FIGURE 53.-Age-specific mortality rates by 5-year age groups for cancer of the pancreas for white males, United States, 1950-1977 SOURCE: Natxmal Cancer Institute (198). groups without significant increases in the rates of the younger age groups, as is readily apparent when age-specific death rates for white males and females are plotted on a three-dimensional graph (Figures 53 and 54). Pancreatic carcinoma is generally undetected until late in its course, due to difficulties in diagnosis and the nonspecific nature of the presenting symptoms. Metastasis occurs relatively early in the 126 -----T------ IlllIllIlIlIlIIIIIIIIIIII 1 FIGURE 54.-Age-specific mortality rates by 5-year age groups for cancer of the pancreas for white females, United States, 1950-1977 SOURCE: National Cancer Institute W8b course of this disease, contributing to the poor S-year survival rate of 2 percent (194) and a mean survival time after diagnosis of less than 6 months (187). The most common form of pancreatic cancer is adenocarcinoma. Pancreatic cancer is more common among men than among women in the United States, but the male to female ratio has been decreasing steadily from 1.6:1 during the period of 1940-1949 to 1.2:1 estimated in 1960 (27G). 127 Several epidemiological studies have `established an association between cigarette smoking and pancreatic cancer. Causal Significance of the Association Consistency, Strength, and Specificity of the Association A number of retrospective studies have examined the relationship between smoking and pancreatic cancer. In the Third National Cancer Survey (299) and in the Hawaiian Study of Five Ethnic Groups (1131, there was a significant positive relationship between smoking and pancreatic cancer. An earlier retrospective case control study of 81 cases of pancreatic cancer (3.201 found a dose-response relationship with a relative risk of 5.0 for males smoking more than two packs of cigarettes per day (Figure 551. A recent report found a positive association for both males and females who had ever smoked and cancer of the pancreas (relative risk of 1.41, but not for pipe or cigar smokers. They also reported a significant dose-response rela- tionship for females. A similar but not significant dose-response relationship was noted for males (169). Several of the large prospective investigations have reported mortality ratios of approximately 2.0 for smokers as compared with nonsmokers. These data are presented in Table 36. The dose- response relationships from four of the major prospective studies are presented in Table 37. Smokers consuming more than one pack of cigarettes per day had mortality ratios two to three times greater than those of nonsmokers. These data consistently support an association between smoking and pancreatic cancer, although the strength of the association is less than that noted for smoking and cancer of the lung, larynx, oral cavity, and esophagus. Temporal Relationship of the Association Support for the temporal relationship of the association is provided by the prospective studies that observed subjects over varying periods of time for the development of pancreatic cancer. Support for the temporality of the association is advanced by a histological study showing a greater frequency of premalignant changes in pancreatic tissue of smokers when compared with tissue of nonsmokers (1621, and by cohort analysis showing correlation between trends in smoking patterns and pancreatic cancer mortality (22, 128). 128 5, 4 - 3 - 2 - l- O- N: CASES CONTROLS 13 5; -0 e-w- NEVER SMOKED l-10 11-M 2140 41* NO. OF CIGARETTES SMOKED PER DAY FIGURE BEi.--Relative risk of pancreatic cancer in males, by number of cigarettes smoked SOURCE: Wynder (320). Coherence of the Association Dose-Response Relationship The coherence of the association is supported by the dose-response relationship noted above, although it is not as marked as those noted for smoking and other cancers. Correlation of Pancreatic Cancer Among Populations With Different Tobacco Consumption The finding of a low incidence of pancreatic cancer in special groups (e.g., Mormons and Seventh Day Adventists) with a small proportion of smokers (79, 165, 166, 211, 294) is consistent with a causal relationship. 129 TABLE 36.-Pancreatic cancer mortality ratios-prospective studies Study Size of Population Nonsmokers All Cigarette * Smokers Comments ACS 9.State Study 188000 white males Canadian 78,000 Veterans males ACS 25Stat.e Study 358,000 males 483,ooO females U.S. Veterans 290,000 males Japanese Study 122,OlXl males 143,COO females California occupations 68,OflO males Swedish Study 55,ooo males and females 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 3.1 males 1.00 2.5 females 1.50 Based on 117 microscopically proved cases 1.96 2.14 1.42 1.79 1.57 males 1.94 females 2.43 British Physicians 34,000 males 1.M) 1.60 TABLE 37.