Tobacco Documents Online

u m • I| DHEW Publication No. (NIH) 74-544 DEPARTMENT OF HEALTH, EDUCATION AND WELFARE Public Health Service National Institutes of Health T109781644

merit completinn with or without case monitoring assistance study suggests a positive effect ef implementing a case mort- is the same. If this is the case, then additional time and ex- itoring system aimed at diagnosis and treatment initiation in pense of extended case monitoring (through treatment corn- the EPSDT program. pletion as received by Group III) may not be cost effective. Of special interest is the large proportion (78 Fer cent) of R~g~..K~,Iz'e~C ES treatment completions involving multiple visits achieved by I. Cowan DB~ Suarex MM: Child health screen- individuals in Group I who received no monitoring assist- ing for the~a~rpraetitioner. Nurse Practitioner, Vel. 1, No. ance. These individuals may represent a high degree of self Jan.-F~t~1976, pp. 109-120. ............... 2~reen~ng, in A. O~csby and H. m.o.uvatmn ano aaequ, ate gnowie.agc, o~ me neaatn care prO~aI InstitUte-on Earlier vtder system needed m order to obtain care. In Dallas~itions in Childhood, Berke- merit providers are readily available and a large, ~hco~e Health, 197I, pp. public children's clinic an~ hospital is accessib~a~ public 52-56. transportation. In areas ,,~here providers ransportation .... :lers.~ransportation ACKNOWLEDGMENTS are scarce, results may miter . . . ..... • ~~ich this paper was Several hmltatlons ot ~n i-c ~Ad-m~nlstration The population within th~~n contract was held by The ties and the low r~niversity o_fTexas Health aram (10 net c~h the grantee, The Texas State raoly s sample ,'uso, expcnmentaJ conataons were " The authors would Hkc to acknowledge the support of the time-sequenced, so that changes in program operation over Health Services Research Institute and in particular, assistance giv- time could have influenced the results. Nevertheless, this en by Arthur Bfitt. Smoking and Li ng U.S. Veterans Estimates of life expectancies for white males according to cigarette smoking habits have been reported by Ham- mond3 These were based on life tables constructed from a 5- year mortality follow-up of 447,196 men with known smok- ing information, of whom 39,178 died in the study period, July 1, 1960-June 30, 1965. The present study reports on smoking and life expec- tancy among U.S. veterans who held government life insur- ance policies in December 1953 and were followed for 16 years, from January 1, 1954 through December 31, 1969. In previous reportsfl, a mortality experience was described in terms of 16-year (and 13-year) probabilities of death. By an approach described below, l-year probabilities of death for single years of age are derived for selected groups of smokers, ex-smokers and nonsmokers and life tables are constructed. Estimated life expectancies are then compared between groups and with values obtained by Hammond. The nature of the study population, the study design, and other details have been fully described elsewhere.~-~ A brief review follows for purposes of this paper. In January 1954, questionnaires on smoking habits were mailed to 293,958 U.S. veterans who held insurance policies mad were in the age group 3I--84. A total of 198,820 or 6S per Address reprint requests to Eugene Rogot, Epidemiology Branch, Division of Heart ~.~d Vascular Diseases, National Heart, Lung. a~d BlmrJ Institute, Federal Bailding, Re,am 2C~. Bethesda, MD 2~i4. This paper, submitted to ~e Journal February 13, 1978, was re,.'ised and accepted for publication April 14, I978. cent responded. These respondents comprise the "1954 co- hort," and their survivorship experience is the basis of the present report. Attalnedage 1954 1955 1956 1957 ... 