Search

On the 30th anniversary of the U.S. Surgeon General's report Smoking and Health, we present updated results from one of the original cohort studies that comprised the groundbreaking document. A 26-year follow-up of 248,046 U.S. veterans evaluating the risks of cigarette smoking revealed strong dose-response effects between smoking and total cancer and a large number of cancer sites. Over 50% of cancer deaths among current smokers and 23% of cancer deaths among former smokers were attributable to cigarette smoking. These findings further demonstrate the large and unique role cigarette smoking plays in cancer etiology and the importance of smoking cessation to reduce this enormous public health burden.

—A case-control study was undertaken to investigate the role of residential radon exposure for lung cancer. The study included 210 women with lung cancer diagnosed from 1983-1986 in the county of Stockholm and 191 hospital and 209 population controls. Interviews provided information on lifet

The relationship between the use of cigarettes and other tobacco products and the risk of multiple myeloma was examined in a cohort of nearly 250,000 American veterans followed prospectively for 26 years. Compared with men who had never tobacco, the risk of death from myeloma was not increased among current (relative risk [RR] = 0.9, 95 percent confidence interval [CI] = 0.8-1.2) or former (RR = 1.0, CI = 0.8-1.3) cigarette smokers, nor among users of chewing tobacco or snuff (RR = 1.0, CI = 0.4-2.3). Risk was only slightly and nonsignificantly increased among pipe or cigar smokers (RR = 1.2, CI = 0.9-1.5). There was no indication of increasing risk with amount of tobacco used or earlier age at first use. With over 90 percent power to detect a 30 percent increased risk of this tumor occurring among current cigarette smokers, this study provides the strongest evidence to date against an association of cigarette smoking with multiple myeloma.

A cohort of 1041 men who were discharged from the hospital with a diagnosis of acromegaly were examined for subsequent cancer. With a mean follow-up time of 8.3 years, an increased rate of cancers of the digestive organs was observed (27 cases; standard incidence ratio [SIR], 2.0; 95% confidence interval [CI], 1.3 to 2.9). Rates were elevated for cancers of the esophagus (7 cases; SIR, 3.1), stomach (4 cases; SIR, 2.5), and colon (13 cases; SIR, 3.1). The increased risk of colon cancer in acromegaly is consistent with previous clinical reports and suggests opportunities for etiologic research and early cancer detection. It would seem prudent to also evaluate this risk in current research on the use of growth hormone in older individuals to increase muscle mass and reduce body fat.

Lung cancer risk in relation to indoor radon was examined in three case-control studies in Stockholm (Sweden), New Jersey (United States), and Shenyang (People's Republic of China). Year-long measurements of radon gas were made in current and past homes of 966 women who developed lung cancer and of 1,158 control women, included in the combined analysis. Nearly 14 percent of the participants were estimated to have a time-weighted, mean, radon concentration in their homes of more than 4 pCi/l (150 Bq/m3) during the period from five to 35 years prior to the date of lung cancer diagnosis (or comparable date for controls). There was a tendency for risk to increase with increasing levels of radon in NJ and Stockholm, but the trends for individual studies and overall were not statistically significant. The estimates of the excess relative risk for indoor exposure per pCi/l were 0.18 (95 percent [CI] = -0.04-0.70) in NJ, 0.06 (CI = -0.05-0.34) in Stockholm, and -0.02 (CI = -infinity-0.03) for Shenyang; these estimates did not differ significantly from each other. The overall excess RR per pCi/l was 0.00 (CI = -0.05-0.07); the confidence limits were sufficiently broad, however, that the overall estimate was still compatible with extrapolations of risks from miners. Cigarette smoking was the predominant cause of lung cancer with the RR significantly elevated in all studies. Within smoking categories, the trend in risk with increasing mean radon concentration was inconsistent. Analyses of data from several studies are complicated by the possibility that there may exist important differences in study bases which might affect results, and which may be controlled only partially through adjustment procedures. Future efforts to combine various residential studies will need to be attentive to the intrinsic limitations of studies to detect low levels of risk as well as the unique uncertainties associated with estimating, accurately, cumulative exposure to indoor radon.