-Mortality ratios for cancer of the pancreas by amount smoked-prospective studies Study Swedish Study Population 5waJ males and females Amount Smoked per Day Comments Males Females NS 1.00 NS 1.00 l-7 1.60 l-7 2.40 8-15 3.40 8-15 2.50 15 + 5.90 15 + 3.00 British Physicians 40,ooo NS 1.00 NS 1.00 Males based 1-14 1.35 1-14 0.44 on grams of 15-24 1.42 15-24 2.66 t&WC0 25 + 2.07 25 + 1.77 per day Japanese 265,ooO NS 1.00 NS 1.00 Study males and l-19 1.42 l-19 1.M) females 20-39 1.57 20-29 1.60 40 + 0.69 30+ 1.90 U.S. Veterans 290,ooo NS 1.00 n-G&S 1-9 1.60 lG20 1.71 21-39 2.00 40 + 2.20 NOTE NS Nonsmoker 130 Correlation of Sex Differences in Pancreatic Cancer With Different Smoking Habits The declining male to female mortality ratio discussed above is consistent with the delayed initiation of cigarette smoking by women as compared to men. Two studies have performed cohort analyses of the relationship of time trends in smoking patterns among males and females and mortality rates from carcinoma of the pancreas. Bernard and Weiss (22) examined the relationship in the United States for the period of 1939 to 1969; Moolgavkar and Stevens (185) examined these relation- ships in England and Wales for the period of 1941 to 1975. Both studies found a positive association between changes in smoking habits in males and females and pancreatic cancer death rates. Smoking and Histologic Changes in the Pancreas A recently reported study (162) found evidence for premalignant changes in pancreatic tissue of smokers. The authors collected 108 specimens of pancreatic tissue. In 44 percent of the series, there were some focal acinar cell abnormalities, which the authors state were similar to atypical acinar cell nodules in carcinogen-treated animals. These findings were more common in tissue from patients with a history of smoking as compared with tissue from nonsmokers. Tissue from heavy smokers (67 to 100 pack-years) showed a 1.8 times higher incidence of such nodules than tissue from all smokers. Pancreatic Cancer and Non-Cigarette Tobacco Use The U.S. Veterans Study found an elevated risk of 1.5 for pancreatic cancer in cigar, but not pipe, smokers. Experimental Studies Dietary factors, the presence of underlying diseases, such as chronic pancreatitis and diabetes mellitus, and chemical exposures have been suggested as potential determinants for this disease (187). The pathogenic mechanisms by which tobacco smoking influences the development of pancreatic cancer are obscure. It has been suggested that a carcinogen derived from tobacco smoke (either directly or after metabolism by the liver) ts excreted into the bile (321). It is then refluxed into pancreatic ducts and induces cancer. One group of investigators (145) has reported that nicotine inhibits pancreatic bicarbonate secretion in the dog by direct action on the organ. This has led to speculation that inhibition of duct cell secretion of bicarbonate could lead to intracellular pH changes and subsequently play a role in carcinogenesis. It has also been suggested that a protease-antiprotease imbalance may be capable of promoting carcinogenesis. Cigarette smoke is known to affect the protease- 131 antiprotease balance in vivo and in vitro. In a study of beagle dogs smoking 12 cigarettes per day for 600 days, the authors reported significant changes in pancreatic proteases as compared with their sham-exposed controls (189). Conclusion Cigarette smoking is a contributory factor in the development of pancreatic cancer in the U.S. The term "contributory factor" by no means excludes the possibility of a causal role for smoking in cancers of this site. Stbmach Cancer It is estimated that in the United States there will be 24,200 new cases of stomach cancer and 13,800 deaths in 1982 (2). For unknown reasons, mortality rates and the number of deaths have fallen dramatically over the last 28 years. The age-adjusted mortality rate for stomach cancer has continued to decline for both males and females. Since the period of 1951-1953 through 1976-1978, the age-adjusted rate has decreased by 59 percent in males and 65 percent in females. Rates for both males and females adjusted to the 1970 population are presented in Figure 56. Figures 57 and 58 give age-specific death rates for cancer of the stomach for four separate time periods by race and sex. In 1950, cancer