1969 31 ~ 31 /32 /33/34 32 32 "~ 33"~ 34- 46 47 48 49 84 84 85 86 87 ... 99 An important concem~whether bias exists between smoking and response--cannot be determined directly from the data. However, indirect evidence presented in the earlier reportsz.a suggests there is no bias between smoking and response to the questionnaire. Almost all policyholders were white males, drawn from the middle and upper socioeconomic classes. Most of the policyholders were veterans of" World War 1. The Veterans Administration provided names of all pol- icyholders known to have died in the study period. The over- all mortality follow-up with respect to the fact of death and year of death, is considered to he almosl If)0 p~r cent com- plete. The full details of the various~l~,~.~u,p_.p,~ocednres appear in the previous reports."-~ AJPH Octcber, 1978. VoL 63, No. 10 1023 TI09781645

PUI~LID H~Jd.TH BRIEFS For each group of interest, a I-year probab~ity of death, ct~, was estimated as follows. Consider a given year of attained at any time in the 16-year period. Count a Ferson who is x years old in 1954 as x + 1 years old in 1955, x + 2 years old in 1956, etc., provided he survives. Ifhe dies, he is counted as entering emch )'ear from 1954 up through the year of death. A diagram may serve to illustrate. The numbers entering a given age are poole(i (along a diagonal) and constitute the total persons at risk for the given age. The numbers of deaths for a given age are collected in the same way. Dividing the latter by the former gives a 1- year probability of death, q~, for that age, x. This approach is justified if we assume: 1) no secular trend in mortality from 1954-1969, and 2) no change in a per- son's smoking status from 1954-1969. The first assumption was checked by single years of age and held up. The second assumption could not be tested. Values of qx were obtained as described above, for each year of age, x, from 3 ! to 99, for each group of interest. Since numbers at risk tended to be small for ages under 35 or over 90, life tables were constructed for the shorter age span, 35- 99. No smoothing of probabilities was undertaken. The clos- ing life expectancy value in each table, that for age 90, was taken as 3.16 years, which is the value given in the 1959-61 U.S. Life Tables for White Males.7 Total deaths for attained ages 35-89 numbered 69,643 for the entire 1954 cohort. The main results are shown in Table I where estimates of the average remaining years of life, ~x, at ages 35, 40, .... 85 are given for selected smoking groups. S -t,xndard er- rors of these estimates were also calculated but for simplicity are shown only for age 35. ~,Vhere the stand~d error is great- er than .49 years, the ~x value is asterisked in Table 1. A comparison of nonsmokers with cigarette smokers by amount smoked shows fairly substantial and consistent dif- ferences in estimated life expectancies between groups, at each age shown. Life expectancy varies inversely with num- ber of cigarettes smoked Per day. The most pronounced dif- ferences are between nonsmokers and heavy cigarette smokers (40+ per day). These differences were greatest at lhe younger ages (8.7 years at age 35 and 8.8 years at age 40) and least at the older ages (2.9 years at age 80 and 1.6 years at age 85). In relative terms, these differences may be di- vided by the llfe expectancies of nonsmokers at the corre- sponding ages, and a "per cent loss" in average remaining lifetime derived. Thus, for heavy cigarette smokers, per cent losses were: At age % loss in ~x 35 20 40 23 45 24 50 27 55 29 60 3I 65 34 70 36 75 36 80 39 85 31 TABLE 1--Life Expectancy in Years for Selected Smoking Groups at Specified Ages: U.S. Veterans' study, 1954 cohort, January 1, 1954--December 31, 1969. Life Expectancy in Years Ago in Years at Next Birthday: Standard Error of Smoking Status in 1954 35 40 45 50 55 60 65 70 75 80 85 (}35 in years Nonsmokers 43.5 38.7 34.0 29.4 25.0 20.8 17.0 13.5 10.2 7.5 5.2 .15 Cigarette smokers 38.2 33.5 29,0 24.7 20.7 16.9 13.6 10.7 8,3 6,3 4.7 .10 No. c~garettes per day: < 10 41.0 36.3 31.7 27.5 23.1 19.0 15.4 12.0 9,2 6.7 4.7 .33 10-20 38.7 34.1 29.5 25.2 21.0 17,2 13.8 10.8 8.3 6.3 4.7 .14 21-39 36.7 32.0 27.5 23.4 19,4 15.8 12.6 9.8 7.7 6,0 4.8 .18 40+ 34.8 29.9 25.7 21.6 17.8 14.4 11.2 8.6 6.5 4.6" 3.6" .45 Age began (years); < 15 36.1 31.5 26.9 22.9 19.1 15.6 12.5 9.9 8.0 7.3* 7.4" .34 15o19 37.4 32.7 28.2. 24.0 20.0 16.3 13.1 10.2 7.9 6.0 4.5 .14 20-24 38.7 34.0 29.5 25.1 20.9 17.1 13.6 10.6 8.0 5.8 4.1 .20 25-34 40.8 36.0 31.3 26.5 22.4 18.5 15.1 12.0 9.4 7.4 5.3 .38 Ex-cigaretto smokers 42.1 37.3 32.5 27.9 23.5 19.6 15.9 12.6 9.7 7.2 5.4 .16 (not on doctor's orders) Max. amt. smoked (no. cigar~tteslda): < 10 43.5 38.9 34.3 29.5 25.0 20.8 17.0 13.5 10.5 7.7 5.7 .42 10-20 42.2 37.3 32.6 27.9 23.5 19.5 15.7 12.4 9.4 6.9 5.1 .23 21-39 41.1 36.1 31.2 26.8 22.3 18.6 15.1 11.9 9.1 6.9 5.0" .30 40+ 40.4" 35.9" 31.0" 26.4* 22.3 18.5 15.0 12.1 9.5° 7.6" 6.1" .67 Smok~_rs of cigars only 42.4" 37.7 32.8 28.3 23.9 19.7 15.9 12,6 9.6 7.0 5.2 .52 Smokers of pipes only 42.8" 38.0" 33.4 28.8 24.7 20.3 16.7 13.2 10.2 7.6 5.7 .55 Tot~11954 coho~ 40.8 36.0 31.4 27.1 22.9 19.0 15.5 12.3 9.5 7.1 5.1 .07 error ct ~x ;~ .50 ye~,~s AJPH October, 1978. VoL 63, I'lo. 10 T109781646

Life expectan=ics for cig'-~ette smokers varied directly with a~e began smoking. Fairly large differences are served at the younger ages, especially between those who began smoking early (before age 15) compared with those starting late (age 25-35). For all ages, differences in life expectancy between non- smokers and ex-cigarette smokers who stopped for other than doctor's orders were less than those between cigarette smokers and nonsmokers. For the ex-cigarette smoker .group, life expectancy is seen to vary inversely with maxi- mum amount smoked. Among those who had been light smokers (< I0 cigarettes per day), values of Sx were equal or close to those of nonsmokers for nearly all ages. Pure cigar and pure pipe smokers had much smaller deficits in 8= than cigarette smokers. Over most of the age range shown the cigar smokers had life expectancies about 1 year less than nonsmokers while pipe smokers had life ex- pectancies .6 or .7 of a year less than nonsmokers at the younger ages and exhibited less of a deficit thereafter. In Table 2, our results are compared to Hammond'sI for life expectancy at age 35. Since age in the Veterans' Study is TABLE 2--Comparisons of Life Expectancies Obtained in the Hammond Study (H) with those in the Veterans' Study (V): Selected Smoking Groups, White Male-, age 35. <}~s in yearn Smoidng Status in x 100 years Total 39.3 40.3 1.0 3 Nonsmokers* 42.4 43.0 .6 1 Cigarette smokers* No. of cigarettes per day: <10 37.8 40.6 2.8 7 10-193 37.1 38.3 1,2 3 20-393 36.5 36.2 -.3 - 1 40+ 34.7 34.3 -.4 -1 began (years): < 15 34.6 35.7 1.1 3 15-19 36.0 37.0 1.0 3 20-24 37.7 38.3 .6 2 25-34 38.8 40.4 1.6 4 1Hammond EC: Life expectancy o! American men {n relat;on t~ th8~" smo,V~ng ha~s. J Nail Cancer [nst 43:951,962. 19~9. 2U.S. Veterans" Study. 