at helmet- mountin~ the apparatus for progressively greater ease of use. The essential stimulus control and data acquisition software will be implemented on a small dedicated computer for ease of portability, and application in environments where the use of a larger machine is inconvenient. In general~ the technology described for eye movement measurement is well adapted to applications requiring high-precision, real-time interaction with the human visual system. This is particularly so where control of an exter- nal system or environment is contingent on eye position, velocity, or acceleration. The ability to miniaturize the essential components makes possible applications in which limitations on mass or volume occupied are critical. The investigation of tobacco dependence is an important specific example of how this technology has been applied in the clinical research environment. U.S. Public Health Service. Smoking and health. A report of the Surgeon-General. Rockville, Md.: U.S. Dept. Health, Education and Welfare, 1979. (DHEW publication no.(PHS) 79-50066). Royal College of Physicians. Smoking or health. A report of the Royal College of Phvslcians.. London: Pitman Medical, ~,7.~" World Health Organization. Controlling the smoking epidemic. Report of the WHO Expert Committee on Smoking Control. Geneva: World Health Organization, 1979; Tech Rep Series #636. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 3rd ed. Washington: ,American Psychiatric Associa- tion, 1979. Lux JE, Frecker RC. The generation of a nicotine eerosol for inhala- tion. Manuscript submitted for presentation at the llth Canadian Medical and Biological Engineering Conference, Kingston, Canada, June 1985. Eizenman M, Frecker RC, Hallett PE. Precise non-contacting measurement of eye movements using the corneal reflex. Vision Research 1984; 24(2): 167-174. Frecker RC, Eizem~n M, Hallett PE. ~igh-precision real-tlme measure- merit of eye position using the Ist Purkinje image. In: Gale AG, Johnson F, eds. Theoretical and applied aspects of eye movement research. Amsterdam: North Holland Publishing Com~pany, 1984: 13-20. T108350812

Ei~en~an H. Precise non-contacting eye-movement ~onitoring system. PhD Thesis, Department of Electr~cal Engineering, and Institute of Biomedical Engineering, University of Toronto, 1983. Frecker, RC. Effects of diazepam, pentobarbital, and dextro amphetamine on human saccadic eye movements. PhD Thesis, Department of Pharmacol- ogy, a~ Institute of ~iomedical Engineering, ~nlversity of Toronto, 1973. I0. Leigh RJ, Zee DS. The neurology of eye movements. Philadelphia: F.A. Davis Company, 1983. 11. Frecker RC. Problems of pharmacodynamic measurement related to psycho- active drug interactions in humans. Washington: Department of Health and Human Services, National Institute on Drug Abuse, Research Monograph Series. In press. 12. Linnoila M, Erwin CW, Cleveland WP, et al. psychomotor performance of men and women. 39(5): 745-758. Effects of alcohol on J Studies Alcohol 1978; 13. Netter P. Somatic factors as predictors of psychotropic drug response. In: Janke W, ed. Response variability to psychotropic drugs. Oxford: Pergamon Press Lid, 1983: 67-95. 14. Itil TM. Effects of psychotroplc drugs in qualitatively and quantita- tively analyzed EEG. In: Clark WD, Del Giudice J, eds. Principles of psychopharmacology. New York: Academic Press, 1978: 419-443. 15. Brodie BB, Mitchell JR. The value of correlating biological effects of drugs with plasma concentration. In: Davie DS, Priehard BNC, eds. Biological effects of drugs in relation to their plasma eoncentratlons. London: Macmillan, 1972: 1-12. 16. Russell MA~. Tobacco smoking and nicotine dependence. In: Gibbins RJ, et al, eds. Recent advances in alcohol and drug problems, Volume 3. New York: John Wiley & Sons, 1976: 1-47. 17. Eussell MA~, Feyerabend C. Cigarette smoking: a dependence on high nicotine boli. Drug Metab Rev 1978; 8(I): 29-57. 18. Benowitz N, Jacob P, Wilson M. Improved gas chromatographic method for the determination of nicotine and cotinine in biologic fluids. J Chromatography 1981; 222: 61-70. 19. ~enowitz N, Jacob P, Jones ~T, Rosenberg J. Interindividual variability in the metabolism and cardiovascular effects of nicotine in m~n. J Pharmacol Exp Ther 1982; 221(2): 368-372. 20. Wynder EL, Roff~u D. Tobacco and tobacco smoke. Press, 1967. New York: Academic T108350813

21. Pomerleau OF, Pomerleau CS. Neuroregulatora and the reinforcement of smoking: towards a behavioural explanation. Neurosci Biobehav Rev 1984; B: 503-513. 22. Sellers EM, Frecker RC, Romach ~1. Drug metabolism in the elderly: confounding of age, smoking, and ethanol effects. Drug Metab Rev 1983; IA(2): 225-250. 23. Frecker RC. The effects of tobacco smoking on drug metabolism. Canadian Pharmaceutical Journal 1984; i17(4): 158-162. 24.. British Standards Institution. British standard for methylene blue particulate test for respirator canisters. London: British Standards House, 1955; BS 2577: 3-38. 25. Wilkinson IMS, Kine R, Purnell M. Brain 1974; 97: 785-792. Alcohol and human eye movement. 26. Rothenberg SJ, Selkoe D. Specific oculomotor deficit after diazepam. I: Saccadic eye movements. Pychopharmacology 1981; 74: 232-236. 27. Rothenberg SJ, Selkoe D. Specific oculomotor deficit after diazepam. II. Smooth pursuit eye movements. Psychopharmacology 1981; 74: 237-240. 28. Westheimer G, Rashbass C. Barbiturates and eye vergence. Nature 1961; 191: 833-834. 29. Davson H. Physiology of the eye, 4th ed. London: Churchill Livingston, 1980. 30. Greenberg WN. An automated, high resolution eye movement transducer for clinical research. MASc Thesis, Department of Electrical Engineering, and Institute of Biomedical Engineering, University of Toronto, 1978. 31..Young LR, Sheena D. Survey of eye movement recording methods. Behaviour Research Methods & Instrumentation 1975; 7(5): 397-429. 32. Kozlowskl LT, Herman CP. The interaction of psychosocial and biological determinants of tobacco use: more on the boundary model. J Appl Social Psychol 1984; 14(3): 244-256. T108350~14

OF ~ECO USE Seymore Herling, Ph.D. Lynn T. Eozlowski, Ph.D. Clinical Institute Addiction Research Foundation Toronto, Ontario Canada MSS 2Si Although nicotine has long been suspected as being the primary constituent of tobacco responsible for the maintenance of smoking, it had often been difficult to demonstrate experimentally the importance of nicotine in controlling tobacco smoking behavior. In studies involving human subjects, the role of nicotine in smoking has been assessed in a number of different ways. Various indices of smoking behavior have been measured, Ill in response to changes in the nicotine yields of cigarettes (i), [2] following preloading with nicotine hy routes other than smoking e.g., intravenous, oral (2-5), [3] in response to blockade by the centrally-acting nicotinic antagonist mecamyalmlne (6), and [4] following the manipulation of the urinary excretion of nicotine by manipulatlng urinary pH (7). More recently, the role of nicotine in cigarette smoking has been examined in human volunteers given the opportunity to self-administer intravenous nicotine (4). While most studies in which nicotine-lnduced changes in smoking have been measured, have shown some degree of nicotine regulation (e.g., compensatory smoking - the number of cigarettes smoked is generally inversely related to the nicotine yield of the cigarettes), the resultant changes in smoking behavior are often small (8,9,10). Schachter for example (9), showed that heavy smokers smoked only 25% more low nicotine-yield (0.3 mg) cigarettes than cigarettes containing more than 300% more nicotine (nicotine yield = 1.3 mg). Results such as these implicate factors other than nicotine, both pharmacological (carbon monoxide, perhaps) and noupharmacological, as being important determinants of cigarette smoking. The importance of nicotine in maintaining cigarette smokiu~ in humans had also been questioned because o~ the difficulty in demonstrating consistent reinforcing effects of nicotine in animals. Drug self-administratlon procedures in animals (in which laboratory animals learn to nmke operant responses, such as lever presses, or appropriate turns in a maze, which result in drug delivery to the animal) are often used to assess the abuse llability of drugs. In general, drugs which serve as reinforcers of operant respondin~ in animals are drugs of abuse in humans (e.g., cocaine, amphetamine, morphine, alcohol). In the earliest studies of nicotine self- administration in rhesus monkeys, nicotine was shown either not to maintain above those maintained by ~line (II-14). In most of these s=udles, nicotine T108350815

132 injections ~ere available on a continuous reinforcement (CRF) schedule (i.e., each lever press m~de by the animal resulted in ~n injection of nicotine) 24 hours per day. Under these conditions, most injections o~ nicotine occurred during the daytime or during the 12-hour light phase of the light-dark cycle; the rate at which nicotine was self-lnjected only slightly exceeded rates maintained by saline; and, response rates maintained by nicotine generally did not vary as a function of nicotine dose. ~ore receotly, however, it has become clear ~hat (just ~ ~th ~re co~nly acknowledged drugs of abuse) behavioral history and current access condi- tions (i.e., the way in which injections are scheduled to occur) are impor- tant determinants of nicotine's behavioral effects. For example, nicotine appears to maintain responding more effectively under intermittent schedules of relnforce~nt (i.e., when behavior results ~n nicotine injections occur- ring every 4 ~o 6 minutes) than when the drug is continuously ~v~ilable (15). Indeed, recent studies in squlrrel monkeys have sho~ that whet~er nicotine, [I] acts to increase behavior leading to its adminis=ratlon (i.e., serves as a positive reinforcer), [2] acts to increase behavior leading to its termination or postponement (i.e., serves as a negative reinforcer)~ or [3] acts to decrease behavior leading to its administration (i,e., serves as a punisher), is dependent in an important way on the ongoing behavior of the ani~l and the schedule under which nicotine is delivered (15). These diverse functions of nicotine cannot be explained simply by different nicotine dosing schedules, since the sa~ doses of nicotine (I0-I00 ug/kg/ injection) could either ~intaln or punish responding depending upon the prevailing environmental conditions. These results in animals suggest that the way in which nicotine might func- tion to control smoking behavior in humans is also determined by past and current environmental conditions. Hennlngfield et al. (4) have recently shown that human subjects will self-administer intravenous injections of nicotine. Moreover, in subjects with histories of drug dependence (includ- ing cocaine abuse), intravenous injections of nicotine were identified as cocaine from a list of commonly abused street drugs. One interesting aspect of these studies in humans was that although self-administered injections of nicotine produced dose-related increases in self-reported "drug strength" and "drug liking" scores, nicotine, in the same subjects, also produced dysphoric effects (e.g., burning sensations at the site of injection, momentary shortness of breath, ~eetlngs of fear, ¢oughlng~ sweating, and nausea), which became more intense with repeated injections over the course of the 3-hour experimental session. Yet despite these untoward effects, most subjects continued to self-administer nicotine (4). In some subjects, however, rates of responding maintained by nicotine were lower than those maintained by saline, suggesting, as in the animal studies~ that in these subjects nicotine served to punish behavior leading to its administration. When these subjects were tested under conditions in which lever-press responding resulted in the postponement of nicotine injections, nicotine (but noc saline) resulted in high rates of avoidance responding, so that few or no nicotine injections occurred (16). Thus it appears that, in humans, nicotine can serve as either s positive or negative reinforcer and produce concomlttant self-reports of liklng or disliking. These functlonal effects nicotine (i.e., s~eating, respiratory distress, n~us.ea, etc.), since these latter effects tend to occur irrespective of whether nicotine serves as a T108350816

~IC~.- ~l) PSTC~LO~ICAL F~fOe~ 133 positive or negative reinforcer, or h~ether nicotine produces increases in liking or disliking scores (16). These preliminary studies of intravenous nicotine self-administration in humans indicate that nicotine may function in different ways to control smoking behavior. The results from the animal self-administratlon studies suggest that nicotine's control of ~Jman smoking is modified by environmental conditions such as behavioral history and the schedule of nicotine availability. One way in Which ~ironmental conditions have been sh~n to affect respond- ing maintained by nicotine (and other drugs and reinforcers) in animals, has been in studies of second-order schedules (17). Under second-order sched- ules, lever pressing produces environmental stimuli (e.g., brief visual stimuli) which are only occasionally associated with the intravenous injection of drug or the presentation of some other reinforcing stimulus. Table 1 compares results from different studies in which the effect of brief visual stimuli on response output maintained by drug or food was assessed. Although these studies differ in a number of respects, such as session length and the frequency with which the reinforcers were available, in each of these studies, when brief stimuli (colored lights) were presented as a consequence of lever-pressing during the session, the rate at which lever presses were emitted increased by at least two-fold as compared to when no stimuli were presented. Importantly, this enhancement of behavior by an environmental stimulus occurred whether drug or food maintained the behavior, or whether the drug maintaining the behavior was cocaine, morphine or nicotine. Second-order schedules of nicotine injection have not been studied extensively in humans. However, in one human subject responding for intravenous injections of nicotine, response rates under a second-order schedule were almost three times higher than under a schedule in which second-order stimuli were not presented (16). TABLE I. EFFECTS OF BRIEF VISUAL STIMULI ON RESPONSE OUTPUT MAINTAINED BY DRUGS OR FOOD Reinforcer Dose or Session Presents- Responses/sec* amount per length tions per % presents- (hrs) session No Stimulus Increase tion stimulus Cocaine C1~) Cocaine C19) Morphine (18) Food (19) * Approximate values. T108350817

It should not be surprising, then, given the rather strong influence that relatively simple environmental stimuli have in ev~hancing nicotine-maintain- ed responding in animals (and humans), that environmental factors wuld play an important role in maintaining smoking behavior. A1thou~h the present review has focused on research ~rowlng out of the animal behavioral pharma- cological tradition, other models (20) also insist on the simultaneous attention to both pharmacological and psychological factors when studying determinants of smoking in humans. It is likely that in different indivi- d~Is the relative importance of these factors in controll~ng smoking behavior will vary. An understanding of the contribution of the various psychological and pharmacological influences in individual smokers will be important for suggesting appropriate treatment strategies for different indlviduals. Schachter S. Regulation, withdrawal and nicotine addiction. In: Kras- negor NA, ed. Cigarette smoking as a dependence process, NIDA Research Monograph 23. Rockville, MD., 1979: 123-133. (DHEW Publication No (ADM) 79-800). Lucchesi BR, Schuster CR, Emley GS. The role of nicotine as a determin- ant of cigarette smoking frequency in man with observations of certain cardiovascular effects associated with the tobacco alkalold. Clin Pharmacol Ther 1967; 8: 789-796. 3. Kumar R, Cooke EC, Lader MH, Russell ~A~. Is nicotine important in tobacco smoking? Clin Pharmacol Ther 1977; 21: 520-529. Henningfield JE, Miyasato K, Jasinski DR. Cigarette smokers self- sdminister intravenous nicotine. Pharmacol Biochem Behav 1983; 19: 887-890. 5. Kozlowski LT, Jarvik ME, Gritz ER. Nicotine regulation and cigarette smoking. Clin Pharmacol Ther 1975; 17: 93-97. o $~olerman IP, Goldfarb T, Fink R, Jarvik ME. Influencing cigarette smoking with nicotine antagonists. Psychopharmacolog~a 1973; 28: 247-259. 7. Schachter S, Kozlowski LT, Silverstein B. Effects of urinary pH on cigarette smoking. J Exp Pyschol (Gen) 1977; 106: 13-19. 8. Russell MAH. Tobacco smoking and nicotine dependence. In: Gibbins RJ et al, eds. Research advances in alcohol arm drug problems V.3. New York: Wiley and Sons, 1976: 1-47. 9. Schachter S. Nicotine regulation in heavy and light smokers. J Exp Psychol (Gen) 1977; 106: 5-12. I0. Sutton, SR, Feyerabend C, Cole PV, Russell M~. Adjustment of smokers to dilution of tobacco smoke by ventilated cigarette holders. C1in Pharmacol Ther 1978; 24: 395-405. TI08350818