of the stomach was fatal to 24,257 persons; in 1977, 14,440 died from this cancer in the United States. Death rates are higher for races other than white than for whites; other males have higher death rates than any of the other color sex groups. The age-adjusted rate for other than white males was 31.16 in 1950 compared to 23.86 for white males. The corresponding rates for females were 16.05 and 13.13, respectively. By 1977, the rate for other than white males had decreased to 15.18; the corresponding rate for white males was 8.25. The age-adjusted rate for females other than white was 7.46 in 1977 compared to 3.83 for white females. These differences may represent variations in exposure to undeter- mined dietary and other environmental factors or genetic differ- ences. A limited number of epidemiological studies have examined the relationship between smoking and stomach cancer. The data are not consistent, but overall, the evidence points to a possible association between cigarette smoking and stomach cancer. Olearchyk (204) noted that alcoholism (26.7 percent) and smoking (26 percent) were common habits of 243 patients with stomach cancer. In the popula- tion-based Third National Cancer Survey (299), there was a signifi- cant positive association between smoking and stomach cancer. A few other retrospective studies have also reported a statistical association between smoking and stomach cancer (122,151,302). 132 +=HHITE flALES 3~ rWHITE FEllALES O=NONWHSTE HRLES IJl =NONWHITE FEMRLES 4 1970 1975 0' * ' 1955 1960 1965 1960 CALENORR YERRS AGE IN YEFlRS (BY S-YERR AGE CROUPS1 FEMWES +=1950-1956 L 21957-1963 0=1964-1970 c1=1971-1977 / * MRLES AGE IN YEnRS IBY 5-YERR RGE GROUPS1 FEllfILES + =1960-19.56 Ill =1967-1963 0=1964-1970 [II =1971-1977 10 20 30 40 50 60 70 SO AGE IN YERRS IBY S-YEAR ROE DROUPSl TABLE 38.-Stomach cancer mortality ratios-prospective studies Population Study size NO"- smokers All cigarette smokers Comments ACS O-State 186,OKJ Study white males U.S. Veterans Swedish Study Japanese Study California males in 9 occupations ACS 25&&e Study British Physicians 55,ooo males and females mifm males and females 358,400 males woo0 1.00 1.00 (men) 1.00 (women) 1.00 (men) 1.00 (women) 1.00 1.00 45-64 1.00 6.579 1.M) 1.00 1.61 1.52 Jbed on 176 micrascopically proved cases 1.30 Cigarette and 2.30 pipe smokers 1.59 1.31 1.04 1.42 1.26 1.39 All current smokers TABLE 39.-Stomach cancer mortality ratios by amount smoked-prospective studies Amount smoked Study Population size per day Mortality ratio Comment U.S. Veterans 290.000 males Nonsmoker 1.00 l-9 1.47 10-20 1.49 21-39 1.55 40+ 1.83 British 34,ooO males Nonsmoker Physicians l-14 15-24 25+ California males in 9 occupations Nonsmoker about `I1 pk about 1 pk about 1 `I? pk 1.00 1.20 1.65 1.39 1.00 1.09 0.94 1.25 Japanese Study 122,OCKl males Nonsmoker 1.00 l-19 1.46 20-39 1.53 40+ 1.76 Baaed on grams of t&XX0 per day In contrast with the above investigations, the Hawaiian Study of Five Ethnic Groups failed to show a statistically significant associa- tion between smoking and stomach cancer (123). Haenszel et al. (91) 136 reported an increased relative risk for stomach cancer among smokers in a series of 783 patients living in the Hiroshima and Miyagi prefectures of Japan; however, these findings were not statistically significant. In a similar study of Japanese living in Hawaii, these same authors (92) found a statistically significant increased risk among Issei smokers but not among Nissei. The absence of a significant association between cigarette smoking and gastric cancer has been reported by other authors (236,318). The relationship between smoking and stomach cancer was examined in several prospective studies (Table 38). Although mortal- ity ratios were increased for smokers as compared with nonsmokers, these increases were small. Three of the four major prospective studies noted a consistent dose-response relationship as measured by the number of cigarettes smoked per day. However, the magnitude of these relationships was moderate compared to that between smoking and other cancer sites (Appendix Tables A and B). Conclusion 1. Epidemiological studies have noted an association between cigarette smoking and stomach cancer. The association is small in comparison with that noted for smoking and some other cancers. Cancer of the Uterine Cervix Slightly over 8,300 women died of cancer of the uterine cervix in 1950. By 1977, the total number of deaths attributed to this site had decreased to 5,165. The age-adjusted rate for white females is only about one-third that observed for races