PART 1 Current Information on the Health Consequences of Smoking Introduction I n January 1964, an Advisory Committee appointed by the Surgeon General of the Public Health Service issued its report (16) on the relationship between smoking and health.* The conclusions of that &nmittee were summed up in the sentence: uCigarette smoking is a health hazard of sutllcient importance in the United States to warrant appropriate remedial action." In the 3% years since the publication of that report, an unprece- dented amount of pertinent research has been completed, continued, or initiated in this muntry and abroad under the sponsorship of governments, universities, industry groups, and other entities. This research has been reviewed and no evidence has been revealed which brings into question the conclusions of the 1964 report. On the con- trary, the research studies published since 1964 have strengthened those conclusions and have extended in some important respects our knowledge of the health consequences of smoking. The present state of knowledge of these health consequencB8 can, in the judgment wf the Public Health Service, be summarized as follows : 1. Cigar&a smokers have substantially higher rates of death and disability than their nonsmoking counterparts in the popu- lation. This means that cigarette smokers tend to die &, earlier ages and experience more days of disability than comparable nonsmokers. 2. A substantial portion of earlier deaths and excess disability would not have occurred if those affected had never smoked. 3. If it were not for cigarette smoking, practically none of the earlier deaths from lung cancer would have occurred; nor a sub- stantial portion of the earlier deaths from chronic bronchopul- monary diseases (commonly diagnosed as chronic bronchitis or pulmonary emphysenra or both) ; nor a portion of the earlier deaths of cardiovascular origin. Excess disability from chronic pulmonary and cardiovascular disectsss would also be less. o ?????o? and Health. Report of the Advisory Cmnmittee to the Surgeon ~~~1 of the Public Health Service." It la frequently r&err& to in t& mann- SeriPt a8 "the SW General'8 1964 Report." 3 4. Cessation or appreciable reduction of cigars& smoking could delay or avert a substantial portion of deaths which occur from lung cancer, a substantial portion of the earlier deaths and excess disability from chronic bronchopulmonary diseases, and a portion of the earlier deaths and excess disability of cardiom- cular origin. NATURE OF RECENT RESEARCH FINDINGS Since the Surgeon General's Report was published in January 1964, there has been a proliferation of additional studies and reports on smoking research. In the 12 years preceding that report, some 3,000 articles were published reporting research; since 1964, there have been more than 2,000 additional studies. These studies have helped to clarify the role that age plays in the relationship of smoking to health ; the similarities and diiIerences in the ways in which men and women are affected by smoking; and the influences and effects of stopping smoking, particularly in the case of lung cancer where there is significant data to show thst sharp redue tions in lung cancer deaths follow closely reductions. in cigarette smoking. The studies also suggest the importance of a variety of measures of exposure ; add substantial new information on the magni- tude of the morbidity problem associated with smoking; and provide more adequate data upon which to base estimates of the magnitude of the mortality problem. Historically, concern about the effects of smoking began with ob- servations of the extremely high frequency with which lung cancer patients were identified as cigarette smokers. These observations took on a fuller meaning with the first publication of the prospective studies in 1954 when higher overall death rates among cigarette smokers were identified. The rates were found to exceed the difference that could be accounted for by lung cancer alone. Until that time, the possibility remained that although more cigarette smokers appeared to suffer from lung cancer, if there were no significant excess overall mortality, some other cause or causes of mortality would have had to be underrepresented among cigarette smokers. The Surgeon General's 1964 Report concluded that cigarette smokers do have higher death rates than their nonsmoking counterparts. Thii has changed the emphasis of the present problem away from the ques- tion "does cigarette smoking cause diiP' to the more precise questions of : 1. How much mortality and excess disability are associated with smoking? 4 2. How much of this early mortality and excess disability would not have occurred if people had not taken up cigareti smoking? 3. How much of this early mortality and excess disability could be averted by the cessation or reduction of cigarette SlllOkillg? 4. What are the biomechanisms whereby these effects take place and what are the critical factors in these mechanisms? To answer these questions one must not only study the details of the relationship of ovdl mortality with cigarette smoking, one must also turn to the specific causes of death and disability and to other kinds of evidence. The research carried on since 1964 is of three principal varieties : Epidemiological studies, especially those which involve surveys of large portions of the population ; a health survey which has revealed new information about the relation between smoking and illness; and a vast amount of experimental, clinical, pathological, and behavioral research which adds to the understanding of the precise ways in which smoking affects the body, plus other closely related or peripheral information. In the area of morbidity or `illness, the primary addition to our knowledge is from "Cigarette Smoking and Health Characteristics," a report (16) of the National Center for Health Statistics on the frequency of illness among smokers and nonsmokers in a large proba- bility sample of the U.S. population. Regarding epidemiologioal data, new reports from four of the major population studies have been published since 1964 : 1. The Dorn study of smoking and mortality among U.S. vet- erans (18). 2. Hammond's study on smoking in relation to the death rates of 1 million men and women in 25 States (11). 