135 II. I)eneau CA, Inoki R. Nicotine self-administration in ~nkeys. Ann l~Y Acad Sci 1967; 142: 277-279. 12. Yanagita T. An experimental framework for evaluation of dependence liability of various types of drugs in monkeys. Bull Narcotics 1972; 25: 57-64. 13. Yansgita T. Brief review on the use of self-admlnistratlon techniques for predicting drug abuse potential. In: Thompson T and Unna K, eds. Predicting dependence liability of stimulant and depressant drugs. Baltimore: University Park Press, 1977; 231-242. 14. Yanagita T, Ando K, Oinuma N, Ishida K. , Intravenous self-administratlon of nicotine and an attempt to produce smoking behavior in monkeys. In: Proceedings of the 36th Annual Scientific Meeting, Committee on Problems of Drug Dependence, National Academy of Sciences, 1974; 567-578. 15. Goldberg SR, Spealman RD, Risner ME, Henningfield JE. Control of behavior by intravenous nicotine injections in laboratory animals. Pharmacol Biochem Behav 1983; 19: 1011-1020. 16. Hennlngfield JE, Goldberg SR. Control of behavior by intravenous nicotine injections in human subjects. Pharmacol Biochem Behav 1983; 19: 1021-1026. 17. Goldberg SR, Spealman RD, Goldberg DM. Persistent behavior at high rates maintained by intravenous self-adminlstration of nicotine. Science 1981; 214: 573-575. 18. Goldberg SR, Spealman RD, Kelleher RT. Enhancement of drog-seeking behavior by environmental stimuli associated with cocaine and morphine injections. Neuropharmacology 1979; 18: 1015-1017. 19. Kelleher RT, Goldberg SR. Fixed-interval responding under second-order schedules of food presentation or cocaine injection. J Exp Anal Behav 1977; 28: 221-231. 20. Kozlowskl LT, Herman CP, The interaction of psychosocial and biologic- al determinants of tobacco use: more on the boundary model. J Applied Soc Psychol 1984; t4: 244-256. T108350819

137 Takeshi Hiraya~a, M.D. Epide~iology Division ~ational Cancer Center Research Institute Tokyo~ Japan INT~O~OC'fION Non-smoking wives of heavy smokers have a higher risk of lung cancer, as reported previously. Results of our follow-up study now reported here, not only confirm the results of the previous report (i), but also reveal additional evidence for the health consequences of passive smoking. MAT~LL~T.,S The 16-year follow-up results of a census-population based cohort study for 265,118 adults (m: 122,261, f: 142,857), aged 40 years and above, in Japan were analysed. RESULTS Non-smoking wives with a smoking husband were found to carry a significantly elevated risk of lung cancer (n=200), nasal sinus cancer (n=28), brain tumor (n=34) and ischemic heart disease (n=494) by a large scale cohort study, 1966-81, in Japan. In these diseases, the risk went up with the increase in number of cigarettes smoked by the husband. No other causes of death showed such significant association with husbands smoking, except suicide (n=200) where association of borderline significance was observed (Tables 1,2). A total of 429 deaths from lung cancer in women was recorded during 16 years follow-up (19'66-81). Out of these, 303 were non-smokers and 20,0 of these occurred among 91,540 non-smoking married women whose husbands' smoki~ habits were known. A similar trend of risk elevation in non-smoking women with the extent of the husband's smoking was observed in each age group, both by age of the husbands and by age of the wives, in each occupational group and in most areas ~der study (i~er~al consistency). ~o o¢her chara¢~e¢is¢ics of the husbands, such aa husband's drinking habits, were found to elevate the risk of lung cancer in their no~-smoki~g partners (specificity of association) (2). The results are in lime with a Creek study (3), and a Louisiana study Preventive Oncology, I~I Bldg., 1-4, Sadohara-cho, Ichigaya, Shinjuku-ku, Tokyo 162, Japan. I III T108350820

TABLE i. SPOUSE SMOKING AND CANCER Dose-response relationship Mantel- One-tail (No. of extension p value Deaths) chi Cancer of all sites (2705) 2.659 Ca. Mouth & Pharynx Ca. Esophagus Ca. Stomach Ca. Colon Ca. Rectum 0.00392* (22) -0.829 0.20355 (58) 0.246 0.40284 (854) -0.270 0.39358 (142) 0.463 0.32168 (112) -0.007 0.49721 Ca. Bile Duct & Gall Bladder (91) 0.972 Ca. Liver (226) 0.696 Ca. Pancreas (127) -0.860 Ca. Nasal Sinus (28) 1.963 Ca. Lung (200) 2.915 Ca. Breast Ca. Cervix Ca. Ovary Ca. Urinary Organs Ca. Skin 0.16553 0.24321 0.12500 0.02482* 0.00178* (115) 1.320 0.09342 (273) 1.156 0.12384 (54) 0.394 0.34679 (49) 0.125 0.45026 (23) 1.445 0.07423 Bone Tumor Brain Tumor Malignant Lymphoma Leukemia (17) 0.358 0.36017 (34) 2.673 0.00376* (85) 1.134 0.12840 (51) 1.389 0.08242 Husbands Smoking non ex- 1-14 15-19 20- Rate Ratio Ca. All Sites 1.00 in nonsm~king Ca. Lung 1.00 wives Ca. Nasal Sinus 1.00 Brain Tumor 1.00 1.16 1.13 1.36 1.42 - 1.57 - 3.05 1.04 I. 58 2.02 6.25 1.20 1.9I 2.55 4.32 T10,~350821

139 TABLE 2. SPOUSE SMOKING AND MAJOR CAUSES OF DEATH EXCLUDING CANCER Dose-response relationship Mantel- One-tail (No. of extension p value Deaths) chi All Causes (9106) 4.351 0.00001" Tuberculosis Diabetes Subarachnoid Hemorrhage Cerebrovascular Disease Ischemic Heart Disease Other Heart Disease Hypertensive Heart Disease Ulcer Cirrhosis Emphysema/Bronchitis Suicide (i00) 0.608 0.27159 (227) 0.800 0.21186 (126) 1.622 0.05240 (2609) 1.604 0.05436 (494) 1.979 0.02391* (680) 1.254 0.10492 (226) 0.927 0.17696 (57) 0.772 0.22006 (180) -0.808 0.20955 (106) 0.940 0.17361 (200) 1.859 0.03151(*) Husbands Smoking non ex- !-14 15-19 20- Rate Ratio in nonsmoking wives All Causes 1.00 1.26 1.16 1.06 1.19 Ischemic Heart Disease 1.00 1.03 1.17 1.06 1.30 Suicide 1.00 0.94 1.52 0.85 1.60 ........... !ill ....... [ ........ TIO83508PP

The risks of three other diseases, nasal sinus cancer, brain tumor, and ischemic heart disease in non-smokin~ ~omen were also found to go up signlficantly with the a~ount of the husband's smoking. The former observation appears important in demonstrating carcinogenic potentials of side-stream smoke mainly inhaled through the nose and the latter finding ~ust be of importance in planning control programs of another major killer of modern societies. I. Hirayama T. Non-smoking wives of heavy smokers have a higher risk of lung cancer: a study from Japan. Br Med J 1981; 282: 183-185. 2. Hirayama T. Passive smoking and lung cancer; association. Lancet 1983 Dec 17; 11(8364): 1425-1426. 3. Trlchopoulos D, Kalandidi A, Sparros L, MacMahon B. passive smoking. Int J Cancer 1981; 27(I): 1-4. consistency of Lung cancer and 4. Correa P, Pickle LW, Fontham E, Lin Y, Haenszel W. Passive smoking and lung cancer. Lancet 1983 Sept I0; 11(8350): 595-597. Garfinkel L. Time trends in lung cancer ~ortality among non-smokers and a note on passive smoking. J Nat Cancer Instit 1981; 66: 1061-1066. Preston-Martin S, Yu MC, Benton B, Henderson BE. and childhood brain tumors; a case-control study. 42:5240-5245. N-nitroso compounds Cancer Res 1982; Brunnemann KD, Adams JD, Ho DPS, et al. The influence of tobacco smoke on indoor atmospheres. II. Volatile and tobacco specific nitrosamines in main- and sidestream smoke and their contribution to indoor pollution. In: Proceedings of the 4th Joint Conference on the Sensing of Environmental Polutants. New Orleans 1977. Washington: American Chemical Society, ]978: 876-880. Brunnemann KD, Hoffman D. Chemical studies on tobacco smoke LTX. Analysis of volatile nitrosamines in tobacco smoke and pollu~ed indoor envlronmen~s. In: Walter EA, Griciute L, ~aste~naro M, eds. Environmental aspects of N-nitroso compounds. Lyon: World ~ealth Organization, ~978: 343-356. (IARC scientific publ~catlons No. 19). Nhi~e RJ, Froeb FH, $~!!-ai~sys dysfunction in ~sm~kers chronically exposed to tobacco smoke. New Eng J Med 1980; 302: 720-723, T108350823

(4) (external consistency) but are somewhat at variance with an Americsn C~ncer Society study in the D.$oA. (5). Differences in room size, in proximity between husband and wi£e, in room ventilation and in frequency of wives' office work would be the major reasons. ILeal si~s cancer a~d brain t~or A significant risk elevation for cancer of the pars-nasal sinuses and brain tumor in ~on-smoking wives was observed according to the ~mount of the husbands' smoking. No ocher risk factors studied significantly altered the risk of nasal sinus cancer in women. This finding must strengthen the plauslbillty of carcinogenic hazards of side-scream smoke inhalation through the nose. For brain tumor, a significn~ risk elevation by passive smoking was reported for childhood brain tumor (6). It is of importance that similar risk elevation of adult brain tumor by passive smoking was observed in the current study. Ische~c heart disease A significant risk elevation with increase in the extent of the husband's smoking was observed with ischemic heart disease. Similar results were obtained in the detailed analysis by age and occupation. DISO~SSION The current results of elevated risk of nasal sinus cancer, brain tumor and ischemic heart disease, in addition to the risk of lung cancer, are in llne with the results of measurements of various carcinogens and other toxic substances found to be mostly in higher concentration in side-stream smoke than in main-stream smoke (7,8). The results are also compatible with known evidence showing the possible influence of passive smoking on health, including elevation of CO-Hb levels and nicotine/cotinlne levels in blood and/or urine after exposure to passive smoking and demonstration of small airway dysfunction in those exposed daily to passive smoking in the work place (9). The results of the present study must be utilized effectively in planning control programs for lu~g cancer and other selected diseases. The results clearly indicate that lung cancer, especially in women, can only be controlled when proper measures are taken against passive smoking as well ~s a~a~ns~ active smoking. A similar statement must be valid also with ischemlc heart disease. In the case of nasal sinus cancer and brain tumor, since the influence of active smokin~ on the risk is known to be limited, the role of controllin~ passive smoking must be of particular importance. The previous report that non-smoki~ wives of heavy smokers have a higher risk of lung cancer was confirmed b~ further - emen was ou~ to be valid after considering most of the possible confounding variables, internal and external consistencies and specificity of association. T108350824

Leona Hubac~ov[, M.D., Ph.D. Miloslav Huba~, M.D., Ph.D. Franti~ek Strelka, Ph.D. i~ich ~orsk~, M.D., Ph.D. Research Institute of Preventive Medicine Bratislava, Czechoslovakia Many authors consider in their papers the relationship between smoking and occurrence of some diseases. Primarily, they have analysed causes of deaths in people who were heavy smokers for an extended period (1,2,3). The aim of our study was to find whether smoking has an influence on the health state of employed women in Slovakia. I~ETHODS In 13 industrial plants and two hospitals, 4,000 randomly selected women aged 16 to 67 years were examined for their health status, and also their history of previous diseases was ascertained. Additionally, data about their smoking habits were recorded, i.e.~ the years of smoking and the number of cigarettes smoked in a day. We examined three groups of women: 3,155 factory workers, 424 clerical workers and 421 medical nurses. From these groups of women, 832 were currently smokers (mean age 30.0 ± 9.0 years), 194 women were ex-smokers (mean age 29.9 ± 9.4 years), and 2,974 women were non-smokers (mean age 38.3 ± 11.3 years). Twenty-two percent of the examined factory workers were smokers, 33% of the clerical employees and 42% of the nurses. In the group of current smokers and ex-smokers (1,026 women), 584 women were ~ to 29 years of age, 343 were in the age group 30-45 years and 99 women were older than 45 years. In the group of current smokers, 465 women were younger than 29 years, 286 women were in the age group 30-45 years and 81 women were older than 45 years. Correspondence and requests for reprints should be addressed to: Leona Huba@ovI~ M.D., Ph.D., Research Institute of Preventive Medicine, Li~ovE 1~ 833 0~ Bratislava. C~eBn~.o,,.u~. TI08-350825

In data processing ve evaluated 16 parameters - including factors relating to the patient's general medical history and to the diseases which were assumed So be the harmful consequences of smoking. We evaluated the follow- i~ indices: [I] Subjective difficulties of the examined person, [2] sleep disturbances, [3] frequent headaches, [4] dizziness and fainting symptoms, [5] neurotic symptoms, |6] chronic inflammation of the upper respiratory cr=c~, [7] chronic laryngitis, [8] chronic bronchitis, [9] hypertonlc disease, {10] myocardial infarction, acute and chronic heart disease and angina pectoris= [11] peptic ulcer, [12] gastritis, duodenitis and chronic intestinal inflam-- marion, [13] complications £n pregnancy, [14] spontaneous abortion, [15] premature delivery, [16] lung tuberculosis. Health status of smokers and ex-smokers was compared with non-smokers. We calculated the percentage ~, to indicate the degree to which smoking may have been associated with health impairment in smokers and ex-smokers: P - P0 ~ - p . 100% , where P is the percentage of ill persons in the entire sample P0 is the percentage of ill non-smokers in the reference sample. RgSDLTS An increased occurrence of health impairments and subjective difficulties with increased age of examined women was found. FIGURE 1. SUBJECTIVE DIFFICULTIES AND MORBIDITY OF NON-SMOKERS IN COMPARISON TO SMOKERS AND EX-SMOKERS IN THE AGE GROUPS UNDER 29 AND 30-45 YEARS. T108350826