other than white (3.53 versus 9.63) (Figure 59). The age-specific rate for races other than white was 17.92 in. 1950 and decreased to 7.99 by 1977. The agespecific rate for white females decreased from 10.12 to 4.12 over the same time period (Figure 60). Squamous cell carcinoma is the major cell'type. The overall 5-year survival for patients with carcinoma of the cervix is 60 percent, but survival ranges from 86 percent for those with localized disease, to 50 percent for those with regional involvement, and to 22 percent for those with distant metastases (2). Cervical cancer appears to be more common among women who have early and frequent coitus, who have early or multiple mar- riages or partners, and who become pregnant at an early age or frequently (140, 264). In addition, a number of other variables have been studied that may affect the risk for cervical cancer, including 137 25 20 15 10 5 0 Cb %=WHITE FEtlALES l?l=NONNHITE FEtlALES - 4 1950 1955 1960 1965 1970 1975 CRLENORR YEARS WHITE FEIIALES 0 L 0 10 20 30 40 SO 60 70 60 RGE IN YEARS IBY 5-YEAR RGE GROUPS1 j NONWHITE FEMRLES + z1950-1956 X=1957-1963 D -1964-1970 c!l=1971-1977 ,,-., I . . L--L---_L1 IO 20 30 40 50 60 70 60 ROE IN YERRS IBY S-YEAR AGE GROUPS1 venereal infections, circumcision status of consort, and exogenous hormones (264). A limited number of studies have attempted to identify an association between cigarette smoking and cervical cancer. One study (192) reported a relationship between smoking status (never smoked, ex-smokers, present smokers) and suspicious or positive cervical cytology. Thomas (264) administered a home questionnaire to 324 females with abnormal cervical cytology and reported that the prevalence of smoking was 70 percent in cases with carcinoma in situ and 58 percent in controls (0.02 5 p 16 3.40 All smokers 3.00 diseases and matched for age, race, hospital, and hospital status (semi-private versus ward). Socioeconomic status was determined by the subject's education and occupation and by the husband's occupation. Their analysis showed an overall positive association between cigarette smoking and cervical cancer. However, after Mantzel-Haenszel adjustment for age and socioeconomic status, the authors did not find a statistically significant association. The authors suggest that the association between smoking and cervical cancer is highly confounded and not consistent with a causal hypothesis. This study also, however, failed to include direct measures of potential confounding variables, such as sexual activity. It should be noted that in the Swedish (42) and German (201) studies, differences in socioeconomic status did not affect cervical cancer incidence. The associations described between cervical cancer and many other variables, in addition to the variation in results of studies of. the possible association of cigarette smoking and cervical cancer, do not permit a conclusion on the character of this relationship at this time. Conclusion 1. There are conflicting results in studies published to date on the existence of a relationship between smoking and cervical cancer; further research is necessary to define whether an association exists and, if so, whether that association is direct or indirect. 141 Smoking and Overall Cancer Mortality Introduction Several investigators have estimated the proportion of all cancer deaths attributable to tobacco use in the United States to range from 22 percent to 38 percent of all cancer deaths (70, 78, 106). The authors of a recent review of cancer causes (701, commissioned by the Congressional Office of Technology Assessment, conclnded that 30 percent of all U.S. cancer deaths are attributable to tobacco use ~Appendix Table 0. These estimates reflect a growing consensus that smoking is the single largest contributor to cancer mortality in the United States. Overall Cancer Mortality As early as 1928, Lombard and Doering (160), in a study of 217 cancer patients and 217 controls in Massachusetts, identified an association between heavy smoking (defined as all types of smokers) and cancer in general. This study is of historical significance in light of our present day knowledge about the relationship, between smoking and specific cancer sites. Over the last two decades, four of the eight major prospective studies have examined the relationships between smoking to overall and site-specific cancer mortality. Two of these studies (98, 120) included observations on females as well as males. Male smokers, regardless of the amount smoked, have approxi- mately twice the risk of dying from cancer than do their nonsmoking counterparts (Table 41). Data from these studies also showed a gradient increase in overall cancer mortality with the amount smoked. These data are presented in Table 42. Males who consumed more than one pack of cigarettes daily had overall cancer mortality rates almost three times greater than did nonsmokers. Mortality TABLE 4l.