The risk of childhood cancer was evaluated among 35,582 twins born in Sweden between 1952 and 1967. Cancers were identified through linkage with national cancer and mortality registries. Overall, 59 childhood cancers and 41 deaths from cancer occurred before the age of 16. For both sexes combined, the cancer incidence was similar to that in the general population of primarily single-born children (number observed/number expected [O/E] = 1.0, 95 percent confidence interval [CI] = 0.7-1.2). For males under age five, cancer incidence was reduced significantly (O/E = 0.3, CI = 0.1-0.7). There was a substantial increase in all-cause mortality among twins (O/E = 3.7), attributable to a high mortality during the first year of life. Excluding this first year, the O/E for death (all causes) was 1.1 (CI = 1.0-1.3). For cancer mortality of both sexes, the O/E was 0.9 (CI = 0.6-1.2), with no significant reduction of risk in any sex-age group. We conclude that with the possible exception of males aged 0-4 years, the childhood cancer risk of twins appears similar to that of singletons.

A 26-year follow-up of nearly 250,000 US veterans who responded to a questionnaire revealed 4,607 deaths from prostate cancer, providing the largest number of cases to date for evaluating relation to tobacco habits. Cigarette smokers had a significant increase in relative risk (RR = 1.18; 95% confidence interval 1.09-1.28) and a dose response reaching 1.51 among smokers of 40 or more cigarettes per day. Risks were elevated, but not significantly, among users of smokeless tobacco and pipe/cigar smokers. Despite limited data in the literature to support this finding, our study suggests that cigarette smoking may be related to prostate cancer, perhaps through its effect on sex hormone metabolism.

The relationship of tobacco use with risk of primary liver cancer was investigated using data from a 26-year mortality follow-up of nearly 250,000 US veterans, mostly from World War I. Significantly increased risks for liver cancer (289 deaths) were associated with most forms of tobacco use, including pipe and cigar smoking. Elevated relative risks (RRs) were seen for current cigarette smokers (RR = 2.4; 95 percent confidence interval [CI] 1.6-3.5) and former cigarette smokers (RR = 1.9, 1.2-2.9). A strong dose-response relationship (P less than 0.001) was found for cigarette smoking, with smokers of 40 or more cigarettes per day having almost a fourfold risk (RR = 3.8, 1.9-8.0). Risks were also found to increase significantly with years of cigarette use and with earlier age at the start of cigarette smoking. These results are consistent with those of other cohort and case-control studies, suggesting that cigarette smoking may be related to the risk of liver cancer.

The association between X-ray exposure in utero and childhood cancer was studied in a case-control study, nested in a national cohort of Swedish twin births. Records of the Swedish Twin Register for those born 1936-1967 (n = 83,316) were linked to the Swedish Cancer Register (1958-1983) and the Swedish Cause-of-Death Register (1952-1983). Ninety-five cases of childhood cancer before the age of 16 were identified. Two controls from the Twin Register were matched to each case by sex and year of birth. Maternal X-rays during pregnancy were documented for 41% of the cases and 36% of the controls. The crude relative risk of any childhood cancer associated with any prenatal X-ray was 1.2 (95% CI; 0.7-2.1). For the cases 64% of the X-rayed women had had abdominal X-rays, the numbers for the controls were 57%. Fifty-nine per cent of the X-rayed women had had abdominal X-rays, which were associated with a relative risk for all cancers of 1.4 (Cl; 0.8-2.5), leukemias 1.7 (Cl; 0.7-4.1) and tumors of the central nervous system 1.5 (Cl; 0.5-4.2). There was no apparent confounding by mother's age, drug use, obstetric complications, previous miscarriages, social class or length of pregnancy. The observed relative risks of cancer following prenatal X-ray exposure are consistent with previous studies, suggesting that the developing fetus may be more sensitive to the carcinogenic effects of ionizing radiation than are children irradiated post-natally.