Figure I illuatrates subjective difficulties and ~rbldlty of uo~-s~okers in co~ris~ to s~kers a~ ex-s~kers ~n the age groups under 29 ~ars and ~5 years. In the yo~ges~ age grip, ~ ascertained difference~ ~t~een ~ w~h s~k~ng habits ~ n~-s~kers ~n 5 indices. Si~igicsncly hi~- er occurrence (P ~nd ~a~nt~ ay~Coms). In the a~e ~ of 30--~ years, we f~ differences in 6 indices. Si~i- ficantly higher occurrence was fouo~ in index 6: chronic infla~iou of the upper respiraco~ ~racc (P < 0.001) and in index n~ber 8: chronic bron- chitis (P < 0.05). Figure 2 illustrates ~he subjective difficulties and ~orbidity of non- s~okers in comparison to s~okers and ex-smokers in those over 45 years of age. Certain effects of smoking habits possibly were ~anifested in 12 indices. Significant differences were found only in index 1: subjective difficulties of the examined person (P < 0.05). FIGURE 2. SUBJECTIVE DIFFICULTIES AND MORBIDITY OF NON-SMOKERS IN COMPARISON TO SMOKERS AND EX-SMOKERS IN THE AGE GROUP OVER 45 YEARS. 'mx TIO&350827

FIGURE 3. PERCENTAGE p IRDICATI}IG THE DEG~.F. TO ~ICH THE SMDKIFIG HABIT MAY HAVE CONTRIBUTED TO THE HEALTH IMPAIRMENT IN $~0KERS AND EX-SHOKERS IN THE AGE GROUPS UNDER 29, 30-45 A~D OVEE 45 YEARS. OVER 7 6 9 fl 12 IN~X SMOKERS ~O EX-~E~ [] EX-g, IOKERS Figure 3 shows the degree to which smoking may have contributed to at least 30% of the health impairment of smokers and ex-smokers in three age catego- ries: up to 29 years, 30-45 years and over 45 years. In the first category we found p = 56% for gastritis, duodenltis and chronic intestinal inflanmm- tion. In the age category of 30-45 years old women, we found similar results in chronic inflammatlon or the upper respiratory tract (smokers and ex-smokers together: ~ = 46%, ex-smokers: ~ = 43%), in chronic laryngitis (ex-smokers: ~ = 49%) and in premature delivery (smokers and ex-smokers: ~ " 38%). In the last age category, over 45 years, we found ~ higher than 30% in 5 indices: chronic inflammation of the upper respiratory tract (smokers and ex-smokers: ~- 46%), peptic ulcer (smokers and ex-smokers: ~ = 52%), gas- tritis, ~uod~nitis and chronic intestinal infla~tion (smokers and ex- smokers: ~ = 93%) and complications in pregnancy (ex-smokers: DISCUSSION Many authors have been interested in the relationship between smoking and morbidity and mortality in different groups of the population (2,3,4). These papers are based first of all on statistical data of causes of deaths or autopsy records from hospitals. T108350828

147 In our paper ~e tried to find out the harmful consequences of smoking on liviv-g subjects. It is known that women do not smoke to such an extent as men. They start smoking later, smoke less than men and therefore have a lower incidence of the so-called smoking-related diseases, primarily lung cancer (5). It is known that the female organism is more susceptible to the effects of harmful substances cha~ the male and for this reason m~ny co,retries have enacted specific legislation ~o protect the health interests of ~n a~ work (6), In spite of these legislative provisions there are, nonetheless, many poten- tially harmful substances in the working environment. It is also necessary to consider the slt~atlons outside the working environment - at home, in the garden, or resulting from air pollution in the majority of industrial regions. In women who smoke, the harmful effects of their smoking are superimposed on these general environmental factors. Attention should be paid to all influencing factors when evaluating compli- cations in pregnancies, spontaneous abortions and premature deliverles. Physical strain of women should also be considered. We should realize that physical work is performed not only by working women in industrial plants, but by nurses in medical wards too. We should also have in mind the high neuropaychical strain on women employed in some professional groups. Such a strain may have consequences. Smoking may be the compensatory action of people in response to increased neuropsychical strain, such as clerical employees and nurses. In this way, we may explain the higher percentage of smokers in these professional groups. In conclusion, it is possible to say that among smoking female employees of industrial plants and hospitals in Slovakia, we found increased incidence of subjectively expressed problems and some diseases. It is possible to sup- pose that they are consequences of long-term smoking which, together with a number of other factors, may initiate pathological changes in human organisms. TI08350829

o Cherry WH, Forbes W~. Canadian studies aimed toward a less harmful cigarette. J list Can,cer Instit 1972; 48: 1765-1773. Forbes WF, Gentleman JF. A possible similar pathway between smoking-induced shortening and natural aging. J Gerontol 1973; 28: ~02-311. Hammond EC. Quantitative relationship between cigarette smoking and death rates. Hatl Cancer Inst Monogr 1968; 28: 3-B. Gentleman JF, Forbes, WF. Cancer mortality for males and females and its relationship to cigarette smoking. J Gerontol 1974; 29: 518-533. Fingerland A, Hus~k T, Bendlov~ J. Vliv kou~en[ cigaret na morbiditu a mort~litu. I: Sborni~k v~deck~ch prac[ l~ka~sk~ f~kulty D-K. Hradec Kr~lov~ 1971; 14: 221-231. World Health Organization. February 16-18. Report on a W~O meeting. Budapest, 1982 T108350830

IN I'WO SOCIO~C (Z~S~S I~ l)~, II~)I~ Kasturi Jayant, M.Sc. Perin Notani, M.A. L.D. SanEhvi, Ph.D. Cancer Research Institute Tara Memorial Centre Fatal, Bombay 400 012 India INTRODUCTION In developed countries, several studies have implicated smoking as a risk factor for cancers and coronary heart diseases. In particular, the risk of lung cancer is shown to be high for cigarette smokers. In India, varied forms of smoking besides cigarettes are prevalent. Bidi smoking is a common form. It has been shown that the constituents of bldi smoke contain higher levels of tar (23-41 mg), nicotine (1.7-2.8 mg), benzo(a)pyrene (78 ng) and carbon monoxide (7.7 vol%) compared to western cigarettes (1,2). Even ciga- rettes marketed in the country have high levels of tar (18-28 mg) and nlco- tine (1.0-1.8 mg). The filter cigarettes do not necessarily have lower levels of tar and nicotine than nonfilter cigarettes (I). Further, the machine estimates have been estimated with the international standard of puff frequency of I puff per minute for cigarettes and 2 puffs per minute for bidis. A study of smoking behaviour of ~he Bombay smoker has shown that he smokes cigarettes with 2 puffs per minute and bidis with 5 puffs per minute (i). Thus, his smoking behaviour puts him at a higher level of exposure than indicated by machine estimates. In view of these observations, study of health hazards due to smoking in the Indian population assumes great relevance. So far, only retrospective studies have been reported from ~he country, demonstrating higher risks of cancer of upper alimentary tract in bidi smokers (3) and of lung cancer (4,5) and coronary heart disease (6) in bidi an4 cigarette smokers. A pros- pective study to assess risk of t~rtali~y and morbidity d~e to coronary heart disease and cancer in bidi and cigarette smokers has been undertaken in males of two socioeconomic classes in Bombay. The preliminary results of this study, which is still being continued, are presented here. For such a cohort stray, it was essential to ascertain whether the usual method of collecting data by mailing questionnaires, followed by several workers in developed countries, was feasible in the Indian population. When this approach was attempted with a group of lawyers and doctors, the response was very poor - just about 33% inspire of all efforts. As a T108350831

150 methodology ~hich ~ould give good response at enrollment, low rate of loss at follow-up and reliable data o~ ~ortality and ~orbidity in the study pop- ulation. M~~Y FO~ D~TA ~OLLECTI~ It was [elf that if ~e could enroll for our study a~a~b~rs from organiza- tions with a stable work force, by meeting them personally, the response would be good. A pilot study was conducted in an organization with white- collar workers. The social worker approached all the members who were 40 years and above and requested them to enter the study and fill up at their leisure a questionnaire on personal data, including tobacco usage. This gave the social worker an opportunity to clear any misgivings the members might have had regarding entering the study. The following day the social worker went back to collect the completed questionnaire. At times more than one visit was required to collect the questionnaires. This approach gave encouraging results and was followed for white-collar workers in all the organizations entering the study. Altogether 6742 junior and mlddle level executives, supervisory and clerical staff belonging to 19 organizations were enrolled and will be designated as social class W. The response in this class, varied from 51% to 98% in the various organizations, giving an overall rate of 76%. However, for the blue-collar workers personal interviews had to be conducted as, even though the workers were literate, it would have been difficult for them to fill out the questionnaire. In this way, 5,981 persons who were mill-workers, ~aechanics, bus drivers, or conductors were enrolled and this group will be designated as social class B. Unfortunately, we could not assess the response at enrollment for this class, as the organizations sent us persons in batches for interview without giving us the total list of persons. Only those above 35 years were interviewed. There were hardly any refusals. As the study population belonged to organized groups, follow-up did not pose any serious problem. The follow-up procedure entailed perusal of Attendance Registers every six months to see if any member was on long sick leave, had resigned or retired, so that they could be interviewed at their homes to get the necessary health information. In the case of any change o.f address, the new address was obtained from Whe organization or neighbours. If the changed addresses were local, home visits were undertaken. If not, a !citer was sent enquiring about their health. The response to these letters was surprisingly good, perhaps because of the initial personal contact and the rapport established. The period of follow-up for Class W was three to five years and for Class ~ one ~o five years. During the home visits, if any member was found to have any severe illness all the necessary information regarding diagnosis, attending doctor, hospi- talization if any, were recorded from the members' case papers. For cancer the study period, brief health foram regarding chronic ailments ~ere filled out for each individual in the study group by soclal workers through person- al interview to ~ake doubly sure that no cancer or coronary heart disease case was missed. If any ~ea~ber had expired between follow-up visits, the T108350~32

death registratio, nu~er as well as infor.mtlon on the cause of death were collecte~ either from the organization or relatives a~ crosschecked with the Bo~ay Municipal Corporation death records. It was of importance to collect the mortality and morbidity data ourselves as. the information was not readily available from any systematic records. ~h~ also. Even though members below 40 years and above 58 years were enrolled, the present analysis is restricted to the cohort 40-58 years. Further, the attained age at the end of the follow-up period included in the analysis was 41-59 years. Over the duration of the study, 2% person years in class W and 10% in class B were los= to follow-up. The population enrolled were grouped into 4 categories on the basis of tobacco usage. One group comprised those with single usage viz. current bidi smokers or current cigarette smokers or current chewers. The second group comprised those with dual usage viz. current smokers and chewers, and the third group comprised ex-smokers, ex-chewers, occasional smokers and occasional chewers. The last category waz e~mprised ef tbo~e who neither smoked nor chewed, referred to henceforward as the non-exposed group. For the study of risk in smokers we have considered only those with single current usage as this forms a homogenous group. Table 1 shows, for the two classes, the number of persons enrolled in each of the habit groups along with the person years accumulated. TABLE I. NIPMBER ENROLLED Ah'D ACCUMULATED PERSON YEA~RS BY TOBACCO USAGE IN THE TWO SOCIOECONOMIC CLASSES. (Cohort 40-58 Years) (Attained age 41-59 Years) Tobacco Usage: Bidi S~oker Cigarette Smoker C~ewers and Mixed Habit Total data Socioeconomic Class No. Enrolled Person Years B W B W 515 82 1512.5 317.5 344 692 1128.5 3046.0 2327 2922 6947.5 12735.5 3922 6157 11799.5 26745.0 T108350833

152 The health consequences considered ~ere mortality fro~ all causes, a~d incidences of coronary heart disease (CKD) (including deaths), myocardial infarction (CHDI) (a subset of CHD), cancers of the upper alimentary and respiratory tract (CA1) and all other cancers (CA2). Mortality and morbidi- ty in each exposed group for the two classes are shown in Table 2. TABLE 2. PERSON YEKRS, h~JMBKR OF DEATHS, INCIDENC~ OF CORONARY H~ART DISEASE ~ CANCER BY CLASS Person Deaths CHD CHDI CA1 CA2 Years Class B Nonexposed 2211.0 12 8 5 I 1 Bidl Smokers 1512.5 19 18 i0 4 0 Cigarette Smokers 1128.5 13 12 7 3 0 Total 11799.5 75 64 39 ii 4 Class W Nonexposed I0646.5 57 84 29 3 7 Bidi Smokers 317.5 7 4 1 0 1 Cigarette Smokers 3046.0 11 26 13 I 0 Total 26745.0 120 187 65 8 9 Rate ratios were estimated over age strata by using Miettinen's method given by Rothman and Boice (7). Five yearly age strata were considered for estimation of rate ratios for mortality and coronary heart disease and ten yearly for cancers. RESULTS The risk for smokers within each of the classes, along with upper and lower bounds are shown in Table 3. It was found that the bidi smokers in class B have significsntly higher risk, ~wo ~o three fold, for mortality, CKD and CN_DI, and a five fold risk for cancers of the upper alimentary and respira- tory tract compared to the nonexposed of the sa~e class. The cigarette s~okers of this class also have a significantly higher risk, three fold, for CHD, but the risk for mortality and CHDI, though more than two fold, falls short of significance (0.05<P<0.10). The risk for CA1 is 6.19 (O.05<P<0.10). In class W, bidi smokers were too few for a detailed study. The cigarette smokers did not show significantly higher risk for either mortality or any of the morbidity conditions studied. However, when only heavy smokers (~ 20 per day) were considered, significantly higher risk for CHD (2.03, P<O.05) and CHDI (2.90~ P<O.OS) were observed. The risk of mortality was not higher and risk of cancers could not be estlm~ted due to paucity of numbers. T10.~350834