3. The Doll and Hill study on the mortality of British physi- cians in relation to smoking (8,&M). 4. A Canadian Smoking and Health Study of Canadian pen- sioners, including veterans and dependents (1). The principal features of the additional data provided by these four studies are.: (1) The extension of the time period of followup, (2) the additional data available for specific age groups among men, and (3) the inclusion of substantial data on women. In all, the pro- spective study reports now available are based on more than 108,000 dwths, an increase of about 43,000 deaths over the 65,023 summarixed in the 1964 report. About 19,960 of these additional deaths were among women. THE NATURE OF THIS REPORT This report which provides a summary of current information on the health consequences of smoking, is based on the review of the research reports which have become available since the study of the Surgeon General's Advisory Committee was releasexl. Public Health Service staff members consulted the literature and requested additional infor- mation or interpretations of the published data from the research scientists when needed. During this review a complete bibliography, containing some 5,766 citations, was compiled; it is .now in manu- script form and will be published shortly (19). The advice and comments of experts within the Public Health Serv- ice, particularly the Bureau of Disease Prevention and Environmental Control and the National Institutes of Health, as well as of specialists outside the Public Health Service, were solicited especially on matters involving judgment and evaluation. The general criteria used by the Surgeon General's Committee have been followed. First, epidemiological data were evaluated to determine whether an association exists. In judging the significance of the as- sociation, its consistency, strength, specificity, temporal relationship, and coherence were utilized. The convergence of evidence from animal experiments, clinical and autopsy studies, and population studies re- mains the essential basis for evaluation of the significance of the associations identified. This report presents, under the following headings, the major find- ings of research studies published in the past 3 to 4 years : 1. Smoking and Overall Mortality. 2. Smoking and Overall Morbidity. 3. Smoking and Cardiovascular Diseases. 4. Smoking and Chronic Bronchopulmonary Diseases (Non-neo- plastic). 5. Smoking and Cancer. 6. Other Conditions and Research Areas. Each of these sections is introduced by pertinent conclusions from the Surgeon General's 1964 Report, which are followed by discussion and conclusions of the present study. Smoking and Overall Mortality CONGLIJNONS OF TEE SURGEON G ENElrAL's 1934 RsPoRT CIGARETTE smoking is associated with a 70-percent increase in the age-spe.cXc death rates of males, and to a lesser extent with in- creased death rates of females. The total number of excess deaths causally related to cigarette smo -II0 in the U.S. po ulation cannot be accurately estimated. In view of t continuing 811 if mounting evi- dence from many sources, it is the judgment of the Committ.sc that cigarette smoking contributes substantially to mortality from certain specific diseases and to the overall death rate. hi In general, the greater the number of cigarettes smoked daily, the % her the death rate. For men who smoke fewer than 10 cigarettes a ay, according to the seven all causes is about 40 percent tl respective studies, the death rata from who smoke from 10 to 19 ci `gher than for nonsmokers. For those %T rettes a day, it is about 70 than for nonsmokers ; for t ose who smoke 20 to 39 a cir rcent higher hi y, 90 cent her; and for those who smoke 40 or more, it is 120 percent &! KY? igarette `gher. smokers who stopped smoking before enrolling in the. seven studies have a death rate about 40 percent higher than non- smokers, as against `70 percent higher for current cigarette smokers. Menwhobegansmo' 9 death rata than those w before age 20 have a substantially higher o began after age 25. Compared with non- smokers, the mortality risk of cigarette smokers, after adjustments for differences in age, increases with duration of smoking (number of years), and is higher in those who stopped after age 55 than for those who stopped at an earlier age. In two studies which recorded the degree of inhalation, the mortality ratio for a given amount of smoking w&s greater for inhalers than for noninhalers. The ratio of death rates of smokers to that of nonsmokers is hi est at the earlier ages (44-50) %I with increasing age. represented in these studies, and dec * es Possible relationships of death rates to other forms of tobacco use were also investi ted than 5 cigars a r * * *. The death rates for men smoking less ay are about the same as for nonsmokers For men Smoking more than 5 ci T rs daily death rates are slightly higher. There is some indication t at these higher death rates occur rimaril in men who have bean smoking more than 30 years and w o i inhd the smoke to some d if at all hi her than or nonsmokers, even for meu who smoke 10 or ifll Y . The death rates for pipe smokers are little more pipe 30 years. s a day and for men who have smoked pipes more than 271-224o-67-2 7 CURRENT INFORMATION, 1967 The primary addition to knowledge in the nreas of smoking and overall mortality comes from the four major population studies. Ad- ditional periods of followup have provided a broader base from which it .becomes possible to estimate the excB8s deaths related ,to cigarette smoking in the U.S. population and from which firmer conclusions may be drawn as to the role of various exposure factors in the associa- tions found. The contributions since 1964 of each of the four population studies to the relation of smoking and overall mortality, as summarized by the authors, are set forth below. &tTJlY OF U.S. VETERANS (An 854 year followup of 293,658 persons holding U.S. Government life insurance polides. Commonly referred to as the Dom Study after the late Dr. Harold F. Dom. The most recent report is by Kahn (IS).) "* * * the increased mortality risk associated with cigarette smok- ing was found to be higher in the more recent calendar time period than in khe initial years of the study. "* * * mortality ratios of current cigarette smokers compared with those who have never smoked are 1.7 for