-Smoking and overall cancer mortality ratios- prospective studies Smokers Study Nonsmokers Male Female ACS 25State Study 1.00 1.79 1.18 pipe and cigar 1.21 U.S. Veterans Japanese Study ACS 9-State Study 1.00 1.00 1.00 2.12 1.32 cigars 1.29 pipes 1.62 1.41 1.97 cigarettes 1.44 pipe 1.34 cigar 142 TABLE 4Z.--Smoking and overall cancer mortality ratios in males by amount smoked Study Amount smoked per day Mortality ratio ACS S&ate Study Nonsmoker 1.00 l-9 1.87 lo-20 1.92 20+ 2.94 All smokers 1.97 U.S. Veterans Japanese Study Nonsmoker 1.00 1-9 1.42 x-20 1.95 21-39 2.66 40+ 3.31 All smokers 2.12 Nonsmoker 1.00 1-19 1.53 20-39 1.81 40+ 2.06 All smokers 1.62 A B C D E Em Current cqaretle smokers n Ex-cigarette smokers . FIGURE 61.-Mortality ratios for all cancer sites for ex- cigarette smokers by number of years of smoking cessation, U.S. Veterans Study NOTE: A: Stopped less than 5 years. B: Stopped 5-9 years C:Stopped1&14yean. D: Stopped 15-19 years. E: stopped 20 or more years. SOURCE: Roget and Murray 122~. ratios for male pipe smokers and male cigar smokers were 1.44 and 1.34, respectively (224). Female smokers had overall cancer mortali- ty rates 20 to 40 percent greater than female nonsmokers. Hammond (106) calculated that 34.5 percent of all cancer deaths in males were smoking related. These are in close agreement with estimates made by other investigators (70,216). Rogot and Murray (224) examined overall cancer mortality in ex- cigarette smokers compared to continuing cigarette smokers and 143 1 Ex- smokers Current smokers FIGURE 62.-Mortality ratios for all cancer sites for current and ex-smokers by number of cigarettes smoked daily, U.S. Veterans Study SOURCE Roqt and Murray 1Z4) found declining cancer mortality ratios for ex-smokers by the number of years off cigarettes. For those former smokers who had quit for. 20 years or more, the overall cancer mortality rate was approximately 25 percent above those of nonsmokers but substan- tially below those of continuing smokers (1.27 versus 2.12) (Figure 61). These investigators also noted that cancer mortality. among former cigarette smokers was correlated to the number of cigarettes smoked per day. A clear gradient by the amount smoked is evident for ex-smokers as well as continuing smokers for overall cancer mortality (Figure 62). Overall cancer mortality rates for former cigarette smokers were 40 peEcent greater than for nonsmokers. Conclusion 1. Cigarette smokers have overall mortality rates substantially greater than those of nonsmokers. Overall cancer death rates of male smokers are approximately double those of nonsmok- ers; overall cancer death rates of female smokers are approxi- mately 30 percent higher than nonsmokers, and are increasing. 2. Overall cancer mortality rates among smokers are dose-related as measured by the number of cigarettes smoked per day. Heavy smokers (over one pack per day) have more than three times the overall cancer death rate of nonsmokers. 3. With increasing duration of smoking cessation, overall cancer death rates decline, approaching the death rate of nonsmokers. 144 Summary 1. Cigarette smoking is the major cause of lung cancer in the United States. 2. Lung cancer mortality increases with increasing dosage of smoke exposure (as measured by the number of cigarettes smoked daily, the duration of smoking, and inhalation pat- terns) and is inversely related to age of initiation. Smokers who consume two or more packs of cigarettes daily have lung cancer mortality rates 15 to 25 times greater than nonsmokers. 3. Cigar and pipe smoking are also causal factors for lung cancer. However, the majority of lung cancer mortality in the United States is due to cigarette smoking. 4. Cessation of smoking reduces the risk of lung cancer mortality compared to that of the continuing smoker. Former smokers who have quit 15 or more years have lung cancer mortality rates only slightly above those for nonsmokers (about two times greater). The residual risk of developing lung cancer is directly proportional to overall life-time exposure to cigarette smoke. 5. Filtered lower tar cigarette smokers have a lower lung cancer risk compared to nonfiltered, higher tar cigarette smokers. However, the risk for these smokers is still substantially elevated above the risk of nonsmokers. 