1954 Cohort. Averag~ of values for 35 and 36 3Ca~eg~rles were 10-20 and 21-39 in Veterans" Study. "Smoking stalus at tkne o! inflation of study FU3LIC HE~.LTH ER;EFS in years as of next birtkday, we took the, average of~ values for 35 and 36 in order to compare to th~ life expectancies of 35-year-o!ds in Hammond's study. (The. latter figure is as- sumed to be 33.0.) Differences in Table 2 appear to be fairly small. The vet- erans have somewhat greater life expectancies for most cate- gories shown. Relative differences between the studies are fairly small. The inverse pattern of amount smoked (ciga- rettes per day) with life expectancy and the direct relation- ship between age began cigarette smoking with life expec- tancy hold up in both studies. Our results clearly confirm Hammond's basic findings. Life expectancies were estimated for selected groups of smokers, ex-smokers, and nonsmokers based on the results of a 16-year mortality follow-up of 198,820 U.S. veterans. Life expectancy varied inversely with number of cigarettes smoked per day. The most pronounced differences were be- tween nonsmokers and heavy cigarette smokers (40+ per day). These differences in life expectancy were greatest at the younger ages--nearly 9 years at ages 35 and 40. Life ex- pectancies for cigarette smokers varied directly with age be- gan smoking. For all ages, differences in life expectancy be- tween nonsmokers and ex-eigarette smokers who stopped for other than doctor's orders were less than those between nonsmokers and current cigarette smokers. Results in the present study clearly confirmed Hammond's earlier findings. REFERENCES I. Hammond EC: Life expectancy of American men in relation to their smoking habits. J Nail Cancer lnst 43:951-962, 1969. 2. Rogot E: Smoking and General Mortality among U.S. Veterans, 1954-1969, DHEW Publication No. (NIH) 75-544, 1974. 3. Rogot E: Smoking and Mortality among U.S. Veterans. J Chron Dis 27:189-203, 1974. 4. Dora HF: The mortality of smokers and nonsmokers. In Proe. Sue. Stat. Sect. Amer. Stat. Assoc. 34-71, 1958. 5. Dora HF: Tobacco consumption and mortality from cancer and other diseases. Public Hlth. Rep. 74:581-593, 1959. 6. Kahn HA: The Dora study of smoking and mortality among U.S. veterans: Report on 8 ~ years of observation. In Haenszel W., Ed. Epldemiological Approaches to the Study of Cancer and Other Diseases. Bethesda, U.S. Public Health Service, National Cancer Institute Monograph 19, 1966, pp. 1-125. 7. United States Life Tables: 1959-1961. Public Health Service Pub- lication 1252, Vol. 1-No. 1. USDHEW, Washington, DC, Dee. 1964. AJPH Oc~]:~r. 1978. Vc, L 6~, No. 10 10"=5 TI09781647

Cancer Mortality Among Nonsmokers insured. Group of U.S. Veterans, in an ~and J. Murray ABSTRACT---In a 16-yemr folIow-up of insured U.& veterans, stendard~ed mort~il~y r,,tlas (SMR) for carc~r were for nonsmoke~ (55,049}; Is the standMd, cancm- fetes of respondents (24~,046) were used. For a~! cantle, nonsmcke~ had an SMR of 71%; for lung cancer, they h~d an SMR of 21%. The dete for nonsmokers were further an~Jyzed with rL~k of ~ncer assec~ed eccordlng to ris|dl~ce Ind u$~,~ occupation and IndustW. Among the nonsmokerS, SMR were especially low for dentlsf~, carl:enters, and lawyer~, and for the e~ectrtc light end power Ind printing Industde|. By division of the country, SMR for no~smokers were e~peclllly low for the Mountain States.wJNCI 65: 1163-1168, 1980. Follow-up of a defined population of U.S. veterans that yeas planned by Dorn in 1952 with the general objective of studying the relationship of the use of tobacco, residence, and occupation to mortality has furnished da~a for several studies by Dorn (I, 2), Kahn (3), and Rogot (¢-6). The present study examines cancer mortality in a group of veterans at lower risk of de~eloping cancer, i.e., nonsmokers. SUBJECTS AND METHODS With the cooperation of the Veterans Adminisu-adon, policyholders of U.S. Government Life Insurance were selected [or study. This insurance was available to persons who served in the U.S. armed forces from 1917 to 1940. Most of the policyholders were veterans of World War I. All persons with an active policy at the end of 1953 were included except for a few special grou~, such as persons with total and pe~'manem disability. Beginning in January 1954, a questionnaire