ESTIF~kTED RATE RATIO FOR MORTALITY, II~CID~_~CE OF COitOI~A~Y HEART DISEASE AI~D CANCER; WITHIN CLASS COMPARISONS: SMOKERS VS NONEXPOSED Class MORTALITY CHD CHDI CAI CA2 Bidi Smokers 2.22* 3.27* 3.04* 5.22+ (1.10, 4.46) (1.49, 7.16) (1.07, 8.65) (0.76, 35.gl) Cigarette Smokers 2.09+ 3.O1" 2.76+ (0.97, 4.50) (1.28, 7.09) (0.92, 8.33) 6.19+ - (0.84, 45.70) Class W Bidi Smokers Cigarette Smokers 0.69 (0.37, 1,29) Sample Size Small 1.13 1.63 1.16 (0.73, 1.75) (0.86, 3.09) (0.13, 10.35) 95% lower and upper bounds are given in parentheses * P<O.05 + P<0.10 TABLE 4. ESTIMATED RATE RATIO FOR MORTALITY, INCIDENCE OF CORONARY HEART DISEASE AND CA~NCER; BETWEEN CLASS COMPARISON MORTALITY CHD CHDI CA1 CA2 Class ~.vs. !.51" Class W (I.14,2.01) Nonexposed in Class B vs Nonexposed in 1.06 Class W (0.58,1.95) Cigarette Smokers in Class B vs Cigarette Smokers in Class W 3.50* (1.63,7.51) 0.82 1.43+ 3.32* 1.03 (0.62,1.t0) (0.96,2.14) (1.39,7.92) (0.32,3.35) 0.47 0.84 1.58 (0.23,0.97) (0.32,2.21) (0.18,13.60) 1.31 1.53 8.95* - (0.65,2.64) (0.58,3.98) (1.33,60.18) - 95% lower a~i upper bounds are given in parentheses * P<O.05 + P<O.IO T108350835

154 Estimated rate ratios for mortality from all causes, incidence of coronary heart disease and cancer for specific comparisons between the tw~ classes are given in Table 4. Nhen the entire class B is compared to the entire class W, mortality arid cancers of the upper alimentary ar~ respiratory tract are found to be significantly higher in class B. Idyocardial infarction just falls short of siEnificnce (0.05<P<0.10) and there is no difference £n inci- dence of all CflD between the two classes. When comparison is restricted to the ~o~exposed groups of the ~ classes, no significant difference is observed for mortality, CRD, CHDI and CAl. However, cigarette smokers of class ~ have a three fold risk of mortality and a nine fold risk of cancers of upper alimentary and respiratory tract as compared to cigarette smokers of class W, but the risk of CfIDl is 1.5, and it does not attain signifi- cance. Bidi smokers in the two classes could not be compared as the sample size in class W was too sma11. DISCUSSION Significantly higher risk of mortality in bidi smokers compared to the non- exposed group is observed for class B indicating bidi smoking to be a risk factor in early deaths. In a rural Indian population, Gupta et al (8) reported mortality in male bidi smokers to be higher than in male chewers. The three fold risk for ali CHD and myocardial infarction in class B is in agreement with the results of a case control study in a Bombay Hospital, which serves the low socioeconomic class, where the observed risk for bidi smokers was 3.1 for all CHD and 3.8 for myocardial infarction (6). Cigarette smokers in class B have a two fold risk of mortality compared to the nonexposed group though it just falls short of significance. However, in class W the mortality in cigarette smokers is not significant. Mortality ratios of 1.83 and 1.63 for cigarette smokers have been reported by Hammond and Doll respectively and, in the Japanese data, Hirayarna reports a risk as low as 1.25 (cited in U.S. Surgeon General's report (9)). The observed risk of 3.01 for all CHD and 2.76 for myocardial infarction in c~garette smokers in class B compares well with the results of the case con- trol study mentioned earlier. However~ class W does not show significantly high risk of CHD or CHDI even though the values are greater than unity. In western studies, though workers have reported high risk for ali CHD and ~I, there are also studies which do not report increase in risk. Keys reports a risk of unity for those smoking less than 20 per day and Dayton reports risk of 1.00 for those smoking less than 10 cigarettes per day and 1.17 for those smoking more than 20 per day (cited in H.S. Surgeon General's report (9)). As seen earller, in our study risk of Cl~ and CIIDI in heavy cigarette smokers (e 20 per day) is enhanced and attains significance. Cancer risk in smokers varies by specific site in the upper alimentary and respiratory tract. However, for risk analysis of our data on cancer, we had cancers. !t is seen that the overall risk of these cancers is high, five to six fold, in both bldi and cigarette smokers in class B though it just falls short of attaining significance. In class W, larger numbers are required to arrive at a definitive conclusion. TI08350836

differential ~ortality and ~orbldity due to ~yocardial infarction and in the t~o classes m~y possibly be due to differences in life style. In persuance of this hypothesis we have studied one of the components of life style - viz., smoking - in some detail. The proportion who ever smoked or chewed was high in class B as compared to class W (81% vs. 60%) as also the proportion of bidi smokers-nonchewers (13% v~. 1.5%) a~ bi~i ~rs irresp~ctiv~ of chewing (24% vs. ~.1%). ~ there was no differential in the frequency distributions of bldi or ciga- rette s~king in smokers of the two classes, the proportion of cigarette s~kers who started s~ki~ by age 20 was h~gher in class B compared to class W. These tobacco usage profiles point to the fact that class B is a hlgher risk group for Zobacco related diseases. We have already seen in the between-class comparison Ehat this is, in fact, so - class B was observed to have a higher risk of ~rtallty, CHDI and CAl. Interestingly, the non- exposed groups of the two classes did not differ in mortality or incidences of coronary heart disease or cancers (upper alimentary and respiratory tract). The higher risk in cigarette smokers of class B for mortality and CA1 compared to class W needs further study. CONCLI/SION No doubt further data need to be accumulated to reach definitive conclusions regarding health hazards in some of the exposed groups. But the present study has shown that bidi smokers have a high risk of mortality and morbidi- ty due to coronary heart disease and cancer of the upper alimentary and respiratory tract. The study also brings out clearly that class B is a higher risk group com- pared to class W for coronary heart disease as well as cancers of the upper alimentary and respiratory tract, which are mainly tobacco related. Class B should be a target group for smoking control and cancer prevention programmes. We are indebted to the authorities of the various organizations who have made this study possible. Thanks are due to social workers Mrs. gulati, ~iss P.R. Shah, Mrs. V.V. Gadre and Mrs. P. Lentin for their pains- taking work. Pakhale SS, Jayant K, Sanghvi LD. Chemical constituents of tobacco smoke in relation to habits prevalent in India. Indian J Chest Dis Allied Hoffmann D, ganghvi LD, Wynder EL. Comparative chemical analysis of Indian bidi and cigarette smoke. Int J Cancer 1974; 14: 49-53. TI08350837

Jussawalla DJ, Deshpande VA. Evaluation of cancer risk in tobacco chewers and smokers - an epidemlologic assessment. Cancer 1971; 28: 244-252. Notani PN, Sanghvi LD. A retrospective study of lung cancer in Bombay. Br J Cancer 1974; 29: 477-482. Jussawalla DJ, Jain DK. Lung cancer ia Greater Bombay: correlation with religion and smoking habits. Br J Cancer 1979; 40: ~37-448. Jayan~ K, Gulati S, Sanghvi LD. Tobacco usage in relation to coronary heart disease: a case control study in Bombay, India. World Smoking and Health 1983; 8: 15-18. Rothman KJ, Bolce Jr. JD. Epidemlologic analysis with programmable calculator. Washington, DC; 1979. NIH Publication No. 79-1649o Gupta PC, Mehta FS, Iranl RR. Comparison of mortality rates from bidi smokers and tobacco chewers. Indian J Cancer 1980; 17: 149-152. US Dept of Health, Education and Welfare. Smoking and Health. A report of the Surgeon General. U.S. Department of Health, Education and Welfare, Public Health Service 1979. (DHEW Publication No. (PHS) 79-50066). TI08350838

[57 Li Nan-Xlan, M.D. Director, Department of Epidemiology School of Public Health Shanghai First Medical Colle~e Shanghai, People's Republic of China The harmful effects of cigarette smoking on health have been identified for decades in the western countries and various measures have been taken ever since to diminish its risk (1,2). In China, cigarette smoking has become popular since the second world war~ especially after the foundation of the People's Republic of China. But very few papers have been published regard- ing cigarette smoking (3,4), and there have even been some saying that ciga- rette smoking has little relationship to health, including lung cancer, in China (5). This paper demonstrates the evidence of association of cigarette smoking to mortality in a defined population in Shanghai, China. SOgR~K OF DATA ~ND METBODS USED IN ANALYSIS The defined population consisted of 2,923 persons 55 years of age and over, 1,636 males and 1,287 females, from three factories enrolled in the mass survey of coronary heart disease in 1972. The details of that survey have been reported elsewhere (6). Everyone was interviewed and the smoking status, including whether the subject smoked, the average number of ciga- rettes smoked daily and the years he or she had smoked, were recorded at the same time. The subject was defined as a smoker if he or she smoked more than one cigarette daily for more than one year. Those who smoked occasion- ally and those who had stopped for more than five years were considered as non-smokers. To check the reliability of the smoking history already taken, ~94 workers, 75% of one factory, were reinterv~ewed, in the absence of the previous record, by students from the School of P~blic Health, Shanghai Pirst Medical ColleKe in Fall 1982, and the da=a recorded. Using December 31, 1972 as the starting point, mortality data of the 2,923 persons were collected each year up to the end of 1981 with the help of the staff of the medical offices, finance sections and labour unions of the respective factories, and checked with other related medical colleagues to co=flrm the causes of deaths. Durin~ this period, 116 subjects, 81 male and 35 female, moved [o other places and were lost. They were removed from the llst at the years they moved. The number of subjects observed at the end of each year was the number T108350839

158 LI living at the end of the previous year minus the number dying and lost during the year. Person-years of exposure ~ere calculated by adding the number of people observed at the ends of the previous year and the successive year, divided by two. Total person-years of exposure was the sub of the person-years of exposure of each year. The ~u~ber of subjects observed and person-years of exposure ~ere all separated by sex, age group and s~oking status. Direct standardization was used for age adjustment~ using the overall subjects 8s standard population, In comparing the mortalities between different sexes or different smoking status of the same sex, age adjusted mortalities (P'A and P'B) were calculated by using the sum of the two different populations as the standard population. The sum of variance of difference in mortality of each age group (S2dl), and the total number of subjects (N~) of each age group were used in deriving the standard error of the difference in the standardized mortality (S~) by the formula: / ZN:i 2 = 2 (zNi) U test was used to measure the significance of the difference in standard- ized mortalities. U P'A - P' = B , when U > 1.96, P < 0.05; U < 2.58, P < o.oi (7). The percentage of smokers in each sex was calculated by direct standardiza- tion using the total number of subjects as standard population. Relative and attributable risks were also measured. RESULTS i. The Smoking Histories Smoking histories taken i0 years apart were compared, with 90.5% coincidence as shown ~n Table I. TABLE I. COMPARISON OF NO. OF CIGARETTES SMOKED DAILY 1972 Record 1982 Record 0 I0 10-20 20 Total 0 456 IO 5 2 473 <i0 18 69 i0 0 97 10-20 4 21 243 1 269 >20 I 0 13 41 55 Total 479 i00 27I 44 894 809 Coincident rate = BR:)T x I00 = 90.5~ T108350840

2. }~ortalicies There were altogether 161 deaths, 123 male and 38 female, during the nine years follow-up. The number of person-years of exposure a~d deaths are shown in Table 2, TABLE 2. NUMBER OF PERSON-YEARS OF EXPOSURE AND DEATHS Person-years of Exposure Age Male Female Group Smokers Nou-smokers Total Smokers Non-smokers Total <45 1315,5 1560 2875.5 90.5 3360.5 3451 45-54 3161 2118 5279 404.5 4983.5 5388 55-64 3056.5 1247 4303.5 283,5 i539.5 1823 65- 878 508 1386 122 464.5 586.5 Total Age Group <45 45-54 55-64 8411 5433 13844 900.5 10348 11248.5 Number of Deaths Male Female Smokers Non-smokers Total Smokers Non-smokers Total Grand Total 6326.5 10667 6126.5 1972.5 25092.5 2 1 3 1 1 2 5 14 5 19 2 9 11 30 39 10 49 4 8 12 61 38 14 52 3 I0 13 65 Total 93 30 123 I0 28 38 161 Age-adjusted mortality per 1,000 Male: Smokers 8.78 Non-smokers 5.29 Total 7.52 Female: Smokers 10.27 Non-smokers3.80 Total 4.36 U and P values of the difference between different mortalities: U P Male amd female smokers 0.32 >0.05 Male and femnle non-smokers Hale total and female total Male smokers and male non-smokers Female smokers and female non-smokers 1.2~ >0.05 2.81 <0.01 2.39 <0.05 3.58 <0.01 TIO8350,M- 1

160 LI Males had an overall higher ~ortality than females, the difference bei~ significant. But the differences of ~rtalitles between males and females became non-significant when smokers were compared with smokers and non-smokers were compared with non-smokers. Smokers showed a much higher mortality than non-smokers of either sex. Excess mortality of male smokers was 40% while that of feumles was 63%. Males had a much higher age adjusted percentage of smokers (59.5%) as compared with females (9.8%), the difference being significant. TABLE 3. PERCENTAGE OF SMOKERS IN EACH SEX Age Number of Subjects Percentage of Smokers Group Male Fera~le Male Female <45 607 736 50.4 3.67 45-54 598 385 65,4 13.25 55-64 369 136 73.7 16.18 65- 62 30 52.3 26.67 4. Relative Importance of Smoking to Different Causes of DeaLh Causes of deaths may be roughly grouped into three major categories: cardiovascular d~seases, malignanc[es and other diseases. The number of deaths in each category is shown in Table 4. TABLE &. CAUSES OF DEATH AMONG SMOKERS AND NON-SMOKERS OF EACH SEX (1973-1981) Male Cardiovascular Age Malignancy . Disease Other Causes Group Smokers Non-smokers Smokers Non-smokers Smokers Non-smokers <45 0 0 0 0 2 1 45-54 l I 3 2 0 i 2 55-64 24 ! ii 5 4 4 65- 15 6 15 4 8 4 Total 50 10 28 9 15 II T108350842