death from all causes, 10.9 for lung cancer, 12.2 for ssnphysema without bronchitis, and 1.6 for coronary heart disease. Paralysis agitans was the only cause of death associated with significantly lower mortality for smokers than for nonsmokers. "For all categories of current smokers, risk was related to amount smoked. The risk for cigarette smokers was much greater than that for pipe or cigar smokers. Current smokers of cigarettes, cigars, or pipes experienced a mortality risk significantly greater than that for non- smokers if they smoked more than four pipes or four cigars daily or more than an occasional cigarette. "There was a positive relationship between duration of cigarette smoking and mortality risk from all causes of death for at least some classifications of smokers, "* * * probabilities of death for ex-smokers of cigarettes revealed a downward trend in risk as duration of time discontinued increased, when other variables--age began smoking, amount smoked, and cur- rent age-were controlled * * *. The data can be regarded as evidence against the constitutional hypothesis." Calculations are presented to note that observations made during the study suggest the possibility that data from respondents (those who answered the smoking questionnaire) may in fact underestimate 8 the risk associated with smoking. The Surgeon General's 1964 Report, had considered the possibility that di&ences between respondents aud nonrespondents to the questionnaire might have introduced a bias and had attempted to calculate a maximum estimate of that bias. STUDP OF MEN AND WOIKICN IN 25 STATIW (This report is baaed on 3,764,671 Wrson-years of experience and 43,!2!21 deatha occurring among l,OQ3,ZZQ eubject8-M ,668 men and S.3,671 women-between the ages of 35 and 34 from October 1.1969. to Feb- ruary 16,lQ60, when they enrolled in a proepe&ve study and answered detailed queetionnairea including questions on their smoking habits. Hammond. (II) .) "Death rates of both men and women were higher. among subjects with a history of cigarette smoking than among those who never smoked regularly. "Death rates of current cigarette smokers increased with number of cigarettes smoked per day and degree of inhalation. "Death rates were higher among current cigarette smokers starting the habit at a young age than among those starting the habit later in life. Among both men and women, the difference between the death rates of cigarette smokers and nonsmokers increased with age. "Among men, ,the death rates for ex-cigarette smokers were lower than for men currently smoking cigarettes when they enrolled in the study. Death rates of ex-cigarette smokers decreased with the length of time since they last smoked cigarettes. "* * * Total death rates and death rates from most of the common diseases occurring in both sexes were higher in men than women, were higher in men who never smoked regularly than in women who never smoked regularly, and were far higher in men with a history of cigare6t.e smoking than in women with a history of regular cigarette smoking. "The difference ,between the death rates of subjects with a history of cigarette smoking and subjects who never smoked regularly was far greater among men than women. Female cigarette smokers (as a group) have been far less exposed to cigarette smoke than male cigarette smokers of the same ages, as judged by number of cigarettes smoked per day, degree of inhalation, and the number of years they have smoked. Many female cigarette smokers smoke only a few cigarettes a day, do not inhale, and have been smoking for only a few years; their death rates are about the same as the death rat.4~ of women who never smoked regularly." STUDY OF BRITISH P~~~CIANS (The mortality of nearly 41,000 men and women in the medical profes- sion in the TJnH.4 Kinflom has been followed for I.2 yeara During the 9 tit 10 yeara 4,697 of the men and s&B ai the women died. These deaths were analyzed in relation to smoking habits reported by doctors in reply to a questionnaire sent to them in 196l-both sexes-au d again in lOti?, men, and 1090, women Doll and Hill (8, 9).) `;* * * An association with smoking is found, in differing degrees, in men for seven causes of death ,[which accounted for 39 percent of the death rate]-namely, cancer of the lung, cancers of the upper respiratory and digestive tracts, chronic bronchitis, pulmonary tuber- culosis, coronary disease without hypertension, peptic ulcer, and cirrhosis of the liver and alcoholism. No association is found with the remaining 61 percent of the death rate, and this includes such major causes as other forms of cancer, cerebrovascular accidents, hyperten- sion, myocardial degeneration, suicide, and accidents. "In women, the few deaths at present available show an association only between smoking and cancer of the lung. U* * * If the excess deaths in smokers under the age of 65 years from (a) cancer of the lung, (6) chronic bronchitis and emphysema, (c) coronary thrombosis without hypertension be taken as attributable to their cigarette smoking, then the total mortality from all causes at ages 45-64 years is increased thereby by approximately 50 percent." The report states: "One of t.he striking characteristics of British mortality in the last half-century has been the lack of improvement in the death rate of. men in middle life. In cigarette smoking may lie one prominent cause." STUPY OF CANADUN PENEIONPXS (The purpose d the study was to investigate the relationships between residence, mtion, smoking ~habits, and mortality from chronic Waaes .@cularly lung cancer. It was initiated by a que&ionndre which was sent to Canadian veteran pension recipients doriug the period September 1955 through June 196g. Returns from ?S#OO men, and 14,000 women, mostly widows, were analped. The men were mainly World War I and World War II vet- erans, but some Boer War and Korean War veterans, as well aa some non-veteran pension recipients were included. The age of most of the men at the beginning of the study ranged from 30 to 90 years and the distribution was characterised by the ages of men eligible for service in the two World Wars. For each respondent dying between July 1. 