6. Since the early 1950s lung cancer has been the leading cause of cancer death among males in the United States. Among females, the lung cancer death rate is accelerating and will likely surpass that of breast cancer in the 1980s. 7. The economic impact of lung cancer to the nation is consider- able. It is estimated that in 1975, lung cancer cost $3.8 billion in lost earnings, $379.5 million in short-term hospital costs, and $78 million in physician fees. 8. Lung cancer is largely a preventable disease. It is estimated that 85 percent of lung cancer mortality could have been avoided if individuals never took up smoking. Furthermore, substantial reductions in the number of deaths from lung cancer could be achieved if a major portion of the smoking population (particularly young persons) could be persuaded not to smoke. 9. Cigarette smoking is the major cause of laryngeal cancer in the United States. Cigar and pipe smokers experience a risk for laryngeal cancer similar to that of a cigarette smoker. 10. The risk of developing laryngeal cancer increases with in- creased exposure as measured by the number of cigarettes smoked daily as well as other dose measurements. Heavy smokers have laryngeal cancer mortality risks 20 to 30 times greater than nonsmokers. 145 11. Cessation of smoking reduces the risk of laryngeal cancer mortality compared to that of the continuing smoker. The longer a former smoker is off cigarettes the lower the risk. 12. Smokers who use filtered lower tar cigarettes have lower laryngeal cancer risks than those who use unfiltered higher tar cigarettes. 13. The use of alcohol in combination with cigarette smoking appears to act synergistically to greatly increase the risk for cancer of the larynx. 14. Cigarette smoking is a major cause of cancers of the oral cavi.ty in the United States. Individuals who smoke pipes or cigars experience a risk for oral cancer similar to that of the cigarette smoker. 15. Mortality ratios for oral cancer increase with the number of cigarettes smoked daily and diminish with cessation of smok- ing. 16. Cigarette smoking and alcohol use act synergistically to increase the risk of oral cavity cancers, 17. Long term use of snuff appears to be a factor in the develop- ment of cancers of the oral cavity, particularly cancers of the cheek and gum. 18. Cigarette smoking is a major cause of esophageal cancer in the United States. Cigar and pipe smokers experience a risk of esophageal cancer similar to that of cigarette smokers. 19. The risk of esophageal cancer increases with increased smoke exposure, as measured by the number of cigarettes smoked daily, and is diminished by discontinuing the habit. 20. The use of alcohol in combination with smoking acts synergisti- cally to greatly increase the risk for esophageal cancer mortality. 21. Cigarette smoking is a contributory factor in the development of bladder, kidney, and pancreatic cancer in the United States. This relationship is not as strong as that noted for the association between smoking and cancers of the lung, larynx, oral cavity, and esophagus. The term "contributory factor" by no means excludes the possibility of a causal role for smoking in cancers of these sites. 22. In epidemiological studies, an association between cigarette smoking and stomach cancer has been noted. The association is small in comparison with that noted for smoking and some other cancers. 23. There are conflicting results in studies published to date on the existence of, a relationship between smoking and cervical cancer; further research is necessary to define whether an association exists and, if so, whether that association is direct or indirect. 146 24. Cigarette smokers have overall mortality rates substantially greater than those of nonsmokers. Overall cancer death rates of male smokers are approximately double those of nonsmok- ers; overall cancer death rates of female smokers are approxi- mately 30 percent higher than nonsmokers, and are increasing. 25. Overall cancer mortality rates among smokers are dose-related as measured by the number of cigarettes smoked per day. Heavy smokers (over one pack per day) have more than three times the overall cancer death rate of nonsmokers. 