Reports from the United Kingdom have described increases in leukemia and lymphoma among young persons living near certain nuclear installations. Because of concerns raised by these reports, a mortality survey was conducted in populations living near nuclear facilities in the United States. All facilities began service before 1982. Over 900,000 cancer deaths occurred from 1950 through 1984 in 107 counties with or near nuclear installations. Each study county was matched for comparison to three "control counties" in the same region. There were 1.8 million cancer deaths in the 292 control counties during the 35 years studied. Deaths due to leukemia or other cancers were not more frequent in the study counties than in the control counties. For childhood leukemia mortality, the relative risk comparing the study counties with their controls before plant start-up was 1.08, while after start-up it was 1.03. For leukemia mortality at all ages, the relative risks were 1.02 before start-up and 0.98 after. For counties in two states, cancer incidence data were also available. For one facility, the standardized registration ratio for childhood leukemia was increased significantly after start-up. However, the increase also antedated the operation of this facility. The study is limited by the correlational approach and the large size of the geographic areas (counties) used. It does not prove the absence of any effect. If, however, any excess cancer risk was present in US counties with nuclear facilities, it was too small to be detected with the methods employed.

Questionnaire data on about 1200 male twin pairs from the Registry at the Karolinska Institute, Stockholm, and on about 4000 male twin pairs from the Registry of the National Research Council, Washington, DC, have been used to study factors affecting angina pectoris. An operational definition of "angina pectoris" was developed from the questionnaire. In the available data, alcohol drinking, lack of exercise, frequent change of employer, low occupational adjustment and smoking are moderately but significantly related to angina among individuals (disregarding twin relationships) in both Sweden and the US. In monozygous US twin pairs discordant for the above variables, significantly different rates of angina appear only with alcohol drinking. In discordant dizygous US twin pairs, significantly different rates of angina appear with alcohol drinking and with low occupational adjustment. Of the independent variables only smoking and drinking are appreciably associated with each other. These findings suggest that alcohol drinking and to a lesser extent occupational adjustment are related to angina directly and not through their association with other factors such as age, genetic background, smoking, physical exercise and early environment.

Career radiation doses for 8,961 male workers at the Calvert Cliffs Nuclear Power Plant (CCNPP) were determined for both utility (n = 4,960) and contractor (n = 4,001) employees. Workers were followed from the time of first employment at CCNPP (including plant construction) to the end of 1984 (mean follow-up = 5.4 y). Plant operation began in 1975. The mean duration of employment was 1.9 y at CCNPP and 3.1 y in the nuclear industry. Career radiation doses were determined from dosimetry records kept by the utility company and the U.S. Nuclear Regulatory Commission (NRC). For all exposed workers, the average career dose was 21 mSv and was higher for contractor (30 mSv) than utility (13 mSv) workers. Career doses were also higher among those employed in the nuclear industry for greater than or equal to 15 y (111 mSv) and among workers classified as health physicists (56 mSv). Cumulative doses of greater than or equal to 50 mSv were received by 12% of the workers; the maximum career dose reported was 470 mSv. The availability of social security numbers for practically all employees facilitated record-linkage methods to determine mortality; 161 deaths were identified. On average the workers experienced mortalitymore » from all causes that was 15% less than that of the general population of the U.S., probably due to healthier members of the population being selected for employment. Our investigation demonstrates that historical information is available from which career doses could be constructed and that, in principle, it is feasible to conduct epidemiologic studies of nuclear power plant workers in the U.S. Although difficult, the approach taken could prove useful until such time as a comprehensive registry of U.S. radiation workers is established.« less

A cohort of nearly 300,000 insured veterans (Dom Cohort), experienced a much greater percent decline in CHD death rate over the period, 1954–1979, than the population of the U.S., while for stroke, the percent decline in death rate was virtually the same as the U.S. For CHD, greater percent declines were noted over the study period for non-smokers compared to cigarette smokers, for professionals compared to non-professionals and for persons with high socioeconomic scores (SES) compared to those with low scores. In each group, younger persons experienced greater percent declines than older persons. For stroke, non-smokers experienced a somewhat greater percent decline in rate than smokers but this did not hold true for all age groups. Unlike CHID, professionals experienced a smaller percent decline in their stroke death rate than non-professionals, as did persons with high SES compared to those with low SES. The contradictory patterns observed for the two diseases with respect to occupation and SES suggest that the risk factors for stroke and coronary heart disease are not exactly the same. Throughout, the findings were much more convincing for CHD than for stroke.