161 TABLE 4, continued: Female Cardiovascular Age l~ali~nancy Disease Other Causes Group Smokers Non-smokers Smokers Hon-smokers Smokers Non-smokers <~5 0 0 0 I 1 0 45-54 0 3 2 4 0 2 55-64 I 5 2 2 1 I 65- I 3 I 5 I 2 Total 2 ii 5 12 3 5 The difference of standardized mortalities between smokers and non-smokers, either male or female, the risks and percentage of excess mortalities are shown in Table 5. TABLE 5. RELATIVE IMPORTANCE OF SMOKING TO DIFFERENT CAUSES OF DEATH Male Female CVD Tumors Others CVD Tumors Others Difference of Age Adjusted U i.II 2.97 0.68 Mortalities Between Smokers and Non-smokers P Relative Risk 2.19 0.16 4.32 >0.05 <0.01 >0.05 <0.05 >0.05 <0.01 1.51 2.78 0.77 2.71 0.87 8.44 Attributable Risk per !,000 1.03 3.50 2.69 3.86 Excess Mortality (%) 33.7 64 68.2 In male smokers~ mortality due to maligaant tumors was grea~er than ghst of non-smokers, wi~h the difference being significant and with excess mortality of 64%, while in females, in contrast to males, the differences between satokers and non-smokers were significant in cardiovascular diseases and other diseases, with excess mortality of 68.2X and 88Z respectively, but not in malignant tumors. 5. gelative lmportaace of Smoking to Individtml Diseases Nalignancy ranked the first in frequency in cause of death in males. cancer sites and status of smoking of the victims are shown in Table 6. The T108350843

162 TABLE 6. C~I~ER SITES AMD STATUS OF SMOKIMG IN MALE DEATHS Cancer Sites Smoker ~on-smoker Lung 15 Esophagus 4 Stomach i0 Liver and Biliary Tract i0 Colon 4 Bone 1 Nasopharyn~ 2 Mediastlnum I Retroperitoneum Pancreas l Lymphoid Tissue 1 Leukemia Teeth Lung, stomach and liver cancers were the three most frequently encountered cancers. The differences of the age adjusted mortalities of smoker to non- smoker for individual cancers were all non-significant, except in lung cancer where U = 2.33, P < 0.05, RR = 8.3, AR= 1.43/1000, with excess mortality 88%, though for cancer as a whole the difference was significant. Owing to the small number of smokers and deaths, none of the individual causes of death in females showed any association wlth smoking. 6. Distribution of Loss a~ng Smokers and Non-smokers The total number of subjects lost te the study was 116, accounting for 4% of the total subjects. More males were lost than females. The distribution of smokers and non-smokers among those lost is shown in Table 7. TABLE 7. DISTRIBUTION OF SUBJECTS LOST TO THE STUDY Male Female Age Group Smokers Non-smokers Smokers Non-smokers <45 15 2O 0 45-54 9 9 I II 55-64 17 5 3 8 >65 0 6 I 0 Total 41 40 5 30 The difference of loss between smoker and non-smoker was non-signlficant both in males and in females, with U = 0.89 and 1.22 respectively. TI08350844

163 DISCUSSIOM The similarity in the smoking histories taken ten years apart, by different i~dividuals, without utilization of the data from the former survey by the latter interview, indicates that the smoking histories taken in 1972 were reliable. Since all the workers in China are supported with free medical services, free medications and free hospitalizations, usually they would go to see doctors ~heneve~ ~hey get ill, be hospitalized if the disease were severe and be diagnosed before death. Only for very few sudden deaths were the diagnoses made after death, according to the history of the illness. Therefore as a whole, the causes of deaths were accurate too. The percentage of lost subjects was small and there was no predominant loss among smokers or non-smokers. In addition, deaths and losses were frequently checked as subjects were also given repeated electrocardiographic examinations to detect cases of coronary heart disease. This helped in the calculation of person-years of exposure of different age groups for this paper. The use of person-years of exposure as the denominator in the calculation of various death rates minimized the effect of lost subjects and deaths at different time periods. Thus the association demonstrated between smoking and mortality may reflect the real situation. The finding that mortalities of smokers were higher than that of non- smokers, both in males and females, with differences significant statis- tically, shows that smoking has similar harmful effects in Shanghai, China as in other parts of the world. The difference in the overall mortality of males to that of females was chiefly due to a difference in the percentage of smokers in the two sexes. The difference became non-significant when the mortalities were compared either for smokers or non-smokers separately. It is interesting to note also that the association of smoking with different causes of death differed in the sexes. For males, smoking was associated with an excess of malignant tumors, expeclally lung cancer to which smokers had 8.3 times the risk and 88% excess mortality of non-smokers. For females, no association was demonstrated between smoking and malignancy. But smoking was associated with increased cardiovascular and other diseases mortality instead. The reason for the different effect of sm~klng in different sexes remains unclear and awaits further research. The association of smoking to mortality has not been reported elsewhere in China. This paper ind$cates that the situation at present simulates some- what that of the developed countries, and strongly suggests that China should do her part in controlling the harmful effects of ~oking o~ health. The author wishes to thank Professor Zhan~ Zhao-huan for his kind advice, and also is grateful to Wang Hui-zhen, Jin Huai-gen and Qin Hui-di for their help in collecting the mortality and follow-up data, and to Cheng Xing-bao for her help in checking the calculations. TI08350845

I. World Health Organization. Smoking and its effect on health. Geneva, Switzerland: WhO, 1975. W~O Technical Report Series No. 568. 2. World Health Organization. Controlling the smoking epidemic. Geneva, Switzerland: WHO, 1979. WHO Technical Report Series No. 636. 3. Shanghai collaborative group on coronary heart disease. Matched survey for myocardial infarction. Chinese J Preventive Medicine 1981; 15(2): 75. 4. Wang Gui-heng. Smoking and coronary heart disease. Chinese J Cardiovas- cular Diseases 1979; 7:170. 5. Armstrong B. Cancer in China. Int J Epidemlol 1980; 9(4): 305-316. Coronary heart disease survey group. Coronary heart disease survey in factories and countryside. Acta Academlae Medlcinae Primae Shanghai 1979; 6(2): 65. 7. Armitage P. Statistical Methods in Medical Research. London: Black- well, 1971: 384-391. TI0835~346

165 G.H. Miller, Ph.D. studies on Smoking (SOS) 125 High Street, Edinboro Fen~sylvania 16412, U.S.A. D.R. Gerstein, Ph.D. Committee on Basic Research in Behavioral and Social Sciences National Research Council Washington, D.C., 20418, U.S.A. The difference in life expectancy between U.S. males and females has increased from 2 years in 1920 to nearly 8 years in 1980 (I). A number of theories have been proposed to explain this variation: The Genetic Difference, Female Hormone Protection, Stress, and Smoking. The impact of smoking on mortality has had extensive scientific investiga- tion, but few studies have dealt directly with the mele-female longevity difference (MFLD)o Preston (2) concluded that the international increase in I4~LD from 1930 to 1963 was due largely to cigarette smoking, Retherford (3) estimated that leas than half (47Z) of the U.S. ldTLD in 1962 was due to cigarette smoking. Surveys by Haenszel and associates (4) Godley (5), and Enstrom and Godley (6), show some differences in the MFLD in the non-smoking segment of their samples. Casey and Casey (7) and Miller (8) investigated data from certain rural counnunities in the United States and Ireland where cigarette smoking was vlrtually nonexistent and found no female longevity advantage. The mortality data in the studies reviewed generally included all causes of death or deaths from lung cancer. We hypothesized that the differences among results related to male-female longevity might be due to varying incidences of deaths from traumatic causes (fatal accidents and suicides) and to differences i~ the methods used to identify and classify study ~rticipants as non-smokers. Therefore, we tested the hypothesis that no ~le-female longevity differences would be found if the effects of fatal injury were removed and particular care was taken to exclude former smokers from the non-smoker category. The statistical method used in the study was a two-sample, cross-sectional analysis devised by B~ens¢el a~d associates (4). We combined p~eviously collected retrospective data (9) on the lifetime smoking habits of adult men and wo~_en who had died in Erie County, Pennsylvania, in the years 1972-74 (~=~a derived ~ro~ interviews vitn close relatives ot ~e accesses} vit~ new lifetime survey data on the s~kin~ habits of men and women ~o lived in T108350847

Erie County in 1972-74 (data obtained by reverse projection from a 1979 su~ey). Identical items were used in the telephone interviews in both surveys. Erie is a medium-slzed urban area in tF~ northeastern United States. According to the 1970 U.S. Census, the population of Erie County was 263,654. The county has had a history of low migratory rates. The MI~LD in Erle County age 30 and older was approximately 6 years, a difference that is close to the natlonal figure. DATA O~l Tt~ D~,C~D~T POPULATION In 1973 Miller (9) began the Northwestern Pennsylvania Study on Smoking and Health. Death notices for the years 1972-74 provided the names of nearly every 'person who had died in Erie County and the names of their aurvlving relatives. Telephone numbers of up to three surviving relatives were idea- tilled for each death notice. No telephone numbers were obtained for 15% of the decedents. All deaths caused by accident, homicide, or suicide, and all decedents under age 30 were eliminated from the analysis. Information was collected on the exact cause of death and the decedent's age, occupation, .. and smoking history. In the National Mortality Survey (5,6), survivors' reports of the decedent*s smoking habits were compared with the decedents' medical records and the decedents' own reports before death. In respect to recent smoking status, agreement was nearly perfect. However, the distinction between lifetime non-smokers and long-time former smokers was much less reliable. In our telephone survey, therefore, particular attention was paid to probing deeper into an initial response to identify all former smokers. No telephone contact could be made in approxlmately 10% of the cases. Of the 6,930 persons aged 30 and older who had died in Erie County in the years 1972-74, useable interviews were obtained for 63% or 4,394 decedents. For the purposes of this report, only the lifetime non-smokers (2,195) - persons who had smoked less than 20 packs of cigarettes during their ii~etlme - were considered for analyses. DA~A ~ ~ LIVII~ POPUlaTiON, For the years 1972-74, a 2% random sample of household telephone numbers was taken from Erie County telephone directories. The names and telephone n~rs were compared with those in the 1979 directory. A c~rrent listing was not obtained for approximately 14% of the 1972-74 sample. The items used in telephone interviews with the living population were iden- tical to those used in interviewing the relatives of decedents except that inapplicable items such as "cause of death" were omitted. the household 30 years og age or older sam queried. Information was obtain- ed on a total of 3,916 residents of Erie County, a 96~ response rate. ~his information was then analyzed to adjust for the smoking status of each T108350848

respondent in the 1972-74 years. with this segment of the study. ~'ne Rational Research Council assisted A~tALYSIS O~e LL.'~E ~he percentages of non-smoking men and women by age group, as restrospec- cively estimated in the 1972-74 population of Erie County, are shown in Table I. TABLE i. 5-Year Non-smoking Non-smoking Cohorts men women 30 - 34 ............. 38.2 35 - 39 ............. 31.8 40 - 44 ............. 27.7 45 - 49 ............. 24.4 50 - 54 ............. 30.9 55 - 59 ............. 26.7 60 - 64 ............. 21.7 65 - 69 ............. 26.3 70 - 74 ............. 30.6 75 - 79 ............. 35.4 80 - 84 ............. 31.4 85 and over ......... 36.8 59.9 56.7 59.6 61.0 59.8 64.3 62.4 62.2 72.2 82.4 g3.0 89.3 Population estimates for 1973 provided by the Pennsylvania Department of ~ealth were multiplied by 3 because the mortality data covered 3 years. The results were then multiplied by the proportions of non-smoking men and women in Table I. These computations provided the denominators needed for calcu- latlng the mortality rates for each age-sex cohort, the numerators being the non-smoker deaths recorded from the surviving relatives. The llfe table computations were done according to the standard demographic procedures used by ~hryock an~ Siegel (10) and adjusted by Chlang constants (ii), and are reported in ~able 2. DI$C~ISS!ON Table 2 shows a consistent pattern of similarity in the life expectancy for a!l 5-year cohorts of no~-sm~king men and wo~en ~er age 30 in Erie County. ~ nearly identical life expectancy found when traumatic deaths are removed and improved procedures are instituted for classifying non-smokers corrobo- rates our hypothesis that differences in smoking habits are responsible for observed m~le-female longevity differences. The result supports Preston (2) and Retherford's (3) conclusion that ~ach of ~LD since lq~O ~= ~.,t-ble ko ~ne eI~ects of cigarette smklng. ~e data from ~r study, rosined with ~re than four decades of research shying the destructive force of ciga- rette s~ke and ~he fact that ~n ha~e a greater number of s~k~ng years T108350849

168 MILLEE AND GERSTEIN than women, provides ample evidence of the impact that sr-oklng has on the MFLD. Although all studies ~n which the KFLD has been investigated have revealed a substantial detrimental impact of cigarette smoking, several of the investi- gators have reported residual MFLD not accounted for by smoking. There are two likely explanations for this residual difference. First, traumatic deaths occur ~n the greatest numbers among men in the lower age brackets - age 20 through 55 - and these relatively early deaths produced a dispropor- tionate impact on longevity statistics. In our study th~s effect was eliminated. In any research on the MFLD, investigators should take traumatic deaths {nto account. TABLE 2. RESULTS OF CALCULATION OF THE ABRIDGED LIFE TABLES FOR NON- SMOKERS IN ERIE COUNTY, PA., FOR 1972-74, BY S~X AND AGE GROUP ~ ................... 2 4,5-4 9 .................... 11 ~5-69 ................... 41 70-?4 .................... 3~ 75-7~ ................... ~9 ~ .................... 103 8.241 .00024 .52 .00120 6,215 .00048 .54 .00240 5,7~0 .00051 ,54 .00255 5,545 .00198 .5~ ~.983 .00186 .K~ .00926 5,431 .00552 .52 .02724 ~,576 ,01118 .52 .054.¢4 3,121 .01313 .52 2,~4~ .01472 .51 .07104 2.030 ,03399 .51 ,15~8 1,051 .09800 .48 .3~050 1,048 .07~-42 1.00000 30-34 .................. 5 13,975 .00035 .52 ~-39 .................... 6 11,~69 .00050 .54 ,00250 4044 ............... 14 13.641 .00102 .54 .00509 4~49 ................. ~2 15.275 .00144 ,54 .00718 50-54 .............. 40 14,859 .00269 .53 .013,37 55-59 ................ 50 14.257 .00350 .52 .01735 ~0-64 .................... ~2 11,585 .00794 .52 .03~ ~r)-89 .................. 141 9285 .0151B .52 .07323 ?0-74 .................. 188 ~,558 ,021~ .51 .I0419 7~.7~ .................... ~ 7,~42 .044b'~ .51 ,~0132 ~ .................... 411 4,~31 J)B507 .48 .14831 B5 and olcler ............ 4~6 4,482 .10397 . 1,00000 100.000 120 4~,700 5,0~5,~45 50.9 ~,880 240 4~.800 4,589,245 46.0 ~.640 ~54 497,56~ 4.0~7,¢45 40.9 ~.:~S SeO 4~4.480 3.~9.~0 36.2 ~,405 ~11 4~9.752 3.095400 31.5 97.495 2,556 4~0.B35 2,~0564S 26 7 ~4,B39 5,1~3 4~1,287 2.124,B13 22,4 ~9.576 5,707 434,112 1,663.525 18.6 83,969 5.~5 404.932 1,22~.414 14,6 78.004 12.237 ~59,427 ~24.482 10,5 ~5 767 25,6~2 2~4,630 4~5.055 7.1 40.085 40.085 200,425 200,425 ' 5.D 100,000 175 4~.5~2 5.080,248 50.6 ~,825 250 4~.~ 4,~,~6 45.7 ~,575 ~7 4~.~7 4,~2186 ~.8 ~,~ 711 493,~2 3,~5 579 360 ~,357 1,315 ~,4~7 3,072.017 31.2 97,~2 1,~4 ~1.~ 2.~.520 ~.6 ~,~ &715 467,~2 2,102,520 22,0 ~1,~3 ~,71~ ~I.~7 1,~5.018 17.8 ~,~ B,M9 402.~7 1.1~,~1 14.1 ~,~3 15,317 ~2.122 ~1.~ 10,4 ~,7~ 21.1~ ~,~17 ~,922 7.4 ~,~1 ~.~1 1~,~5 I~.~5 '5.0 Second, a review of the methods used in other studies shows ambiguities in distinguishing between non-smokers and forme_r smokers. W~en answering ques- tions about their own smoking habits or those of their relatives, many interviewees classify themselves or their relatives as non-smokers even T[O~sOosO