26. With increasing duration of smoking cessation, overall cancer death rates decline, approaching the death rate of nonsmokers. Technical Notes Age-Adjusted Death Rates Age-adjusted death rates show what the level of mortality would be if there were no changes in the age composition of the population from year to year. The age-adjusted death rates for the U.S. as a whole presented in this Report were computed by the Direct Method, that is, by applying the age-specific death rates for all causes of death or for deaths for a given cause to the standard population distributed by age. The total U.S. population as enumerated in 1940 is used as the standard population by the National Center for Health Statistics for presentation of mortality statistics. Standard popula- tions other than 1940 have been used by other agencies, organiza- tions, and researchers in presenting mortality data. This introduces some problems of comparability in the presentation of the statistical findings drawn from a variety of sources. Cause-of-Death Classification National mortality statistics from the National Center for Health Statistics for the U.S. presented in this Report are classified in accordance with the World Health Organization (WHO) Regulations, which specify that member nations classify causes of death in accordance with the International Statistical Classification of Dis- eases, Injuries, and Causes of Death. The deaths are tabulated and presented in Vital Statistics of the United States, Volume II, Mortality by cause-of-death categories that are consistent with WHO recommendations. Other organizations and researchers whose work is cited in this Report may use different cause-of-death categories. This introduces some problems of comparability in the presentation of the statistical findings drawn from a variety of sources. Another problem of comparability in mortality rates is introduced when comparisons are made over time for specific causes of death. This is because of the practice to periodically revise the Internation- al Classification of Diseases (ED) by which causes of death are 147 classified and tabulated. The ICD has been revised approximately every 10 years since 1900 to keep abreast of medical knowledge. Each decennial revision has produced breaks in the comparability of cause-of-death statistics. For many of the causes of death described in this Report, the reader may refer to the NCHS report (199) for information about comparability in cause of death statistics due to revisions in the ICD during 1950-1977. Appendix Tables APPENDIX TABLE A.-Mortality ratios (smokers vs. never smoked regularly) for smoking-related cancers among females-ACS 25-State Study and Japanese Study Underlying cause of death Mortality ratios Cancer (total) ACS JlZpWW 1.21 1.41 Lung (excl. trachea. pleura) Buccal cavity, pharynx, larynx, and esophagus Pancreas Uterus Uterine cervix Esophagus Stomach , Bladder 3.56 2.03 3.25 6.52 1.42 - 1.18 - - 1.72 4.89 - 1.21 1.31 2.56 2.00 APPENDIX TABLE B.-Mortality ratios (smoker vs. never smoked regularly) for smoking-related cancers among males-ACS 25-State Study and U.S. Veterans Study Underlying cause of death Mortality ratios Cancer (total1 ACS U.S. Veteran.5 Bee- &e 65-79 All 2.14 1.76 2.12 Lung (excl. trachea, pleura) Buccal cavity, pharynx Larynx E-Www Bladder and other urinary Kidney PEState Pancreas Liver. biliary passages 7.64 11.59 9.90 2.93 6.09 8.99 4.17 1.74 2.00 2.96 1.42 1.57 1.04 1.01 2.69 2.17 2.64 1.34 1.42 1.26 11.28 4.22 11.49 6.43 2.16 1.41 1.31 1.79 - 1.52 148 APPENDIX TABLE C.-Cancer deaths caused by tobacco: United States, 1978 Number of deaths Certified cause of deatha Observed Estimated, had Americans not smoked Approximate excess number and percent of deaths attributed to tobacco (percent m parentheses] Cancer, males Lung Mouth, pharynx, larynx, or esophagus Bladder Pancreas Other specified sites Unspecified sites Total, males Cancer, females Lung Mouth, pharynx, larynx, or esophagus Bladder Pancreas Other specified sites Unspecified sites Total. females Total, males and females 71.006 6.439" 64.567 (90.91 14.282 6.771 11,010 loo.799 14,469 218,337 24.080 5.454h 16.626 177.4) 5.100 3,078 9,767 127,642 13,951 183,618 401,955 1.79" . 2 2.9601) 6.585'* - 8,18% 1.458x 2c 2,170'* 7,291b 11.879 10.698 174.91 3,811 (56.3) 4,425 (40.21 5.oood I 5.01 6,281 143.4) 94.782' (43.41 2,184 t42.81 908 (29.51 2,476 125.41 1.m - 2,072 (14.9, 27,266' (14.8l 122,048' (30.41 "Site of origin oicancer bNumber estimated by opplymg the nonsmoker mortality rates reported by Carfinkel (861 to the U.S population of 1978. ~Double the number estimated by the procedure descr&d in footnote b This number was doubled to allow ior the possibility that the subjects in the ACS prospective study were less exposed to alcohol or to some other cause(s) of cancer of the upper respiratory or &grstlve tracti than were average people m the United States. j Some evidence that this was mdeed the case is that even the