When the Weinberg estimate of the proportion of monozygotic pairs is quite deviant from that in the source population, it is likely to be wrong because Weinberg's difference is much less stable than the zygosity proportions. A formula is proposed for the probability distribution of possible compositions of a small sample of twins based on sex concordance in the sample and zygosity proportions in the source population.

Medical histories of the 15,924 male twin pairs in the National Academy of Sciences-National Research Council Twin Registry were examined to determine, within pairs, concordances for alcoholism and its medical end points. Prevalences per 1,000 among individual twin subjects were 29.6 for alcoholism, 4.1 for alcoholic psychosis, 14.2 for liver cirrhosis, and 2.1 for pancreatitis. Prevalences were similar to monozygotic (MZ) and dizygotic (DZ) twins. Prevalences in percent among co-twins of diagnosed subjects, that is case-wise twin concordance rates, were, respectively, by diagnosis: alcoholism: 26.3 (MZ), 11.9 (DZ); alcoholic psychosis: 21.1 (MZ), 6.0 (DZ); and liver cirrhosis: 14.6 (MZ), 5.4 (DZ). No twin pairs concordant for pancreatitis were found. The greater concordance for alcoholic psychosis and for liver cirrhosis among MZ than DZ twins could not be explained by the difference in alcoholism concordance between them. The difference in concordance between MZ and DZ twins persisted when, in addition, it was assumed that only half of the actually occurring cases of alcoholism and of each of the end points have been ascertained. These results provide evidence in favor of genetic predisposition to organ-specific complications of alcoholism and should serve to stimulate searches for the underlying biochemical mechanisms.

Subjects in the National Academy of Sciences-National Research Council Twin Registry of 31,848 male twin veterans were followed for mortality from 1 January 1946, or from the date of entry into military service if that was later, to 31 December 1978. During this time 3,573 deaths occurred among them, 837 due to trauma and 2,712 due to disease. Mortality from all causes for the entire follow-up period was 10.2% among 11,350 monozygotic (MZ) twins and 11.4% among 14,450 dizygotic (DZ) twins. Mortality of veterans is known to be favorable compared to U.S. males. Among U.S. males of the same ages as the two respective twin zygosity groups, a mortality of 13.9% would have been expected during this time period. Observed mortality from trauma was 2.3% for MZ twins and 2.5% for DZ twins, with 3.0% expected in either group. Observed mortality from all disease was 7.9% for MZ twins and 8.8% for DZ twins, with 10.9% expected in either group. For total mortality, the case twin concordance rates, based on individual deaths, were 28.2% among MZ twins and 17.7% among DZ twins. For trauma, respectively by zygosity, these concordance rates were 6.9% and 3.9%. In this sample, familial factors appear to be of little consequence in trauma deaths. For all disease the concordance rates were 30.1% and 17.4%. Estimating heritability of liability to death from disease, as proposed by Edwards (1969), provides values of h2 = r = 0.51 for MZ twins, h2 = 2r = 0.48 for DZ twins, and h2 = 2(rMZ-rDZ) = 0.54 using data for the two zygosity groups combined.

A study was carried out in 1095 first Army hospital admissions for lumbar herniation of the nucleus pulposus (HNP) who were individually matched on age and period of service during World War II to holders of Army National Service Life Insurance policies. For both the cases and the comparison group, data were obtained from military records, particularly records of induction into service. Factors showing a statistically significant positive association (p less than 0.05) with admission for HNP in the entire sample and also among enlisted cases matched to enlisted members of the comparison group were: occupation of craftsman or foreman, married status, rural residence, excess height, excess weight, heavy frame, good posture, defects relating to back or legs, military occupation specialty of ground combat, and rank of sergeant or staff sergeant. Factors showing a negative association with admission for HNP in these same groups were: clerical occupation, two or more battle stars earned and officer rank. Essentially the same relationships were found among cases with surgically confirmed diagnoses of HNP, less than 30 years old at hospital admission, and with recent onset of symptoms, when the cases in these groups were compared with their matches in the comparison group. Mechanical factors related to body build as evaluated by height and weight measurements appear to be of significance in the etiology of HNP. Occupational factors also appear to be of importance. Some cases and some controls were hospitalized prior to the first hospital admission of the case for a disease possibly related to HNP. For that time period there were no remarkable differences between cases and controls in service hospital diagnoses, including those of trauma.