1956, and June 30, 196% the cause oi death was related to information on his questionnaire atint age, history of emoking habits, residence and occopation. Among the respondents during the 6 years of followup there were Q&l deaths of males, and l,?fM deaths of females which were analyaed (11.1 "Current cigarette smokers had a death rate for overall mortality 54 percent higher than that of nonsmokers * * * Ex-cigarette smok- 10 ers had a comparatively lower rate, which was still 36 percent above the rate for nonsmokers * * * Men smoking combinations of ciga- rettes plus cigars and/or pipe also had elevated death rates for overall mortality, but these were not elevated to the same extent as those of men smoking only cigarettes. "The death rates for overall mortality of pipe smokers and cigar smokers were not appreciably different from those of nonsmokers. "For cigarette smokers as compared to nonsmokers, overall mor- tality ratios were elevatd after 5 years of smoking at any time in their life and remained elevated as long as they amtinued to smoke cigarettes. "Male current cigarette smokers who inhaled had a death rate for overall mortality 52 percent higher than that of those who did not inhale. "An urban/rural comparison was made between males of equivalent cigarette smoking habits and nonsmokers. It was found that the death rate for overall mortality of urban dwellem @rsons with a history of 5 years or more of city residence) was 12 percent higher than that for rural dwellers of comparable smoking habits. "Respondents were classified into occupational groups based on their history of occupation. No evidence was found in this study of clear-cut associations between cause of death and occupation. Further, occupation did not appear to modiQ the e&ab%shed association of cigarette smokers with death rates in ~XCCZB of those of nonsmokers." SOME GENERAL CONSIDERATIONS The problem of how best to measure the relationship between smok- ing and mortality has been discussed in the Surgeon General's 1964 Report as well as in some of the proqective study reports. &4 the amount of data available increases, the person-v of observations in the many population subgroups that are worth examining increases so that stable rates may be computed and compared. A brief discussion of three measures of comparison available and their utility seems desirable as confusion frequently arises over these measmes. 1. Mortality Ratios: Obtained by dividing the death rate for a classification of smokers by the death rate of a comparable group of nonsmokers. 2. Differences in Mortality Rates: Obtained by subtracting from the death rate for smokers, the death rate of a comparable group of nonsmokers. 3. Excess Deaths: Obtained by s&ra&ng from the number of deaths occurring in a group of smokers, the number of deaths 11 which would have occurred if that group of smbkers had ex- perienced the same mortality rates as a comparable group of nonsmokers. In the example which follows this has been reported as a percentage of all deaths in the appropriate age group. Table 1 presents in summary form all three measures for five age groups of men from both the U.S. veterans study and Hammond's study and for the same age groups of women from the latter study. The statistics were derived from the cited publications to provide reasonable comparability and may vary slightly from the Bgures combined in other ways. Also it should be noted that the age groups are not defined identically and the experience reported covers some- what different time periods. The smoking group analyzed is "cur- rent cigarette smokers," i.e., those who were smoking at the time of enrollment into the study, and the comparison group is %ever smoked regularly," i.e., those who had never been regular smokers of any form of tobacco. The nmnber of deaths in each age-sex group is given to indicate the relative stability of the figures in that column. The data in the veterans study are largely concentrated in age groups X5-64 and 65-74. In Hammond's study, age group 3544 is less stable than the succesding groups both for men and for women. 1. Morta&y Rat&x-For men, these are at their highest in age group 45-54, diminishing in each subsequent decade. In both studies mortality ratios appear to be somewhat lower in the preceding decade 35-44. However, with the smaller numbers of cases available in that age group, it may be that selective factors contribute to the fluding. For women the mortality ratios are much smaller than for men, although the same pattern is suggested. In general, a mortality ratio has been considered to reflect the degree to which a classiflca- tion variable identifies or may account for variations in death rates. As such, it is a measure of relative risk which indicates the importance of that variable relative to uncontrolled variables-an indicator of phi?&zl biologicaz 8ifpG-8. 2. Difftmncerr ila Mort&y Ratea.-These increase consistently with increasing age in all three study groups, except for the oldest age group in women where there is practically no difference in the rates for smokers and nonsmokers. Differences between smokers' rates and nonsmokers' rates are much smaller for women than for men, as are the death rates themselves for men and women classified simi- larly with respect to smoking. This measure reflects the added proba- bility of death in a l-year period for the smoker over that for the non- smoker. As such it is a measure of peraon.& health &g&@n.ct?, a means for the individual to estimate the added risk to which he is exposed. 12 TABLE l.-Compariam of 3 nmmm+s of relatkhip betwem cigarette smoking and ovem?~ death rates by age and 8ex a8 derived from W major proepective 8tu4698 (11,13)' U.S. VETNRANB: &N Total deaths __________ _____, Death rates per 100,000: Never smoked regularly _______, Current cigarette smokers- _ _ _ _. Mortality ratio * __________________. Difference in death rates per 100,000 * ----_-_--------__-----. Exceaa deaths as percentage of totsl'-------------------__---_. Total deaths __________ ___ __. Death rates per 100,000: Never smoked regularly _______. Current cigarette smokers- _ _ _ _ Mortality ratio f _______ __ ___ _ ___ __. DiSerence in death rates per 100,000'~~~____~~~~~________~~. Excess deaths as percentage of total'----______-_-_______-----. HAMMOND WOMEN Total deaths ___________ _ ___. Death rates per 100,006: Never smoked regularly-- __ _ ___ Current cigarette smokers-----. Mortality ratio * ____________ ______. DSerence in death rates per 100,000~ _-_____________________ Excess deaths aa percentage of total'-----_--_---------_-----. 