requesting informa- tion c.oncerning the use of tobacco, usual occupation, and industry was marled to about 300,000 policyholders. Usable replies were received from about 200,000 per- sons or 68~ (table 1). This group of respondents was labeled the "1954 cohort." A second questionnaire was mailed m the nonrespondents beginning in January 1957. Usable replies were recdved from an additional ,t9,000 policyholders (labeled the "1957 cohort") for a total of 248,000 or about 85% of eligible persons for whom information concerning the use of tobacco, occupation, indnsu-y, and residence was obtained. The mortality experience of non.respondents was recorded. Whenever a claim was filed for the payment of a policy, a copy of the death notice, usually a copy of the official death certificate, was sent to the office conduct- ing the study. Additional medical information, which included verification of the causes of death entered on the death certificate, the procedures used to establish these diagnoses, a diagnosis of cancer even though it was not considered m be an underlying or contributory cause of death, and the histologic tS"pe of cancer, was requested from the physidan who signed the death certificate or from the hospital where the death cuffed. Verification of the cause of death was not requested if the death occurred outside the United States, was due to an acddent, or was certified by a coroner. Replies to more than 99% of the Ietters of inquiry were receivecL These special verification pro- cedures for cause of death were observed for the first part of the study only, i.e., from 1954 through 1962. All smoking classifications used in this report are based on responses to the 1954 and I957 questionnaires. The category "nonsmoker" refers to persons who never smoked dgarettes, dgars, or pipes; also included are the "occasional" smokers (persons who smoke once in a while but not every Almost all policyholderg were white males, drawn mainly from the middle and upper socioeconomic classes. The Veterans Administration provided names of all policyholders known to have died in the study period. For about 75,000 persons whose insurance policies had terminated in 19~-69, special searches were made by the Veterans Adminismadon index staff to ascertain deaths. The overall mortality follow-up, with respect to the fact and year of death, is considered to be almost 100% complete. The full details of the various follow-up procedures appeared in {2-3). With respect to cause of death, special searches were made in state vital statistics offices to obtain copies of certificates for 34,F~I4 deaths needed to complete the latter ~ o[ the l~-year follow-up. All but 2,541 death certificates were located. Causes of death were coded according to the ICD (Seventh Revision, 1955). The underlying cause of death, used in this report, was available on 97.6,~ of all deaths (table 1). The SMR for nonsmokers were calculated with the age-specific probabilities of death in the study period og all respondents used as standard rates, gxpected ASaRgV~T~ONS USSD: ICDffilnternafional Classification of Diseases: SMg~standardized mortality ratio(s). ~ Presented at the Workshop on Populations at Low Risk of Can- cer cond,~-ted a~ Snowbird, Utah, August 2~-25, 1978. ~ Epldemiology Branch, Division of Heart and Vascular Diseases. National ~eart, Lun~, a~d Blo~d Insdtme, National Insdtm~s of H~alth. Public H~lth Service. US. Departmem of Health and Human Services. Bethcsda, Md. ,~1Y205. ~ Biome~'5" Bmn,'h, Division of Cancer Cause and Pr~'endon. Na- tional Cara:el- Institute, National Insdtmes ~[ Health. 1163 TI09781648