Three methods of measuring SSSRn concentration in homes were compared. In 91 dwellings, measurements were made with thermoluminescence dosimeters (TLDs) for two weeks during the winter, with alpha track Track Etch type SF dosimeters (ATDs) for six months during the heating season and for one year. There was a high correlation between the TLD and the six-month ATD values (r = .83) and between the six-month and the one-year ATD values (r = .80). On the average, however, the arithmetic and geometric means of the ATD (six-month) values were approximately 50 Bq m-3 higher than the TLD values (p less than .0001). The six-month ATD values also were, on the average, approximately eight percent higher than the one-year values. This difference is probably caused by decreased ventilation during the colder part of the year. The results indicate that the measurement methods under study may be of value for exposure assessments in epidemiologic studies, but also that estimates of risk per unit radiation exposure should be interpreted with caution.

In a survey having a 90 percent chance of detecting a quadrupling of risk, no significant increase in leukemia has been observed in the offspring of persons exposed to atomic bomb radiation in Hiroshima and Nagasaki. In the children with leukemia who were born to previously irradiated parents, there was no unusual finding in regard to the interval from parental exposure to onset, age at onset, or type of leukemia. Failure to demonstrate a radiation effect does not mean that there is no such effect. It merely means that within the experience thus far open to observation, no effect is to be seen. There now is sufficient experience to justify a long period of observation of the F1 and subsequent generations for a possible increase of leukemia.

A follow-up study was performed on 690 white diabetic patients 2 to 5 years after their blood had been analyzed for lipoprotein and cholesterol to determine if these lipid measures had any predictive value for development of future cardiovascular complications. The work was done in parallel to the national project in which 15,000 normal persons were followed for 1 to 2 years. 5

Survival of first admissions to Army hospitals for coronary heart disease occurring from June 1943 to December 1944 was evaluated in a sample of 1495 white males who were alive 6 months after the date of admission. Included in this study are only cases whose hospital diagnosis was confirmed by an independent review of the history, laboratory data, and ECG recordings in their service medical records. Physicians were found to have a better prognosis than other occupations. When physicians were excluded, professional men with college education had a somewhat better survival than other occupation or education groupings. Physicians were more likely to have the diagnosis of angina pectoris or coronary insufficiency than others; however, their favorable prognosis was also found within diagnostic groups with specified ECG findings. The much better long-term survival of physicians may give some measure of the extent to which early detection and appropriate medical regimens can improve prognosis in coronary heart disease if followed by optimally motivated patients.

The relation between passive smoking and lung cancer was examined by means of a case-control study in a cohort of 27,409 nonsmoking Swedish women identified from questionnaires mailed in 1961 and 1963. A total of 77 cases of primary carcinoma of the bronchus or lung were found in a follow-up of the cohort through 1980. A new questionnaire in 1984 provided information on smoking by study subjects and their spouses as well as on potential confounding factors. The study revealed a relative risk of 3.3, constituting a statistically significant increase (p less than 0.05) for squamous cell and small cell carcinomas in women married to smokers and a positive dose-response relation. No consistent effect could be seen for other histologic types, indicating that passive smoking is related primarily to those forms of lung cancer which show the highest relative risks in smokers.

Men drafted into the Army, hospitalized during 1944–1945 for service-connected trauma to the extremities, and consequently separated for disability were followed for mortality from January 1946 to April 1977. Three groups were established consisting of those whose injury resulted in: 1. (a) limb amputation; 2. (b) disfigurement without loss of body part; 3. (c) loss of part of hand or part of foot. Group (a) had a mortality, standardized for age and calendar time, 1.4 times that of group (b), matched on age and length of service at admission, and 1.3 times that of group (c), similar in age and length of service to group (a). The excess mortality of limb amputees was statistically significant (p

Daily symptom rates in patients with chronic obstructive pulmonary disease and in other subjects with presumed high sensitivity to air pollution who lived near a coal-fired plant were compared with 24-hr ambient air concentrations of nitrogen dioxide, sulfur dioxide, soot, and suspended particles as well as with emissions from the plant. The mean concentrations of each of the pollutants during the 4-month study period were below 30 micrograms/m3, and no single 24-hr concentration exceeded 100 micrograms/m3. There were no consistent associations between plant emissions and pollutant levels or between these two variables and daily symptom rates. The results indicate that the coal-fired plant was not of major importance for the occurrence of acute respiratory symptoms in the surrounding population.