LIFE-EXPECTANg'Y OF 169 though they may have smoked in the past. This imprecision results in the classification of deceased former strokers as non-smokers and increases the mortality rates attributed to non-smokers. In our study we minimized this confounding factor. Investigators conducting studies should make precise classiflcation of smoking status. The results of our study have two implications. First, standard census data on llfe expectancy in the United States merge the two very different mortal- ity rates of smokers and non-smokers, producing inaccurate estimates for each category when these categories are considered separately. These differences in categories need to be taken into account in the calculations of pension and life insurance premiums which use merged data. The second implication is that a person's sex-role behavior has an impact on health and morbidity and, consequently, on longevity. The most significant change in younger women's health habits in the United States over the past decade has been a large increase in teenage and preteenage smoking. The 1980 Surgeon General's Report on the health consequences of smoking for women indicates that teenage girls have surpassed teenage boys in the percentage of smokers. When cohorts of women who have smoked as much as men reach the later decades of life~ the results of our study suggest that their lives will be shortened as much as men's and that the present differences in longevity between men and women (MI=LD) will disappear. A~OWLED~EMENTS Parts of the work reported here were supported by a grant from the National Institute on Drug Abuse (Contract No. 271-76-331) to the Committee on Substance Abuse and Habitual Behavior, National Academy of Sciences, and by the American Cancer Society of Erie County, the Northwestern Pennsylvania Lung Association, and the Heart Association of Erie County. U.S. Bureau of Census. Statistical abstract of the United States. U.S. Government Printing Office, Washington, D.C., 1979: 70. Preston SH. An international comparison of excessive adult mortality. Pop Stud 1970; 24: 5-20. Retherford RD. Tobacco smoking and the sex mortality differential. Demography 1972; 9: 203-216. Haenszel W, Loveland DB, Sirken MG. Lung cancer mortality as related to residence--and sm~king--~i~-tor-i~s--i~--w~i~e-~m-l-~.s~--J--N~--Canc~r--~n-s~-it 1962; 28: 947-1101. Godley FH. Cigarette smoking, estimates from national samples. Maryland, College Park, 1974. social factors, and mortality: new Doctoral dissertation, University of TI08350851

170 MILLER AND GERSTEIN Enstrom JE, Godley FH. Cancer ~ortality a~ong ~ representative sample of non-s~okers in the ~n{ted States during 1966-1968. J Nat Cancer Instit 1980; 65: 1175-1183. 7. Casey AE, Casey JG. Long-lived m~le population with high cholesterol intake, in Slieve Loughner, Ireland. Ala J Med Sci 1971; 7: 21-26. 8. Miller GH. Male-female longevity comparisons among the Amish. Indiana State Med Assoc 1980; 73: 471-473. 9. Miller GH. Smoking and longevity. J of Breathing 1976; 39: 2-14. Illinois Lung Association, Springfield, III. I0. Shryock SH, Siegel JS. printing (revised). Office, 1975. The methods and materials of demography. Third Washington, D.C.,: U.S. Government Printing II. Chiang CL. Introduction to stochastic processes in biostatistics. New York: John Wiley & Sons, Inc., 1968. TI08350852

171 SMOKING AND VARIOUS PHYSICAL C~MPLAINTS" Niroshi Ogawa, MEd, MPH, DSc Suketami Tominaga, MD, MPH Division of Epidemiology Aichi Cancer Center Research Institute Tashiro-cho, Chikusa-ku, Nagoya 464, Japan Kunio Aoki, MD Department of Preventive Medicine Nagoya University School of Medicine Tsurumai-cho, Showa-ku, Nagoya 466, Japan IIq'r~DuC'rlON Physical complaints often motivate people to seek medical advice and to avoid possible causes. Thus the complaints themselves may have the poten- tial for encouraging disease preventive behavior. This study is to clarify the relationship of cigarette smoking to various physical complaints, some of which may be related to diseases associated with smoking habits. I~THODS A health screening questionnaire survey was conducted in 1970 among civil service employees of Nagoya, Japan, on physical complaints, food habits, personality traits, and social background. From 10,681 males over age twenty, 2,828 males were randomly selected for analysls by age and smoking status stratification (about 25% samples for those who had never smoked and for current smokers, and 50% samples for ex-smokers, Table l). A list of physical complaints was printed on a questionnaire, and respondents were asked to check either "yes" or "no", or among "frequently", "sometimes", or "never". The percentage of men who checked "yes", "frequently" or "sometimes" was calculated. These percentages were compared, and the dose- response relationship was examined by linear regress{on analysis in zelation to the amount of smoking and the number of years after having stopped smoking (1,2). Table 2 summarizes the results on smoking and physical complaints. Prevalence of various physical complalnts increased significantly with the number of cigarettes smoked per day for life. These complaints included sputum, cough, frequent use of stomach medicine, stomach pain, diarrhea, sh~rtness~f-b~eath~--~a~itazi~r~ys~ne-a~-~s~ti~ff_sh~ders~eck--ten-s/~on~, lumbago, back pain, hoarseness, numbness of limbs, swollen eyelids, lethargy, chewing difficulty due to poor teeth, history of jaundice, frequent use of health me.dicine, and decreased appetite. Urination at night Tl08350853

170 Enstrom JE, Godley FH. Cancer ~rtality among a representative sample of non-smokers in the United States during 1966-1968. J Nat Cancer Instit 1980; 65: 1175-1183. 7. Casey AE, Casey JG. Long-lived male population with high cholesterol intake, in Slieve Loughner, Ireland. Ala J Med Sci 1971; 7: 21-26. 8. Miller GH. Male-female longevity comparisons among the Amish. Indiana State Med Assoc 1980; 73: 471-47B. 9. Miller GH. Smoking and longevity. J of Breathing 1976; 39: 2-14. Illinois Lung Association, Springfield, III. I0. Shryock SH, Siegel JS. printing (revised). Office, 1975. The methods and materials of demography. Third Washington, D.C.~: U.S. Government Printing Ii. Chiang CL. Introduction to stochastic processes in biostatistics. York: John Wiley & Sons, Inc., 1968. New T108350854

171 AND VARIOgS PHYSICAL COI~LAINTS' Hiroshi Ogawa, MEd, MPH, DSc Suketami Tominaga, MD, MPH Division of Epidemiology Aichi Cancer Center Research Institute Tashiro-cho, Chikusa-ku, Nagoya 464, Japan Kunio Aoki, MD Department of Preventive Medicine Nagoya University School of Medicine Tsurumai-cho, Showa-ku, Nagoya 466, Japan INTRODUCTION Physical complaints often motivate people to seek medical advice and to avoid possible causes. Thus the complaints themselves may have the poten- tial for encouraging disease preventive behavior. This study is to clarify the relationship of cigarette smoking to various physical complaints, some of which may be related to diseases associated with smoking habits. NEI'HODS A health screening questionnaire survey was conducted in 1970 among civil service employees of Nagoya, Japan, on physical complaints, food habits, personality traits, and social background. From 10,681 males over age twenty, 2,828 males were randomly selected for analysis by age and smoking status stratification (about 25% samples for those who had never smoked and for current smokers, and 50% samples for ex-smokers, Table I). A list of physical complaints was printed on a questionnaire, and respondents were asked to check either "yes" or "no", or among "frequently"~ "sometimes", or "never". The percentage of men who checked "yes", "frequently" or "sometimes" was calculated. These percentages were compared, and the dose- response relationship was examined by linear regression analysis in relation to the amount of smoking and the number of years after having stopped smoking (1,2). ~ES~LTS Table 2 suu=narizes the results on smoking and physlcal complaints. Prevalence of various physical complaints increased significantly with the number of cigarettes smoked per day for life. These complaints included sputum, cough, frequent use of stomach ~edicine, stomach pain, diarrhea, shortnes~-of--brea~,--p&~t-~n, dy~e~-f-f--shouldetg~--~e~ te~s-~on, lumbago, back pain, hoarseness, numbness of li=bs, swollen eyelids, lethargy, chewing difficulty due to poor teeth, history of jaundice, frequent use of health =e_dicine, and decreased appetite. Urination at night T108350855

172 OGAWA, TOMIHAGA ARD AOKI was the only exceptional complaint which decreased significantly with the amount of smoking. Almost one-third of all examined complaints 8bowed significant linear dose-response relations (Figure i). Never Age Smoked TABLE I. AGE DISTRIBUTION OF SUBJECTSl Current smokers2 Ex-s~okers3 Total Light Medium Heavy Total Short Medium Lon~ 1-19 20-29 30+ 0-3 20 - 29 192 571 247 241 83 62 47 15 0 30 - 39 183 633 310 249 74 118 63 29 26~ 40 - 49 71 533 305 168 60 86 41 23 22 =. 50 + 47 278 172 79 27 50 15 9 25 Total ~93 2,015 1,034 737 244 316 166 76 74 = Mean 34 37 39 36 36 39 36 39 4& ~ S.D, 11 II 11 I0 I0 I0 9 9 I Male civil service employees (age > 20 years), Nagoya, Japan. 2 Subdivided into three groups by the average numbers of cigarettes smoked per day for life. 3 Subdivided into three groups by the number of years after stopping smoking. S,D. : Standard Deviation Loss of eyesight, weight loss after age 20, and sick leave from work were more frequent for current smokers than for those who had never smoked." However, no significant dose-response relation was observed for these complaints. On the other hand, history of albuminuria was less frequent for, . current smokers than for those who had never smoked. ~'~ Some kinds of physical complaints were found to decrease significantly wlth~. the number of years after stopping smoking. These included sputum, cough,~ numbness, chewing difficulty due to poor teeth, and sick leave from work~- (Figure 2). Complaints of white tongue, swollen eyelids, and decreasedi! appetite were significantly less frequent for ex-smokers than for curre~t.~ smokers, but they were not associated in a linear dose-response manner with the number of years after stopping smoking. In contrast to these~ complaints, nosebleeding, blurred vision in the center of the visual field,. constipation, and history of parasitic disease increased linearly with the.~ number-o~ ye~r~---a~re~--~ro-~p~-,Yg--~m~f~g--(Figure-2)~ ex-s-m~r--comp~a~ ~Dre frequently than the current smoker in regard to a history of ulcer, strangling sensation in the chest, urination at night, illness, and nose disturbance. T[08350856

SMDKIN~ AND PI~fSICAL COMPLAI[~rs 173 TABLE 2. PREVALENCE OF COMPLAINTS BY SMOKING STATUS AND RELATIONS OF COMPLAIITrS TO THE AMOUNT OF SMOKING AID TO THE NUMBER OF YEARS AFTER STOPPINC. SMOKING1 Sp:c~'m 33.9 46.~ 7.50~ 4~.8 37,~ -1.17-- ~ 32.6 ~3.5~ 5.~ ~],5 32.~ -1.~ ~s~ ~u ~ch cold ~2.3 Al,5 0.15 ~1.5 &2.6 0.~ ~bl~l~S 19.7 18.9 1.18 ]8.9 25.1 • 0.97~ As~ A.7 A.Q 0.52 h.9 2.6 ~.20 H£s~cr~ o~ bl~ 8p~t~ 2.1 2.2 0.39 2.2 2.2 ~.~ ~eac~ve St~c~ ~diclne 37.8 43.~ • 2.19 + 43.5 ~.2 0,41 St~ch ~£n ~.6 39.2 2.71~ 39.2 37.9 ~,23 ~arrhes 26.& 29.3 3.1~ 29.3 31.7 0.65 ~t~e ~Eue 18.8 22.5 3.~(~) 22.5 ~7.3 e ~r ~1 ~In 16.1 39.a l.~ 19.4 21.4 0.19 ~scory of at~h ulcer ~.2 6.0 1.07(~) ~.0 7.0 ~.13 ~story of du~en*l ~lcer 3.0 3.3 0.29 3.3 6.1 " O.ll ~rculmcory ~or~ness of broth 22.~ 2~.8 2,3~ 24,B 24.0 By~rtens~on 23.5 20.~ ~.68 20.~ 22.5 ~.~ ~Ip~;on 15.2 18.8 2.~ 18,8 20.0 0.07 ~spnea 13.3 1~.0 1.69 * 16.0 18.~ Arrh~c~a 11.7 ]3.~ 0.99 1~.~ 13.6 ~.~7 Fmscular and skeletal Stiff shoulders 55.6 63.9~' 4.18~,-# 63.9 65.~ ~.15 Neck ~enslon 42.6 49.~* 3.8~ ~9.9 ~3.2 ~.)) ~o 27.7 ~.8~ A.l~ ~.B ~.fi ~.2g Pa~ In ~olnts 23.8 28.2 l.&2 28.2 3).& 0.52 Pain in the Mck 19.2 2&.O ~ 1.~ + 24.0 22.8 ~.19 Psln In the legs 9.2 12.1 1.14 12.1 10.4 0.~ ~ort durlnS ~d ~s~her 10.6 ll.l 0,~ 11.1 13.9 Peripheral clrculstory ~d ~rvous B~rseneas 19.6 24.9 * 3.8~ 2&.9 25.0 ~.~ ~Idneas ~n l~ba 22.] 22.0 0.~ 22.0 2~.I N~bness ~n flabs I4.& 21.~ 3.~ 21.9 15.3~ ~.69 ~r ebral ~;a 19.8 17.3 ~.~ 17.3 18.5 ~.03 ~o~len eje1~ds 12.~ 19.8~ 3.8~ ~9,~ 14.5 s Sit.sling sensstlon in the chest 15.7 16.2 1.32 16.2 21.2 " ~c~rgy 14.2 17.1 2.)~ 17.] 18.4 0.55 ~es~ ~In 15.& 14.2 0.81 ]~.2 15.2 0.&8 S~ech dl~ur~nce I0.9 10.3 0.~ ]0.3 10.8 0.05 M;~tor~ of ~mint%n8 I0.4 I0.5 0.33 lO.~ 9.8 Bie~or~ of cr~pxng 5.9 8.4 1.4~(*) 8.~ 6.8 31urr~ vls;on xn the cen~ez of 5.2 6.8 0,20 6.8 7.0 0.4] ~lle. 1xmbs 4.3 6.1 0.)7 6.l 6.8 0.~ Age-related L~ss of e!~es~ght 40.0 &5.1 • 1.16 &.~.] 41.9 Che~ln~ dlff~cult~ due to 20.8 2~.~ ~.3~ 2~.~ 21.4 ~.71 ~r ~h ~rln~1on ~ nlgh~ 19.7 )4.6~ -2.~ 14.6 23.~ 1.~(~) ~stor~ of rhe~t~ d.4 4.8 0.~ 4.8 5.2 0.12 Clauc~ 1.9 1.0 ~.52 1.0 1.8 0.~ ~rs Bi~or~ of ~rrholds 28.1 28.9 1.85 28.9 33.2 0.~ TXnn~tus 23.3 23.3 0.83 23.3 2&.6 0.07 ~stur~nce in ~ ~ 23.8 21.0 0.12 2].0 28.1~ 0.55 ~ns~x~ion 18.6 19.7 I.OT 19.~ 23.2 0.73 ~s~ ~o develop ~tls 1~.9 12.6 ~.67 12.~ 13.1 0.16 ~s? ~o develo? rashes 12.7 13.1 0.33 13.1 14.5 0.~ ~xs~ory of al~xnur~a 12.0 8.5 • -~ 8.5 9.6 0.02 ~story of ~agn~ce &.2 6.~ 1.4~ 6.~ 7.7 ~.13 Nx~tory of ~ra~lctc d~mse 2.7 2.t 0.27 2.1 3.3 0.29 ~il~h ~tcl~e ~.~ 45.e * 2.22 * 45.8 AS.& 0.65 Fstxg~e ~ the ~rn~g 33.3 33.) 0.~8 33,1 ~.7 ~.4~ ~r :~xcalll- Xll 27.2 24.3 0.22 ~r~d t~xze 19.4 23,& 3.1~ 23.4 17.1 " Ill ~I~ c~rre~ly 21.5 lS.V 0.12 18.~ 21.I ~.23 Slee~ dXs~ur~=e I0.7 10.4 1.23 lO.t II.7 0.20 P~r a~t~e 4.2 5.6 O.S~ 5.~ ~.3 TI08350857