1164 Ro~t and MuR~ 1954 193,820 69,$58 68,446 1,412 98.0 1957 49,226 16,877 16,521 ~ 97.9 No reply 4~,912 20.828 20,~5 773 Total 293,958 107,56~ 105,02~ 2,54I 97,6 numbers of deaths by cause were-calculated separately for the 1954 and 1957 cohorts and then combined. Age groups used were those shown in table 2 except for state data in which the combined age group 30-54 years was used instead of the separate age groups of $0-34, 35-44, and 45-54 years. The more detailed breakdown by states was not readily available. The difference in expected cancer deaths calculated by means of the coarse and more detailed age groupings ~as too small to affect the overall SMR for total cancer or for lung cancer. RESULTS In the tables that follow, a small number of cancer deaths allocated to female sites, as well as a few deaths with uncertain causes, have been combined with the "other and unknown" category and are included with the "total cancer" deaths. These number $8 of the 133 other and unknown group and represent about 1% of the cancer deaths among nonsmokers. The effect on the SMR for total cancer is considered negligible. Some general characteristics of the population under study are treated in tables 9~ and 3. Table 2 shows the age breakdown of the total group of respondents and of the nonsmokers among them at the beginning of follow-up. For each group, the ages of 55-64 and 65-74 years~dominate. About two-thirds of our population were between 55 and 65 years of age at the start of the study. Residence at the beginning of the study for non- smokers and total respondents compared with U.S. Age. ~ra No. Percent Non- Total Non- Total smokers respondents smokers respondents 3O-34 LOgO 7,4~4 2 3 35-44 4~'-'05 £3.891 8 10 45-54 2.078 11.~59 4 5 ~5-64 37,~i0 164,399 69 66 65-74 9,319 ~.6~5 17 16 75-84 ,~7 2,048 1 1 Total 55.049 248.046 lo0 100 Age range is that at I~eginnin~ of foll~w-up. J.xcl. vOL 6~. xo. 3..xox'z.xmzR z~ Residence" Percent Non- Total U.S. stackers respon- white dem~ males Urban~ 80 83 63 Rural 20 17 37 Division~ New England 6 7 7 Middle Atlanti~ 19 20 21 E~t North Central 20 20 21 West North Central 12 11 10 South Atlantic 12 12 12 East South Central 4 4 7 West South Central 7 7 9 Mountain 4 4 4 Pacific 15 15 10 Total IO0 100 100 • Residence is indicated for beginning of follow-up (1954 or 1957) ~or veterarm and an of April L 1950. for U.$. white males. * Urban,fplaces of >_2.500 in population as o! April 1, 1950; rurM= all other. ~ DivLsions are states in designated ~nsus areas. white males in 1950 is given in table 3. More of the nonsmoking veterans lived in urban areas than did their counterparts in the general population: 80 versus 63%. By census divisions of the country, the percent distributions were generally similar, except that a larger proportion of the nonsmokers came from the Pacific States than did the O.S. white males: 15 versus 10%. A comparison o[ nonsmokers and total respondents wi~h U.S. white males with regard to major occupation SMR 0 20 40 O0 SMR study. TI09781649

Cancer Mortality Among Nonsmokers 1165 To~l malignant Lip, ~1 ~W, and ~p (140.~149.9) To~ (141.~141,9) ~~x (1~ O~er ~d uns~i~ pha~x (146.~I46.9. I~.~148.9. 149) ~fi~ o~s and ~d~eum ~p~ (I~) S~m~h (151.~151.9) 8mall in~s~ne (152.~152.9) Colon (1~.~1~.9) ~r (I~.0. 1~.8) G~der (1~.0) ~er biliaW ~s (155.1, I~.1-1~,9) P~e~M (157.~157.9) ~t~HWne~, ~H~na~, and uns~ diz~ti~ o~ in- 159, 1~,~) ~spI~WW ~x (161.~161.9) ~er Rspi~W o~a~ includin~ pleu~, m~i~tinum, ~d in~- tho~elc si~ (160,~1~.9. 1~) ~nes ~ join~ (1~0.~170.9) Conn~d~, su~u~n~, and o~er ~tt ti~es S~n, melano~ (172.~1~,9) Bre~t (174) P~fs (187.0) S~m and o~er male ~ni~l ~m (1~,5. 1~.8. 1~.9) 2.745 24 0 9 5 5 4 1 1,092 19 194 8 444 155 Exp S~R 3,~.7 71 1.6 0 ~.3 39 8~ 61 6.8 59 25.6 4 I,~.5 84 67.0 ~.0 85 10.8 74 4~.I 93 I~.7 94 59 97.5 61 195 241.1 81 18 17.3 104 190 890.9 21 6 33.7 18 175 844.9 21 9 12.3 73 16 14.4 111 11 15.2 72 27 35.8 75 430 460.4 93 423 453.2 93 5 4.5 111 2 2.7 74 Obs E~ SHE Urinaw sys~m 217 273.2 79 Bladder and other urinary organs 125 163.3 74 (188. 