An epidemiologic study was carried out on 1,393 cases of CHD in World War II Army men, median age 39.6 years, by comparing these cases with representative age-matched Army controls using data from military records of both groups, particularly records of induction into service. Factors showing significant (P Other characteristics more prevalent among the coronary cases than among the matched controls and therefore associated with a higher risk of CHD were: geographical residence in the Middle Atlantic States, a higher socioeconomic score, some graduate education, being an officer, being of Jewish religion, and having a blood Group A. A significantly lower risk of CHD was found for rural birthplace and previous outdoor occupation requiring high physical activity. Men whose previous occupation was that of a farmer or farm laborer had a relative risk of 0.54 (P

Recent trends in chronic obstructive pulmonary disease (COPD) mortality are described for a cohort of nearly 300,000 insured veterans who were followed for survival from 1954 through 1979. Death rates from COPD for the veterans by smoking status were compared with U.S. death rates for white males 55 to 84 yr of age over the same period. The main findings were that the veterans experienced sharp increases in COPD mortality over the 26-yr study period, much like the U.S. findings. However, the observed increases were not as great as those for the United States. The COPD rates for veterans who were cigarette smokers showed the same pattern as for all veterans. Some important differences by age were noted. Thus, for ages 60 to 64 and 65 to 69, the direction of the trend among veterans appeared to change from initial increases in the COPD death rate to later decreases while at the same time the three oldest age groups (70-74, 75-79, and 80-84) continued to experience increases in mortality over the whole period. These findings were similar to those for the United States. Several possible explanatory factors for these changes are discussed.

Concentrations of 226Ra were determined in native vegetation and underlying substrate (soil and tailings) at various sites around a conventional open-pit, acid leach U production operation in Wyoming. Plant/soil concentration ratios (CRs) for 226Ra were estimated for various sites, including weathered tailings; a tailings impoundment shoreline; downwind from exposed tailings; a mine overburden reclamation area; and several background locations. Radium-226 concentrations for vegetation and substrate and CR values from the perturbed sites were elevated above background. The highest vegetation concentration (1.3 Bq g-1) was found in a grass which had invaded exposed, weathered tailings. Levels of 226Ra in soil and vegetation and CR values decreased with distance from the tailings impoundment edge. CR values varied significantly among sites, but few differences were found between plant species groups. The observed CR values ranged from 0.07 at the background and reclamation areas to 0.4 downwind from the tailings area. Average CR values for plants growing on exposed tailings and within one meter from the impoundment edge were 0.15 and 0.3, respectively. CR values of 226Ra for plants on tailings substrates were comparatively low in contrast to other radionuclides in the U chain. We speculate that in the case of sulfuric acid leachedmore » tailings-derived material, 226Ra is sequestered as sulfate, which is highly insoluble relative to the sulfates of the other elements (e.g., U and Th) resulting in reduced availability for plant uptake.« less

A group of 286 adolescent children, all of whom were in utero at the time of the atomic bombing in Nagasaki, Japan, were examined as part of a long-term program to determine possible differences in growth and development that might be attributable to ionizing radiation. Three comparison groups were studied: group 1, children whose mothers were within 2,000 meters of the hypocenter (point on the ground above which the bomb exploded); group 2, whose mothers were located between 3,000 to 4,999 meters; and group 3, whose mothers were not in the city at the time of the bomb. The age at menarche and degree of epiphyseal closure in the wrist were determined. In addition, measurements were made of head circumference, standing and sitting heights, weight, and chest circumference. Differences were found which were consistent with a radiation effect. This was most significant in head circumference measurements.