OGAWA~ TOMIRAGA ARD AOKI FIGURE 1. AMOUNT OF SMOKING AhD AGE-ADJUSTED PREVALENCE RATE OF COMPLAIh'~S Only Complaints ShowinE S~Enificant Associations ]e ~4~ ~ro~,eed TI0~350353

~]40KING AND PHYSIC%L FIGURE 2. SMOK~I~G STATUS A~D AGE-ADJUSTED PREVALENCE RATE OF COMPLAINTS Only Complaints Showing Significant Associations sputum cough noseb|eed T[08350859 -- ÷ -- numbness blurred vlelon Chewing of limbs In center of difficulty vleuef field i | constipation history of sick leave 80 (- parasitic from work • :'SO t disease 0~ C SM L

176 OCA~, T~I~L~A A~D A~KI COMMENTS AND ~ONCLUSIONS Those physical complaints in various parts of the body which were found to be related to smoking may be largely attributed to toxic substances in inhaled smoke, although a stimulus-seeking tendency in the smoker's person- ality, food habits, and lower socio-economlc conditions may also contribute to these complaints. Some physical complaints which were more common among ex-smokers may suggest that these work as motlvational factors for smokers _ to quit smoking. The observed relationships between smoking and physical complaints in this study were largely consistent with the results in a U.S. study (3). Further study is needed to establish a definite causal relation between smoking and physical complaints. People may be more concerned about present physical complaints than about remote, life-threatening diseases which usually appear much later in life. The relationship between smoking and physical complaints should be given more attention for the purpose of providing better education on smoking and health .... Chochran WG. Some methods of strengthening the common X2 tests. Biome- trics 1954; 10: 417-451. Armitage P. Test for linear trends in proportions and frequencies. Biometrics 1955; 11: 375-385. Hammond EC. Smoking in relation to physical complaints. Arch Environ Health 1961; 3: 146-16~. TI0~50~o0

THE EFFECT OF CIGARETTE-SMOKIN~ 0~ THE 2.414 KM R~N OF 155 ADI~LT MALES IN SINGAPORE Teck Chin, Ong, D.Phil. Pui Yong, Tan~ Ph.D. Department of Physiology Faculty of Medicine National University of Singapore INTRODUCTION 177 The fact that smoking contributes significantly to the increase of morbidity and m~rtality in some important diseases threatening public health is well documented (1,2). It is also widely known that the most important noxious components in cigarettes include carcinogenic substances, nicotines, carbon monoxide and traces of metallic ingredients liberated by the burning of tobacco. Performance in sports is impaired by chronic tobacco consumption. Even though the decrease of physical work capacity is not always significant when tested splroergometrically (3), it becomes more pronounced at high workloadsj with growing age and increasing duration of smoking. This study describes the relationships of smoking habits with cardiorespira- tory fitness and suggests explanations for the observed resulcs. Cardio- respiratory fitness is one of the key parameters in determining the aerobic fitness and hence total physical fitness of an individual. The 2.414 km (1.5 mile) running test is used as a measure of the cardiorespiratory or aerobic fitness. MATERIALS AND METHODS The subjects were male volunteers (age range 20 to 29 years). All subjects underwent a detailed medical examination and assessment. They were cleared of any significant medical problems, and each subject was requested to com- plete a smoking history questionnaire. A consent form to participate in the test was signed by each subject in accordance with the Helsinki Declaration of 1975, The height and weight of all subjects were recorded. The classi- fication used for the subjects was as follows: I) non-smokers, 2) those who smoked I to i0 c~garettes/day~ 3) those who smoked Ii to 20 clgarettes/day, 4) those who smoked 21 to 30 cigarettes/day, 5) those who smoked more than 30 cigarettes/day. Address correspondence to: Dr. Teck Chin, One, D___~aKt_ment of Physiology, Fa=ulT>.~-of--Medicine~ ~at~--Dnlverslty of Singapore, Lower Kent Ridge Road, Singapore 0511, Republic of Singapore. TI08350861

The test adopted in this study is based on Cooper's studies (4,5) and measures the mini=um tim in minutes and seconds taken to complete the dis- tance of 2.414 km (1.5 mile) and compared against a similar table (4,5) for the predicted Max VO2 values. The run was conducted in the early morning and before breakfast on an accurately measured, flat, even and firm track, in groups of about 80 runners at a time. The time of the run was kept by using stop-watches (Seiko Quartz, Digital Type cal. 5023, Japan). Pacers were provided to encourage runners to do their best to reach their respec- tive performance levels. Smokers were not allowed to smoke cigarettes for several hours before the test. In this way it was hoped that the acute effects of smoking would not influence performance significantly. RESULTS Table 1 shows that non-smokers as a group were significantly older (p<0.001) compared to smokers. There was no significant difference in the height and weight of the smokers compared to non-smokers. However, since the age groups of both smokers and non-smokers fall within the same age category of 20 to 29 years in Cooper's table of standards for the 2.414 km run-walk test~ which was later used for predicting the Max VO2 values, no significant influence on the results is expected. TABLE I. THE MEAN AGE (yrs), HEIGHT (cm) AND WEIGHT (kg) OF 155 ADULT MALES IN SINGAPORE Group n Mean Age Mean Height Mean Weight ± ISD ± ISD ± ISD Non-smokers 77 (49.7%) 22.6 ± 1.6 167.3 ± 6.2 58.1 ± 6.1 Smokers 78 (50.3%) 21.8 ± 0.8 165.8 ± 7.1 58.9 ± 7.7 All 155 (100%) 22.2 ± 1.3 166.5 ± 6.7 58.5 ± 6.9 Table 2 shows the mean 2.414 km performance of smokers and non-smokers. Smokers performed poorly compared to non-smokers (p<0.001). As expected, the predicted Max VO2 value for smokers was lower than for those who did not smoke. TABLE 2. THE MEAN 2,414 km RUN TIME AND PREDICTED MAX VO2 VALUES OF 155 ADULT ,MALES IN SINGAPORE Group Non-smokers Smokers Mean 2.414 km Run Time (mins:sec) ± ISD Predicted Max VO2 (ml/kg/min) ± ISD 77 12' 54" -+ I' 15" 39.8 + 3.3 78 13' 44" ± I' 22" 37.3 ± 3.6 155 13' 19" -+ i' 22" 38.5 -+ 3.6 p (Non-smokers vs Smokers) < 0.001

2.414 179 Table 3 shows that there is no statistical difference when the mean age, height and weight of categories A, B, C, D of smokers are compared with one another and with non-smokers. TABLE 3. COMPARISON OF MEAN AGE (yrs), HEIGHT (cm) AND WEIGHT (kg) BETWEEN DIFFERENT CATEGORIES OF SMOKERS AND NON-SMOKERS Category n Mean Age ± ISD Mean Ht ± ISD Mean Wt ± ISD A 23 21.9 + 0.8 165.7 ± 8.6 58.9 ± 6.2 B 26 21.8 + 0.8 167.5 + 6.7 59.2 ± 8.6 C 17 21.7 ± 0.9 164.0 + 5.9 58.0 ± 7.6 D 12 21.9 ± 0.5 164,9 ± 6.3 60.1 ± 7.8 All 78 21.8 ± 0.8 165.8 ± 7.1 58.9 ± 7.7 NS 77 22.6 ± 1.6 167.3 + 6.2 58.1 ± 6.1 A : I-I0 cigarettes/day B : 11-20 cigarettes/day C : 21-30 cigarettes/day D : >30 cigarettes/day All : A + B + C + D NS : Non-smokers In Table 4, a comparison of the 2.414 km run time between different cate- gories of smokers and that of non-smokers can be seen. Notice that non- smokers have the shortest running time compared to all the categories of smokers. In addition, the 2.414 km running time increases with the number of cigarettes smoked. TABLE 4. Category n RELATIONSHIP BETWEEN THE NU~IBER OF CIGARETTES SMOKED PER DAY AND THE 2.414 KM RUN TIME 2.414 km Run Time (min:sec) ± ISD Predicted Max V02 (ml/kg/min) ± ISD A 23 12 B 26 13 C 17 14 D 12 14 All 78 13 NS 77 12 59" + 0' 48" 30"-+ I' 20" 20" ± I' 01" 49" ± I' 45" 44" + I' 22" 54" + I' 15" 39.5±2.1 38.0±3.5 35.8±2.2 34.9±3.8 37.3±3.6 39.8±3.3 A : I-i0 cigarettes/day B : 11-20 cigarettes/day C : 21-30 cigarettes/day D : >30 cigarettes/day All : A + B + C + D NS : Non-smokers

180 0~3 A~D ~ Non-smokers have signlficantly shorter running time compared to category B (p<0.05), C (p<O.O01) and D (p<O.O01) though there was no significant dif- ference when compared to category A. Amongst the categories of smokers, category A have no statistical difference when compared to B but have sig- nificantly shorter running time compared to category C (p<0.001) and catego- ry D (p<0.01). Similarly, category B smokers have significantl~ shorter running time compared to C (p<O.05) and D (p<0.05>. However, category C smokers have no significant difference in their running time when compared to category D (p>0.05). Table 5 shows that the running time for 2.414 km increases with the number of cigarettes smoked per day in smokers who have smoked for less than 5 years and for 5 to 9 years. Comparing like categories, that is, A against A and so on~ in the two groups of smokers, it was found that the running time was similarly affected. Those who smoked for a shorter duration have better (that is, shorter) running time compared to those who had smoked for a long time. This was found to be statistically significant (p<0.O01). TABLE 5. EFFECT OF NUMBER OF CIGARETTES SMOKED PER DAY AND DURATION OF SMOKING ON 2.414 km RUN Smoked < 5 Years Category n n Mean 2.414 km ± ISD Run Time Smoked 5-9 Years Mean 2.414 km ± ISD Run Time A B 16 C 6 D 4 All 45 12' 51" + O' 33" 4 13' 12" ± 1' 22" I0 14' 09" ± O' 57" II 14' 18" ± i' 46" 8 13' 17" ± i' I0" 33 13' 35" -+ i' 31" 13' 58" + i' 35" 14' 25" +-I' 06" 15' 04" -+ i' 49" 14' 21" ± l' 24" A : i-i0 cigarettes/day B : 11-20 cigarettes/day C : 21-30 cigarettes/day D : >30 cigarettes/day All : A + B + C + D p (5 years vs 5-9 years) < 0.001 Comparing the running time of non-smokers with that of each of the different categories of smokers in the group who smoked for less than 5 years dura- tion, it was found that there was no statistical difference except with category C (p<0.001). Comparing the running time of the different cate- gori~es-~th-one--~norher--i~n--£h-i-s-grbup ~-V~fh~ there was a significant difference only between A and C (p<O.01) and between B and C (p<0.05). Likewise, comparing the categories of smokers in the group who smoked for a duration of 5 to 9 years with the non-smokers, it was found that non- smokers have significantly shorter running time compared to B (p<O.05),