383 366 13,840 17,550 1,932 127 264 1, fm 2# 411 6,214 232 723 1,819 4032 s, 471 1. 83 2 76 L 72 L 67 1.36 105 464 763 1,621 5257 33 43 21 17 8 631 5,297 210 406 397 925 1. 89 228 187 519 33 33 4125 39968 9168 7, a63 4788 9,674 1. 51 123 1, 620 1,811 13 4 727 105 186 1. 13 21 5 Kg= 304 334 L26 30 9 3,915 698 333 1. 20 5,115 1,913 s= L 17 146 4 316 2 r,l@ 5,914 5,846 .99 68 ----- 48-k 65-74 76-84 --- 3. Eacea Deaths aa a Percentqe of Total Death-h with mor- tality ratios, this statistic appears to be highest in the age group 45-54 where it reaches 43 percent in one group of men and 38 percent in the other. Hammond's data by B-year age groups show the highest rate at ages 45-49, where it is 44 percent. Reviewing both study groups it appears that for men between the ages of 35 aud 60 approximately one-third of all deaths that occur are excess deaths in the sense that they would not have occurred as early as they did if cigarette smokers had the same death rates as the nonsmoking group. For women, the percentage is much lower, reaching a peak of 9 percent of all deaths in age group 45-54. It should be noted that this measure not only de; pends on the differences in death rates between the smokers and the nonsmokers, but also on the proportion of smokers in the group. Thus, even with a large difference in rates between smokers and nonsmokers, a population with very few smokers would have very few excess deaths. This measure is therefore an indicator of the p&Z& hea.Zth Bisnificymrce of the differences found since it measures the number of people affected and therefore the magnitude of the problem for society as a whole. Once the magnitude of the excessis identified the problem becomes one of determining (1) how much of the excess would not have oc- curred if it had not been for cigarette smoking and (2) how much would have occurred anyhow. It should be noted that much of the ex- cess has already been identified as belonging in the first category. Of the remainder, little of the excess has been clearly identified as belong- ing in the second category-t.hat is, not caused by smoking. With most of that remainder there is uncertainty as to the category in which it belongs. Studies involving smoking, whether epidemiological or ,behavioral, have been concerned with measures of exposure to tobacco smoke. For the most part, these studies have been restricted principally to the in- dex of number of cigarettes smoked over a spe&ed period of time, usually an "average day." The heavy reliance on numbers of cigarettes alone as a measure has produced important findings but it has possi- bly obscured others. The new reports on the prospective studies have provided a substantial amount of data to support the concept that many elements should entsr into an overall measure of exposure. Such factors as age at beginning smoking, duration of smoking, and inhala- tion have all shown some independent contributions to the overall effect, along with numbers of cigarettes. A recent report (16) has at- tempted to develop a more adequate measure of exposure in which various individual components of dosage would be combined to form compositescores. 14 A dosage score was developed as a function of the average number of cigarettes smoked per day, the "tar" (smoke solids minus moisture) rating of the brand of cigarette smoked, and the portion of the ciga- rette actually smoked. In addition, questions on both depth aud fre- quency of inhalation were developed. Normative data have been ob- tained from a national survey sample of smokers. In general, although the various measures reflecting exposure are interrelated, there are many individuals with high exposure on one measure but low ex- posure on another. Furthermore, there are sysbma& differences in some of these measures of dosage between men and women, between heavy and light smokers (by the usual criterion of numbers of ciga- rettes) , etc. The existence of a dose-response relationship between ex- posure to cigarette smoke and the risks most clearly associated with cigarette smoking is now generally accepted. Wynder .and 4Hoffmann (90) have shown in laboratory experiments with animals that the tumorigenicity of cigarette smoke can be reduced by alteration in the cigarette which reduces the "tar" and nico- tine content. They use the term "indicator" for %r" and nicotine con- tent (the two measures tend to be used jointly since when one is high the other tends to be high unless the nicotine has been removed in processing), or other measures which reflect this Qpe of relationship, lacking the identification of specific agents which are responsible for the effect. Bock, Moore, and Clark (8) have independently shown a similar variation in carcinogenic activity of tobacco Yati' obtained from different types of cigarettes. The preponderance of scientific evidence strongly suggests that the "tar" and nicotine content of cigarette smoke is a meaningful factor in the measurement of dosage. The cessation of smoking is, of course, an extreme example of the reduction of dosage. Data from the prospective studies show a reduc- tion in both overall mortality axid mortality from specific diseases among those who have stopped smoking when compared with those persons who continue to smoke. This finding has been somewhat ob- scured by the fact that ill health is a frequent cause of giving up smoking so that death rates and disability rates for ex-smokers as a group tend to be high for an initial period of time following cessation. In this connection, the Study of British Physicians shows that among the total group of physicians in the &udy (smokers, ex- smokers, and those who never smoked, combined) there was a reduc- tion in the standardized lung cancer death rate from 0.69 per 1,000 in the first 5 years of the study (1951-56) to 0.64 per 1,000 in the sec- ond 5 years of the study (1956-61). This reduction occurred during 16 the time when there was also a substantial drop in cigarette smoking among physicians in general, and during the time that lung cancer rates were risii in the male population of Great Britain. This situa- tion is not unlike that of a controlled cessation experiment in which the effect of giving up