cigarette smokers I" the ACS study had mortality rates ior these types of cancer that were somewhat below the netional US rates l9H) ) However. it makes IntIe difference to our grand totals whether the small number of cancers of the mouth and throat "expected" irom the ACS nonsmokerexperrence we leit unaltewd. are doubled, or are trebled. dOther specified sites include some. such as ludney. that may truly be aiiected by tobacco, and some, such as stomach or liver, that include a proportion of misdiagnosed cases ofcigarette-mduced cancer of the lung, pancreas, and other organs Some iractio" oi the cancers certified as being of other specified sates IS thus due to smoking, which in part explains the excess mortality among smokers I" the aggregate of all such cancer that IS iound in the American prospective studies (Appendix Tables A and BI. We have suggested. without firm evidence. that of these other cancers. perhaps 5.ooO male and l.OCQ female cases may have been due to tobacco These suggested ligures. totaling 6,000, may shghtly underestimate the actual figures, but renders may substitute any estimate that they consider more plausible, e.g. some other estimate between 1.ooO sod 20.(X3.3, leadmg to an estimate of `Z9 to 34 percent oi 1978cancerdeaths aszribable to tobacco. *Estimated to match the proportions (43 percent male. 15 percent female) ofspecined wtes attributed to tobacco `The percentage ascribable to tobacco is gradually increasing as lung cancer death rates are ~ncrerwng among older Americans. SOURCE: Doll and Pete, 70,. 149 References (I) ABELIN, T., GSELL, O.R. Relative risk of pulmonary cancer in cigar and pipe smokers. Cancer 20(S): X88-1296, August 1967. (2) AMERICAN CANCER SOCIETY. 1982 Cancer Facts and Figures. American Cancer Society, Inc., New York, 1981,31 pp. (3) ANTHONY, H.M., THOMAS, G.M. Bladder tumours and smoking. Interno- tional Journal of Cancer 5f2): 266-272, March 151970. (4) ANTHONY, H.M., THOMAS, G.M. Tumours of the urinary bladder-An analysis of the occupations of 1030 patients in Leeds, England. Journal of the National Cancer Institute 45(5): 879-895, November 1970. (5) ARCHER, V.E., WAGONER, J.K., LUNDIN, F.E., Jr. Uranium mining and cigarette smoking effects on man. Journal of Occupational Medicine 15f3): 204-211, March 1973. (5a) ARMSTRONG, B., GARROD, A., DOLL, R. A retrospective study of renal cancer with special reference to coffee and animal protein consumption. British Journal of Cancer 33: 127-136,1976. (6) ASHLEY, D.J.B., DAVIES, H.D. Lung cancer in women. Thorax 24(4): 446- 450, July 1969. (?I AUERBACH, 0. The pathology of carcinoma of the bronchus. New York Stute Journal of Medicine 49: 900-907, April 151949. (8) AUERBACH, O., GARFINKEL, L., PARKS, V.R. Histologic type of lung cancer in relation to smoking habits, year of diagnosis and sites of metestases. Chest 67(4): 382-387, April 1975. (9) AUERBACH, O., GERE, J.B., FORMAN, J.B., PETRICK, T.G., SMOLIN, H.J., MUEHSAM, G.E., KASSONRY, D.Y., STOUT, A.P. Changes in the bronchi- al epithelium in relation to smoking and cancer of the lung. New England Journal of Medicine 256t3): 97-104, January 1957. (10) AUERBACH, O., GERE, J.B., PAWLOWSKI, J.M., MUEHSAM, G.E., SMO- LIN, H.J., STOUT, A.P. Carcinoma-in-situ and early invasive carcinoma occurring in the tracheobronchial trees in cases of brcnchial carcinoma. Journal of Thomcic Surgery 34(3): 298-309, September 1957. (11) AUERBACH, O., HAMMOND, EC., GARFINKEL, L. Histological changes in the larynx in relation to smoking habits. Cancer 25(l): 92-104, January 1970. (12) AUERBACH, O., HAMMOND, E.C., GARFINKEL, L. Changes in bronchial epithelium in relation to cigarette smoking, 19551960 vs. 1970-1977. New England Journal of Medicine 300(8): 381-386, February 22,1979. (13) AUERBACH, O., HAMMOND, E.C., KIRMAN, D., GARFINKEL, L. Effects of cigarette smoking on dogs. II. Pulmonary neoplasms. Archives of Enoiron- mental Health 21W): 754-766, December 1970. (I* AUERBACH, o., PETRICK, T.G., STOUT, A.P., STATSINGER. A.L., MUIIH- SAM, G.E., FORMAN, J.B., GERE, J.B. The anatomical approach to the study of smoking and bronchogenic carcinoma. A preliminary report of 41 cases. Cancer9(1): 76-83, January-February 1956. (1.5) AUERBACH, O., STOUT, A.P., HAMMOND, E.C., GARFINKEL, L. Changes in bronchial epithelium in relation to sex, age, residence, smoking and pneumonia. New England Journal of Medicine 267(3): 111-119, July 19, 1962. (16) AUERBACH, O., STOUT, A.P., HAMMOND, EC., GARFINKEL, L. Histologi- cal changes in esophagus in relation to smoking habits. Archives of Environmental Health 11(l): 4-15, July 1965. 150