189.2-1~9~) K~dney and rer~tl pelvis (189.0. 92 104.9 .~ 189.1) Eye (lgO) 6 33. 158 Newous s~tem 102 I01.9 1(~0 Br~in (191) Other ~ervous system (192.0-I92.9) ~.~docri~e system 17 16.8 101 Th>coid ~]and (193) 15 11.1 135 Other e~docri~e gZaeds 2 5.7 35 (194.0-194.9) I#'mphoma~ 193 183,5 105 Lympho~reoma and reticulum cell 132 1~23. 108 sareomM (2{)0.0-200.1) Hod~kin's d~secse (201) ~6 49.8 88 Other lymphom~s ineludin¢ myeo- £5 20.5 1~2 sis funEoides (202.0-£02.9) Multiple myeloma (203) 87 "/4,1 117 Leukemht 162 200.7 81 Lymphatic (lymphocytic) Acute (204.0) Chronic (204.1) Other (204.9) Other leukemi~ (e.g.. granulocytic or monocytic) Acute (205.0, 206.0, 207.0) Chronic (205.1, 206.1. ,°07.1) Other (205.9. 206.9. 207.9, 2073.) Other and unknown primary~ 171 197.4 87 (195.1-195.9, 196.0-196.9, 197.0-- 197.7. 197.9. 198.0-198.9. 199.0- 199.1, 173.0-173.9) • These records were coded with the use of the ICD (Seventh Revision~ Their equivalent numbers in the ICD (Eighth Revi- siqn) are shown in parentheses, " The .ICD (Eighth Revision) code numbers for 208 (poly- cythemi~ veto) m~d 209 (myelof~rosis) are excluded. If any" tot~! malignant neoplasm numbers are to be used, the~e must be included here. and major industry (dam no~ shown) revealed that nonsmokers were distributed among occupational and industrial categories in ~he same manner as was the total study population..ks expected, the smd.v and U.S. populations were no~ distributed in the same way on these factors; the veterans occupied more professional and managerial positions in business and public ad- ministration. The main resuhs of our study are summarized according to SMR in rubles 4 and 5 and in rex[- figures 1-7. In text.figure 1, 5MR ar~ shown in rank order for nonsmokers by body system. Sites at the ~op (respirator, s.vstcm and lip, oral c~viw., and phaD'nx) are strongly related to smoking, with SMR of 21 and ~, respecdvel.v. Those at the bottom (nervous system, endocrine sys{em, l.vmphomas, bones and joints, and multiple myeloma) have SMR of 100 or greater and nre not related ~o smoking. Comparative cancer mortality levels for nonsmokers are shown in text-figures 2 and 3 by major occupation and industry, respectively. Farmers had the lowest SMR (66), whereas sales workers and private household and service workers had the highest (79). Overall, SMR were not s~stly different by major occupation nor by major industry. The lowest SMR seen was for agricul- ture (65), and the highest was for the wholesale and retail trades (77). SMR for nonsmokers for selected occupations (n> 400) are ranked in text-figure 4. Dentists had. the lowest value ('}2) followed by carpenters (,t8) and lawyers (56). At the other end of the scale were auto mechanics (79), mail carriers (80), and doctors (82). At the lower end o[ the scale for nonsmokers in selected industries (n>'}00), ranked in text-figure 5. were the electric light and power services with an SMR of 57, the printing indust~, with 58, and legal services JNCL vor 65. NO. 5. NO%'EMBER 1~$9 Tl09781650

1168 Rogot and Murray Divi~,'on of T~I car,:er L, ar.~ caz:~- res~de~ce~ Obs Exp SMR Obs Exp S~FR New 201 244.0 82 6 ,~3.5 11 En;land MiddIe ~ 744.7 76 40 163.8 24 Atlantic EMt North ~4 E~l.6 69 31 1762 18 Cen~l W~ No~ ~ 470.6 65 19 103.4 18 ~u~ 344 4~.6 78 ~ ~.1 " 24 A~nfie E~t ~u~ 110 I~.2 71 6 ~.8 ~n~! W~t ~u~ 185 2~.5 65 11 ~.0 17 ~n~l Moun~in 97 I~.9 ~ 8 ~.9 P~ifie ~9 ~1.3 71 31 116.0 ~ide 6 9.5 ~ 0 2.1 0 U~ To~! 