smoking is judged by the mortality results in an entire population in which the giving up of smoking is common as against another population in which it is not common. A more recent report by Doll (7) suggests that this trend is becoming more marked as the rate of smoking among British physicians decreases and the length of the cessation period increases. These Endings are shown in Table 2, which has been derived from Doll's report (7). The lung cancer death rate among men in England and Wales increased from 1.49 per 1,000 in the period 1954-57 to 1.86 per 1,000 in the period 1962-64, a rise of 25 percent. At the same time, the lung cancer death rate for British physicians dropped from 1.09 per 1,000 in &he first period to 0.76 per 1,000 in the second period, a reduction of 30 percent. This reduction in death rates from lung can- cer among all physicians is larger than would have been anticipated from examining only the experience of those physiciaus who had stopped smoking before the study began and indicates that the ex- perience of ex-smokers in prospective studies probably understates the benefits of giving up smoking. With these fmdings the case for cigarette smoking as the principal cause of lung cancer is overwhelming. The reduction of rates experi- enced in ex-smokers as compared with continuing smokers is clearly shown in the case of lung cancer to be a reflection of a sign&ant change in risk. Since the concern that selective bias might have ac- counted for the earlier findings has been contraindicated, a stronger case csn now be made for interpreting reduced rates of overall mor- tality for those who give up smoking as also reflecting a direct ahera- tion of risk compared to those who continue to smoke. There are no adequate data to evaluate the benefit of reductions in exposure that are more mode& than those achieved by complete ceesa- tion, although it seems reasonable to assume that a substantial reduc- tion in exposure is likely to be accompanied by some reduction in risk relative to those who do not reduce their exposure. 16 TABLE 2 .--changes sicians in the lung canw death rate in male British phy- (age 36-84) compared wit% chunges in the &es f&r talc ma& popu.?ation of England and Wales for 3 time G&sv& bebxcn 1964 and 1964 (7) The period 1954to 1957___________________________________ 1958tJo 1961------______-______________________ 1962to 1964--------____--_____________________ Percentage change: 1st to 2nd period- __ _ ___ _ __ _ ___________ _____ 2nd to 3rd period---- _____ -- ________________ lstto 3rd period---_________________________ +15 ++a% -24 -8 -30 17 Smoking and Overall Morbidity T TEIE TIME of the Surgeon General's 1964 Report there was no A information available on the overall disability associated with smoking. To investigate the relationship between smoking and mor- bidity, the National Center for Health Statistics of the Public Health Service introduced questions about cigarette smoking into its National Health Survey, beginning in July 1964. This Survey is a continuing study conducted since 1957. In carrying on this Survey, interviewers each year visit 42,000 fami- lies (selected as a probability sample of the civilian, noninstitutional population of the United States) and question them about illness, dis- ability, and days absent from work ,because of illness, as well as the nature of the illness. In the year ending in June 1965, they inquired (after all other questions about health had been asked) about the smok- ing habits of persons in the family who were 17 years of age or over. The National Health Survey is concerned with three overall meas- ures of the impact of illness. 1. Daya Loat Frcnn Work.- These are days absent from job or busi- ness because of illness or injury. They apply only to those persons who are currently employed and are therefore heavily concentrated in age groups 17-64. 2. Bed Day&-These are days when the person is su5iciently ill or disabled so as to spend all or most of the day in bed, either at home or in a hospital. All days spent as a hospital patient are included. 3. Days of Restricted Activity.-These are days when a person cuts down his usual activities for most of a day because of an illness or an injury. Days lost from work because of illness and bed days are, of course, counted as days of restricted activity. This represents the most general measure of, disability available in the United States today. Table 3 summarizes the findings in a form similar to that used for summarizing the overall mortality utilizing three measures of mor- bidity effect : Morbidity ratios, differences in rates, and excess days of disability. 19 TABLE 3.-tZlnnpa&m of 3 mexuurea of relakhip between cigar& smoking and 3 types of didiZity days bg age and 8cx as cikiuedfrom the Nat&maZ Et&h &uvcg (16) WOESLoss DAYS E&hated total days (millions)----.. Rate: ' 112 127 21 80 5s 4 Never smoked cigarettea _______ 3.4 5.6 9.8 4.5 5.3 History of cigarette smoking---- 4.4 8.5 9.8 6.5 6.9 Morbidity ratio *-- _______________ 1.3 1.5 1.0 1.4 1.3 DBerence in morbidity rhea * I--- _ _ 1.0 2.9 0 2.0 1.6 Excess days as percentage of total a- 20 28 0 18 11 RESTEICTED ACTIVITY Days Estimated t&al days (millions)----- Rate: 1 305 386 271 543 489 395 Never smoked cigarettea _______ 7.5 15.0 32.9 13.3 22.6 40.1 History of cigarette smoking- _ _ 10.6 22.9 37.9 17.8 25.3 44.8 Morbidity ratio *--- ______ _ _ _ _____ 1.4 1.5 1.2 1.3 1.1 1. 1 DiiTerence in morbidity ratea I4--- _ _ 3. 