SMOKING AND ~E 2.414 KM KND~RANCK FJIN 181 C (p<O.001) and D (p<0.001). There was no statistical significance when non-s=okers were compared to category A. The individual categories A to D were not significantly different when co~pared with one another. Table 6 shows the predicted Max VO2 values extrapolated from Cooper's stand- ard for 2.414 km run time - Max VO2 (ml/kg/min) conversion table. As to be expected, the predicted Max VO2 values in ml/kg/min follow a similar trend to that of the running time; that is, Max VO2 values decrease, suggesting a decrease in aerobic fitness, for smokers who have smoked for a longer duration. The Max VO2 of smokers who smoked less than 5 years against Max VO2 of smokers who smoked for 5 to 9 years was statistically significant (p<0.O01). TABLE 6. Category PREDICTED MAX VO2 VALUES OF THE DIFFERENT CATEGORIES OF SMOKERS Smoked < 5 Years Smoked 5-9 Years n n Predicted Max VO2 Predicted Max VO2 (ml/kg/min) (ml/kg/min) A 19 39.9 + 1.6 4 37.7 +- 3.7 B 16 38.9 + 3.6 I0 36.6 +- 3.1 C 6 36.2 + 2.2 II 35.6 + 2.2 D 4 35.8 _+ 3.5 8 34.5 -+ 4.1 All 45 38.7 + 2.9 33 35.7 + 3.2 A : I-i0 cigarettes/day B : 11-20 cigarettes/day C : 21-30 cigarettes/day D : >30 cigarettes/day All : A + B + C + D p (5 years vs 5-9 years) < 0.001 DISCUSSION i . In this study, the effect of cigarette-smoking on the performance of the subjects in the 2.414 km run reveals that non-smokers, as a group, perform better in the run compared to smokers. Since the 2.414 km run has been used as an indicator of cardiorespiratory endurance, the results suggest that ,.~ non-smokers as a group have better cardiorespiratory endurance fitness com- pared to smokers. The number of cigarettes smoked per day appears to influence the performance of smokers in the endurance run in that the 2.414 km run time increases with the number of cigarettes smoked per day. Non- ~ smokers in general perform better than all the different categories of s~ok- ers in t~e 2_.~I~__ ..... their performance when compared to heavier s~3kers in the categories of those %~o s~oked more than 10 cigarettes a day. There appears to be a rela- tionship between the number of cigarettes smoked per day and duration smoking with the performance in the 2.414 km run. Heavier s=okers who smoked for a longer duration (5 to 9 years) compared poorly with those who TI08350865

smoked a comparable number of cigarettes per day but for a shorter duration(less than 5 years). Also, the predicted Max VO2 values of smokers are consistent with the 2.414 km run ti~e, smokers having lower values compared to non-smokers and heavier smokers having lower values than lighter smokers. These results appear to be consistent with the observation by other workers (6,7) and with the belief that smoking does affect the physical fitness, particularly cardiorespiratory fitness, of individuals and that the severity of the apparent detrimental effects appears to be related to the duration and consumption rate of cigarettes. 1. Fletcher CM, Horn D. WHO-Chronicle 1970; 24: 345. 2. US Department of Health, Education and Welfare. The health consequences of smoking. DHEW Publication 1974; No. 74, 93: 8704. 3. Schwalb H, Flackler R. Untersuchungen uber die kardiopulmonale Leis- tungsfahlgkeit bei Rauchern. Arch Kreislaufforschg 1970; 62: 167. 4. The New Aerobics. New York: Bantam Books, Inc., 1970. 5. The Aerobics Way. New York: M. Evans and Company, Inc., Cooper KH. Cooper KH. 1977. Cooper KH, Gey GO, Bottenburg RA. Effects of cigarette smoking on endurance performance. JAMA 1968; 203: 189-192. Montoye HJ, Gayle R, Higgins M. Smoking hablts~ alcohol consumption and ~mximal oxygen uptake. Med Sci Sports Exerc 1980; 12: 316. TI08350866

183 TOBACCO ADDiCTiON Arid OTHER D~tUG A.~SE AMM A~P~CA~ YOUTH R.T. Ravenholt, M.D. Assistant Director for Epidemiology and Research* William Follin, M.D. Director, National Institute on Drug Abuse Alcohol, Drug Abuse, and Mental Health Administration 5600 Fishers Lane Rockville, Maryland 20857 U.S.A. By every important measure of addiction - psychoactive effects, habitual use leading to dependence and compulsive abuse, physiological and psychological distress upon discontinuance, and tendency to recidivism- tobacco smoking is addictive. In fact, when measured by morbidity and mortality, cigarette smoking is now the most serious, as well as the most widespread, form of addiction in the world. Despite significant improvement in some countries, more persons are smoking in the world today than at the time of the First World Conference on Smoking and Health 16 years ago. Although the propor- tion of the population smoking has been reduced in the U.S. and in some other countries, the total number of smokers has increased along with popu- lation increase, especially in the developing world (i). Two main factors have combined to limit the effectiveness of preventive programs: the power- ful promotional activities aimed at increasing tobacco sales, and the pheno- menon of cigarette addiction. ADDICTIVE NATURE OF CICARETTES Although tobacco has been widely smoked during four centuries, widespread addiction to smoking is a phenomenon of the last century, closely linked to the invention of machine-rolled cigarettes in 1880 (2). Since then, much more has been learned about the addictive nature of ciga- rette smoking - the fastest and most efficient way of getting nicotine to the brain, where it acts through specialized cell formations (3). Nicotine, the opium derivatives, and perhaps the benzodiazepines, are the only drugs of dependence for which specialized receptors of this kind have been identi- fied and studied in detail. One laboratory-determined indication of the abuse liability of a substance is ratings by post drug addicts of their liking for the substance. Nicotine is similar to drugs with known potential for abuse, most notably, opioids and psychomotor stimulants. When the * Now: Director, World Health Surveys, Inc. 4960 Sentinel Drive, Bethesda, Maryland 20816 U.S.A. T108350867

184 RAVENHOLT AND POLLIN receptcr~ ~_zn.,al the presence of nicotine a wide range of physical reactions occur. 2:~-~es occur in heart rate and skin temperature, blood pressure rises, .:e__~_~heral blood circulation slows, changes occur in brain waves, and hormon~ ~ff~==ing the central nervous system are released. With inhalation of tol>a:'_=- ~ke, these r~any physiological changes cause the smoker to experier~.-e_ ~sychic effects which many find pleasurable. Cigarette smoking shows • -.--y~-i:-al substance abuse pattern: experimentation leading to regular use le~_ u= increasing dependence and escalation in the number of ciga- rettes ~m:~ daily. If there ~e_:e such a thing as a typical pattern of dependent smoking, it would be__~-- ~ the morning with the first cigarette. This sends a burst or "bolus" =f --~cotine to the brain, which produces an almost immediate feeling of euphc.-i~= ~nd satisfaction. For the rest of the day, the smoker tries to maintai= -~:-!a feeling by manipulating his or her intake of tobacco smoke, by inhalin~ ~-r~ or less deeply, taking more or fewer puffs, and smoking at differeuz ~==ervals. When more than a certain number of cigarettes are smoked, ~e toxic effects resembling nicotine overdose are experienced, such as :a~ea, light-headedness, and a marked rise in heart rate. When fewer th~ a minimum number of cigarettes are smoked, which appears for many people := :e about ten cigarettes a day, blood levels satisfactory to the depende~: _--~ker cannot be maintained (3). What at f~r-== appears to be a casual, unordered routine, in short, turns out to be he: :~ual at all, but a controlled behavior. Nicotine ~ affect the body in different ways. In stressful situations, it can act a~ ~ anti-anxiety agent like a tranquilizlng drug, while in serene situatioca, !~ can act as a stimulant llke the amphetamines. Some effects of smok~r~_- are psychological, and some people enjoy handling cigarettes and fussing w!:z ~a=ches and ash trays; others smoke out of habit, sometimes un- aware ev~: :~:a: :hey are smoking. With no :::e: drug do people so busy themselves with administering it as they do i: ~:e case of cigarettes. In an 18-hour waking day, a two-pack-a- day smoker ~:euds about four hours with a cigarette in mouth, hand or ash tray, tal~ ~:cu= 400 puffs for the day, and inhales up to 1,000 m{lligrams of tar. To the qu~=:.-u often asked by the young - "How many cigarettes can a person smoke bef:r~ ~e.=oming addicted?" - there is no simple numerical answer. Tobacco ~::::::=~, llke addiction to marijuana, cocaine or heroin, is a complex ~ey$::ic~ic and psychologic process, with the intensity of addiction and depe=~::e varying greatly, and with the numbers of cigarettes required to reach ~ :~ level of addiction modified by the tobacco content of the cigarettes, e~:ent of inhalation, filtration, length of butts, etc. Many parameters ~-~= ~ used to measure the extent of addiction to nicotine, e.g., ti~e lapse f:.-~ awakening until llgh~ing first cigarette, or szoking when A rough inlet :f a person's current dependence upon tobacco is the au~ount of tobacco ~:r:~:: used daily. This measure also provides an index of toxin exposure :: ~:e individual. A meaningful index to the extent of life-long dependence -:.:u :~bacco is provided by lifetime consumption, estimated by T10~,350,868

185 charting the average number of cigarettes, pipefuls and/or cigars smoked per day per year of life and converting the area under the curve to the number of units smoked. The configuration of the charted tobacco consumption indi- cates tobacco dependence trends (4). EPIDEMIOLOCY OF ADOLESCENT S~OKINC Intensely curious about diverse life styles, adolescents are powerfully inclined towards emulation of their elders and peers. If parents or older siblings smoke, the adolescent is particularly likely to do so, because of the intimate example and the ready availability of smoking materials in the home. For most youths however, especially those from non-smoking homes, the initial source of tobacco and encouragement to smoke comes from their peers. Peer leadership and pressure for or against smoking continues to be a powerful determinant of smoking behavior throughout the adolescent and young adult years. The proportion of youths smoking increases rapidly during adolescent years, with roughly equal increments during junior and senior high school. By the time they are seniors, more than 4/5ths of U.S. youth have experimented with cigarette smoking, and close to one-third of those who experimented have become daily users (5). As shown in Figure I, the proportion of youths using cigarettes daily exceeds the daily use of all other drugs combined. Not only is cigarette smoking itself a most serious form of drug abuse, but because of its legal status and the general availabillty of cigarettes, cigarette smoking serves as a "gateway" to the use of other psychoactive drugs. Cigarette smokers are math more likely to use marijuana, alcohol, cocaine/halluclnogens/heroin than non-smokers. The dynamics of smoking habit formation vary greatly by time, place, and person: in the pastp urban males traditionally led the adolescent procession to smoking and rural females lagged substantially behind. A comprehensive study of entering students to the University of Washington in 1965 revealed "smoking hot spots" in a number of Washington State communities generated by increased transmission of smoking behavior in combined junior/senlor high schools and by residence in private boarding schools (6). Further studies of smoking habit formation among nursing students adduced evidence that the foremost determinant of rapid acquisition of the smoking habit upon entry to the University was the indiscriminate mixing of smokers and non-smokers when assigning dormitory roommates. Cigarette smoking behavior is in m~ny ways just as communicable as tuber- culosis - greatly £acilitated by crowding and prolonged intimate contact between smokers and non-smokers - and preventable by many of the same time- honored methods employed in the conquest of tuberculosis: case-findin~ surveys, early diagnosis, segregation of "infected" from "non-infected" p~r~on~, avb-~dan6e~o£ crowdlng, treatment of those infected, and preventive health education. Tobacco dependence is n~ a much mDre lethal condition than tuberculosis infection, and deserves far ~ore intense~ comprehensive and determined preventive action. It is sobering to reflect that adolescents who become addicted to cigarettes and smoke a package per day during the T108350869

186 balance of their lives, will thereby increase their chances of dying of lung cancer ten-fold (from 1% to 10%) and approximately double their chances of dying during m~ddle age from any cause (7). FIGURE i. THIRTY-DAY PREVALENCE OF DALLY USE Eleven Types of Drugs, Class of 1982 30 I0 21.1 From: Johnston LD, Bachman JG, O'Malley PM (5). A powerful inverse relationship exists between cigarette smoking and academic achleve=ent (Figure 2). Comprehensive studies of smoking and academic achievement in Seattle high schools during the 1960's demonstrated this phenomenon (5). When the proportion of 12th grade students then or TI08350870

TOBACCO ADDICTION 187 previously regular cigarette smokers was plotted against grouped cumulative grade point averages, a straight llne negative correlation was evident for boys, and so=ewhat less for girls. Three-fourths of failing male students were regular cigarette s=okers, and al=ost none of the top students. ~nile it is unlikely that it was smoking, per se, that caused =ost of the poor scholastic performance by s=okers, it may have contributed thereto. Students who smoke heavily may study less effectively because of frequent interruptions of their concentration, because of high blood levels of carbon monoxide, and because of increased respiratory illness. FIGURE 2. SMOKING AND ACADEMIC PERFORMANCE Percent of senior high school students who are or have been regular smokers by sex and grouped grade point average. (Numbers in ( ) indicate number of students in each grade point average group.) • O 100 75 50 25 0 <I .50 (79) ~56) boys 16~)~"~ " \l,ool oirls ~(9_0) ~ ~ ~ (2) ~ (34) 1.50-1.99 2.~-2.49 2.50-2.99 3.00-3.49 3.50-3.99 4.00 Grouped Grade Point Averages Data are from a representative sample of senior students in Garfield, Hale, Rainier Beach, Roosevelt and Sealth High Schools. From: Johnson WE, Ravenholt RT, Haroldson ~, Perrin EB (6). T108350871

188 I~h-HOLT &.~ POLLIN DISCUSSION Addictive substance abuse - tobacco, alcohol, heroin, cocaine, marijuana, stimulants, hypnotics, and hallucinogens- is now the foremost preventable cause of death in the U.S., causing approximately 630,000 deaths annually, nearly one-third of all deaths from all causes. Foremost ~mong addiction killers is tobacco, now causing approximately 500,000 deaths per annum and accounting for 80% of addiction deaths and 25% of all deaths from all causes (8). The 2 millionth death from lung cancer in the U.S. occurred in 1983, and the total number of excess deaths during this century attributable to tobacco probably exceeds I0 milllon. The fundamental reason for this carnage is that tobacco combines a powerful addictive substance (nicotine) with a powerful mutagen (polonium 210) (9). Inhalation of tobacco smoke results in the absorption of nicotine and mutagens into the pulmonary circulation and their distribution via the systemic circulation to every tissue and cell, causing damage to cellular genetic structures, deviation of cellular charac- teristics, and accelerated occurrence of cancers, degenerative cardiovas- cular diseases, and degenerative diseases of every kind (10,11). Because such pathologic changes occur gradually during many decades, young people are inadequately aware of the mortal price they may ultimately pay for what initially seems a harmless and pleasurable indulgence. Therein lies the challenge to contemporary epidemlology and public health; to measure more adequately the broad spectrum of diseases and deaths caused by tobacco, and to educate youth to the hazards of tobacco so that they will recoil from its use - as they would from a hot stove or a known source of ionizing radiation. US Dept. of Agriculture. World Tobacco Situation, Foreign Agriculture Circular: Tobacco. Washington, D.C.: Department of Agriculture, 1983 July. Ray 0. Nicotine: drugs, society and human behavior. Louis: CV Mosby Co., 1983. 3rd ed. St. Pollin W. Why people smoke cigarettes. (From a presentation to the U.S. Congress, 1982 March 16.) Public Health Service Publication No. 83-50195. 4. Ravenholt RT. Charting lifetime smoking experience. World Health Forum 1982; 3: 104. -5"--Jokns-ton-LD~chma~~--O~R~[ley PM. Student drug use, attitudes and beliefs. National trends, 1975-1982. National Institute on Drug Abuse and the Institute for Social Research, University of Michigan, 1982. Johnson ~, Ravenholt RT, Haroldson W, Perrin EB. So~e relationshiPS of smoking and teenagers' achievements. Washington Education 1965 November. T10.8350872