2,745 3,~3.0 71 175 ~4.8 21 • Obs=observ~d. exp=expected. s Divisions are states in desig~tted census areas. with 59. Higher SMR were observed for both motor vehicle manufacturing (85) and retailing (102). Text-figure 6 gives the SMR for total cancer for nonsmokers by population sire o[ place of residence at the beginning of the study. No consistent pattern was observed, and little difference by size o[ place of residence was seen. This was also true [or each of the following major cancers studied: stomach, colon, pan. c~-as, lung, prostate gland, bladder, and leukemia (dam not shown). SMR 0 20 40 60 I ~ JillI IIII ! I ! ! I ! I T~'T-~Ct'aE 2.--SMR for to',l cancer for non~,mvkers by m:jvr c~:- cupadvn: U.S. ~,etemr~ s~u:iy. jyc3. VOL. ~. .~o. 5. NOVF-MBER I__C~,3 l I I I I I I 0 20 ~0 60 SMR TLXT.HGURE 3.--SMR for total ~ncer for nonsmokers by major indumT: U.S. ~e~mns study, 1954-~9. The SMR for total cancer for nonsmokers by state of residence at the beginning of the study ranged from 32 for. Wyoming to 109 for ~ew Hampshire (text-fig. 7). For all cancers, the Mountain States had the lowest mortality ratio (58), and the New England States had the highest (82), whereas New England States had the lowe~t SMR (11) and Pacific States the highest (27) for lung cancer (table 5). DISCUSSION In this population of insured veterans, total cancer SMR 0 20 40 60 I I I 1 1 ! t 0 20 40 I I TL\"r.FgGt';~E 4.--SMR f~r t~ml cancer for nonrmoken by ~el~tmt c~cupadon: U.S. x~temns s~ly, 1954-69. Tl09781651

I I I 0 20 40 T£.'cr.nGt-&g 5.--SMI~. for total cancer for nonsmokers by ~elected industr]c U.S. veterans study, 1954-69. mortality among ~e nonsmokers was 71% of that for all respondents. Among the nonsmokers, SMK were especially low for certain workers, e.g., dentists, car- penters, and la~yers and for certain industries, e.g., electric light and power, printing, and levi services. By clivlsion or section of ~e country, SMR for non- 43* 78 103* 61 47 ~ 81 61 TE.~n'-~Gt'~t~ 7.--SMR for to~al cancer (or nonsmokers by state o[ residence: U.S. ~'eter~n.s study. 1954-69..4~te~k indicates <10 ob~en~-d c~r~:er d~ths. j~cL VOT ~. .~0. 5. .~OI;E3~ER 1~0 1167 TI0978t652

smokers were especially low for the Mountain States, with several of these States exhibiting values below 50. The key question as to how mortality levels for veterans compare with those of U.S. white males needs to be addressed. For total mortality, estimates by Dora (1) and Kahn (3) placed the veterans at about 70,~ that of the U.S. mortality levels. If total cancer mortality among veterans is also assumed to be about 70~ that for the United States, we may multiply our SMR by 0.70 to obtain a figure for our nonsmokers compared with the U.S. white males. Thus we have for total cancer mortality: 0.71 × 0.70--50%. For the major sites we would ha~: stomach, 85 × 0.70=6~,o; colon, 93 X 0.70=65%; pancreas, 81 × 0.70~-57%; lung, 21 X 0.70= 15%: prostate gland, 93 × 0.70=65%; and for leukemia, 81 x 0.70=57~. These estimates should be considered as rough ap- proximations because wc have no simple way to veri~y the accuracy of the 70.~ figure as applied to any sFecilic cause of d~th. REFERENCES (I) Do~.'q HF. The msrtality of smokers and nons'n~kers. In: Pro-- ceedings of the Social S:afistkal Se~don o[ the Arr, er~can S~a- dstlcal ,~saciadon. Washington, D,C.: Am Staz Assoc. 1955: $4-71. (2) ~. Tobacco consumption and mortality [rora cance~ and other di~ases. Pub]it Health Rep 1959:74:581-595. - (3) KA~ HA. The Dora study of smoking and mortality among U.S. veterans: Report on eight and one, half yean of tlono Natl Cancer Inst Monogr ~4) Rc;oo'r g. Smoking andgeneral mortality among U.S. veterans, 1954o1969. Washington. D.C.: U.S. Govt Print O(f, 1974 [DHEW publication No. (NIH)75-544I. ($) --. Smoking and mortallt¥ among U.S, veterans. J Chronic Dis 1974;27:189-205. (6) --. Smoking and life expectancy among U.S. veterans. Am J ~,°ublic Health 1978;68:1023-1025. JNCL VOL. 65. NO. 5. NOVF-M~ER TI09781653