1 7.9 5.0 4.5 2.7 4.7 Excess days as percentage of total I- 23 28 8 14 5 2 BED DAYS E&ii&xl total daya (millions)----- Rate: ' 111 118 Never smoked cigarettes _______ 2.7 4.6 History of cigarette smoking- _ _ 3.9 6.9 Morbidity ratio I-- _ --_-----______ 1.4 1.5 Difference in morbidity rates * 4--- _ _ 1.2 2.3 Excess days as percentage of total I- 23 28 AhIn T 17-44 46-M Bsnlld OVW 100 13.4 13.0 .97 -0.4 -1 5.0 : : 210 168 146 5.4 8.0 15.1 6.7 9.2 15.2 1.2 1.1 1.0 1.3 1.2 0.1 10 6 0 B5aud OVW ' DiU- ln hfmbldity Ii&a-itforbidifv rate for c&art& amokem minas morblditv rota for thuw who twmbmob4d~df.w. ` Exceaa deatha among c@rette amokas (I.E., addltionnl days o[ dlmbfflty that omur among c@mtte smokem prr yew dove thorn whkb would hme ocamed li amokexa had the name rated aa thm who nevs smoked cigamttss). TUJ Is expmmmd a~ a percmm of oil diw&Uity dnys ooxmfng in that ale-m gronp. DAYS LOST FROM WORK For those with a history of cigarette smoking, classified by heaviest amount smoked, the average number of days was 7 percent higher for men and 15 percent higher for women who had smoked less than 11 cigarettes per day; 33 percent higher for men and 60 percent higher 20 for women who had smoked 11-20 cigarettes per day ; 48 percent higher for men and 79 percent higher for. women who had smoked 21-40 cigarettes per day; and 33 percent higher for men and 140 percent higher for women who had smoked more than 40 cigarettes per day. The relationships expressed by all three measures are somewhat higher among men aged 45-64 than among men aged 17-44, but lower among women aged 45-64 than among women aged 174. In the survey year, there were an estimated 399 million workdays lost in the United States because of illness. A total of `77 million days, or 19 percent, were excess workdays lost because of the higher rates which exist among persons who have ever smoked cigarettes as compared to those who never smoked. This excess loss is highest in men 45-64 where it represents 28 percent of all days lost. BED DAYS For those with a history of cigarette smokiug, class&d by heaviest amount smoked, the average number of days was 10 percent higher for men and 4 percent lower for women who had smoked less than 11 cigarettes per day ; 22 percent higher for men and 17 percent higher for women who had smoked 11-20 cigarettes per day; 22 percent high- er for men and 57 percent higher for women who had smoked 21-40 cigarettes per day; and 53 percent higher for men and 192 percent higher for women who had smoked more than 40 cigarettes per day. Relationships with smoking are higher for men than for women for all three measures except for age 17-44 in which the d.ifTereuces in mor- bidity rates between smokers and nonsmokers are about the same. For the entire population 17 years of age and older there were an estimated 853 million bed-days in the survey year. A total of 88 million of these days, or 10 percent, were "excess" days lost because of the higher rates which exist among persons who have ever smoked cigarettes as com- pared to those who never smoked. Excess days as a percentage of total bed-days is highest for men aged 45-64, where it is 23 percent. DAYS OF RESFRICTED ACTIVITY For those with a history of cigarette smoking class&d by heaviest amount smoked, the average number of days was 12 percent ,higher for men and 4 percent higher for women who had smoked leas than 11 cigarettes per day ; 32 percent higher for men and 22 percent for women who had smoked 11-20 cigarettes per day; 39 percent higher for men and 43 percent higher for women who had smoked 21-40 cigarettes per day ; and 81 percent higher for men and 146 percent higher for women who had smoked more than 40 cigan&.es per day. Again rates are higher for men than for women in all three measures except for age group 1744, in which differences in morbidity rates are higher for women. There were an estimated 2,369 million such days in the survey year; 306 million, or 13 percent, were excess days lost because of the higher rates which exist among persons who have ever smoked cigarettes as compared to those who never smoked. Excess days as a percentage total restricted activity days was highest in men aged45-64. To help evaluate these general indices of morbidity as measured by various kinds of disability days it is necessary to turn to the conditions which are reported more frequently by cigarette smokers than by non- smokers. Since these are either self-reports or reports made by a re- sponsible member of the household for others in the household, the diagnostic accuracy of the reports is obviously less than one could ob- tain from direct medical examination. Nevertheless, the bulk of the reports on chronic conditions reflect what a physician has previously told the patient or the family with regard to a diagnosis of the condition. Chronic conditions (one or more) are reported by 11 percent more of the men and 9 percent more of the women who have ever smoked cigarettes than by those who have never smoked. cigarettes. This is especially high in those who have reported their highest consumption rate to have ,been over two packs a day (32 percent higher for men and 43 percent higher for women). At the lower levels of consumption the rates reported are 21 percent and 25 percent higher for those smoking 21-40 cigarettes per day, but only 5 percent higher for men and 7 percent higher for women for those smoking 11-20 cigarettes per day and only 1 percent higher for both men and women who have never smoked more than 10 cigarettes per day. The differences are especially marked among present smokers of more than two packs per day whose rate of reporting three or more chronic conditions is 73 percent higher for men and 143 percent higher for women than for those who have never smoked cigarettes. Applying differences in prevalence rates to the entire U.S. popula- tion 17 years of age and over yields the estimate that there are approxi- mately 11 million more cases of chronic illness annually than there would be if all people had the same rate of~`sickness as those who had never smoked cigarettes. A large portion of these are accounted for by conditions classified as "chronic bronchitis and emphysema," "heart conditions," "peptic ulcers," and "sinusitis." All but the last of these have previously shown substantially higher mortality rates among cigarette smokers. Si.nusitis,`being a nonfatal condition, has not been identified in the studies of mortality previously reported. The "heart condition" relationship is most marked in the category "arteriosclero- tic heart disease including coronary disease." The age-adjusted incidence rate of acute conditions for persons who had ever smoked was 14 percent higher among men and 21 percent higher among women than the rates for "never smokers." However, particular caution must be taken in interpreting the results relating specific acute conditions to cigarette smoking because of the relatively large sampling error connected with the estimates for the several types of acute conditions. Since the National Health Survey is not a prospective study, it does not identify the rate at which various types of morbidity develop in comparable groups of smokers and nonsmokers, but reports the recent existence of such disability. Therefore, the tidings are much more significant when they support relationships previously identified than when new relationships are identified. It should not be surprising that causes of mortality which are associated with cigarette smoking have a counterpart in disease or disability associated with smoking. As the primary source of data in the United States on disability, the Survey report, being based on a national probability sample, provides a solid base for estimating the excess overall disability asso- ciated with cigarette smoking. HIGHLIGHTS OF CTJRRENT INFORMATION ON OVERALL MORTALITY AND MORBIDITY 1. The previous conclusions with respect to the association between smoking and mortality are both confirmed and strengthened by the recent reports. The added period of followup and analysis of deaths of nonrespondents as well as of respondents in the Darn Study sug- gests that the earlier reports may have understated the relationship. 2. More information is now available for specific age groups than previously. A comparison of three ways of measuring the relationship indicates that cigarette smoking is most important among men aged 45 to 54 both in terms of mortality ratios and exc855 deaths expressed as a percentage of total deaths. Nevertheless, although both of these measures decline with advancing age, the increment added to the death rate, which reflects one's personal chances of !being affected, continues to increase with age. For men between the ages of 35 and 59, the excess deaths among currant cigarette smokers account for one out of every three deaths at these ages. For women, with their lower overall exposure to cigarettes, the comparable figure is about one death out of every 14 at ages 35 to 59. 3. Women who smoke cigarettes show significantly elevated death rates over those who have never smoked regularly. The magnitude of the relationship varies with several measures of dosage. By and large the same overall relationships between smoking and mortality are observed for women as had previously been reported for men, but at a lower level. Not only are the death rates for men who have never smoked regularly higher than those for women who have never smoked 27l-3a4o-674 23 regularly, but the effect of smoking as measured either by differences in death rates or by mortality ratios is greater for men than for women. At least part of this can be accounted for by the lower ex- posure of female cigar&e smokers whether measumd by number of cigarettes, duration of smoking, or degree of inhalation. 4. Previous .&dings on the lower death rates among those who have discontinued cigarette smoking are confirmed and strengthened by the additional data reviewed. Kahn's analysis of ex-smokers in the U.S. veterans study-controlling for age at which they began smoking, amount smoked, and current age-reveals a downward trend in risk relative to those who continued to smoke as the duration of time dis- continued increases. The British physician study, in which a downward trend is reported in lung cancer death ra&s for the entire group (smokers, ex-smokers, and those who never smoked, combined) along with a very sharp reduction in cigarette smoking by the physiciaus, is the beet available example of a controlled cessation experiment with reduction of risks resulting from reduction of smoking. The Sndings of this report support the view that epidemiological data showing lower death rates among former smokers than among continuing smokers cannot ,be dismissed as due to selective bias and that the bene- fits of giving up smoking have probably been understated. 5. Cigarette smokers `have higher rates of disability than non- smokers, whether measured by days lost from work among the em- ployed population, by days spent ill in bed, or by the most general measurdays of "restricted activity" due to illness or injury. Data from the National Health Survey provide abase for estimating that in 1 year in the United States an additional 77 million man-days were lost from work, an additional 88 million man-days were spent ill in bed, and an additional 306 million man-days of restricted activity were experienced *because cigarette smokers have hiiher disability rates than nonsmokers. For men age 45 to 64,28 percent of the disability days experienced represent the excess associated with cigarette smoking. 24 Smoking and Cardiovascular Diseases CONOLUkUONS OF m &JRGlWN G- `8 1964 Rnronr Male cigarette smokers have a higher death- rate from coronary artcry disesse than nonsmoking males, but it IS not clear that the association has causal sign&ance. CURRENT INMIRMATION, 1967 Important additional epidemiological information from five pros- pective mortality studies confirms that cigarette smokers have sub- stantially higher death rates from coronary heart disesse `than do nonsmokers. This is true for `both men and women although the relationships are less marked in women. Cigarette smoking also markedly increases an individual's susceptibility to earlier death from coronary disease. In general, mortality rates increase with increasing amounts emoked. Cessation of cigarette smoking is followed by a reduction in the risk of coronary heart disease mortality relative to those who con- tinued to smoke. Epidemiological evidence indicates that there is little risk of coronary heart disease associated with cigar and/or pipe smoking. The Surgeon General's 1964 Report indicated a median mortality ratio of 1.7 for current cigarette smokers, with a range from 1.5 to 2.0. Additional evidence from the Hammond study (11) indicates that young smokers between the sges of 45 and 54 have the highest mortality ratios-three times as great for men, and twice as great for women if they smoke 10 or more cigarettes per day, as compared with non- smokers. In general, the mortality ratio shows the most marked in- creases with increasing amount smoked for the ages under 65. While the cigarette smokers older than 65 have lower mortality ratios than those under 65, the public health significance of the relationship in the older population is subetantial because of the large numbers of people over 65 who die of coronary heart disease. Studies of U.S. veterans (18)) Canadian pensioners (1)) British physicians (8,9, JO), 25