The relationship of chronic respiratory symptoms and residence-associated air pollution was evaluated through use of questionnaire data on twins in the National Research Council Twin Registry. Estimates of air pollution were based on residential and work place history. Two different scores were developed. One score evaluated residence only in terms of an urban-rural continuum and the other provided quantitative estimates of air pollution in postal Zone Improvement Program locations. Among individuals, respiratory symptoms were found to be related to urban residence as measured by the first score, but not to the more specific measure of air pollution exposure. Smoking, the drinking of alcohol and socioeconomic status were also positively associated with respiratory symptoms. No significant differences were noted between members of twin pairs if one twin maintained a more urban residence than the other or if one twin had a higher measure of exposure to air pollution than the other. The latter comparisons us...

1. 1. Survival tables over a 15 year period are presented for white males between the ages of 20 through 64 who were diagnosed to have a first clinical occurrence of coronary heart disease while in Army service. Diagnostic groupings for this analysis were separated into two categories: myocardial infarction-coronary thrombosis-coronary occlusion and angina pectoris-coronary insufficiency. 2. 2. Both the immediate prognosis for survival and the prognosis over the 15 year period were less favorable for cases initiating as myocardial infarction-coronary thrombosis-coronary occlusion, than for cases initiating as angina pectoris-coronary insufficiency. 3. 3. The first 24 hours after onset of clinical manifestations of either group of coronary disease syndromes was the period of the highest concentrated mortality rate. 4. 4. At all ages the myocardial infarction-coronary occlusion-coronary thrombosis group had a higher mortality rate (16.7 per cent) in the first year of follow-up than in any subsequent year. After the first year, the annual mortality rate in this group was about 5 per cent and did not vary much with length of follow-up. In the angina pectoris-coronary insufficiency group, the annual mortality rate was about 3 per cent throughout the follow-up period. 5. 5. Comparison of the observed mortality rate in the infarction-thrombosis-occlusion group with that expected for American males of the same age and race in the same calendar period reveals a steady decrease in the mortality ratio (observed/expected) from about 30 in the first year to approximately 2 in the interval 11–15 years after diagnosis. In both diagnostic groups, mortality ratios of older patients were lower than those of younger patients.

A Cox proportional hazard regression analysis was carried out that evaluated age-specific death risk among 21,890 twins born in Sweden during 1886 through 1925 and followed during 1962 through 1980. Cotwin's survival was used as the primary covariable, and auxiliary covariables were smoking, marital status and, among men, police registration for alcohol abuse. In each age, sex and zygosity group, except the oldest DZ males, cotwin's mortality had a significant, independent, positive relationship to the mortality risk of the individual. The auxiliary covariables, except marital status among females, had significant, independent, positive relationships to mortality among the youngest twins of both zygosity groups and in the middle age group of MZ twins. In the oldest age group, the death of MZ cot wins was the only variable significantly related to the individual's mortality. Heritability estimates for the age-specific probability or death risk, developed by different methods for different analysis groups, range between 0.4 and 0.6. They have reasonable internal consistency, are not much affected by the covariates, and are in agreement with other studies that did not control covariates.

Height, weight, and body mass index (BMI) were assessed in a sample of 1974 monozygotic and 2097 dizygotic male twin pairs. Concordance rates for different degrees of overweight were twice as high for monozygotic twins as for dizygotic twins. Classic twin methods estimated a high heritability for height, weight, and BMI, both at age 20 years (.80, .78, and .77, respectively) and at a 25-year follow-up (.80, .81, and .84, respectively). Height, weight, and BMI were highly correlated across time, and a path analysis suggested that the major part of that covariation was genetic. These results are similar to those of other twin studies of these measures and suggest that human fatness is under substantial genetic control.

In Reply.— We appreciate Dr Knight's pointing out the importance of context in interpreting the findings of our twin study of obesity. As he notes, concordance rates and heritability estimates are critically dependent on the environment in which they are measured, and different environments can give rise to different estimates of heritability. The role of environment is particularly important in assessing obesity, in which gene-environment interaction must surely be of great importance. And as he (and we) have noted, such interaction is not measured in twin studies. This limitation makes it important to assess the role of genetics in human obesity by methods that can assess gene-environment interaction. Adoption studies constitute one such method, and a recent adoption study 1 has confirmed the importance of genetic factors in human obesity. The data of our study do not, as Dr Knight says, support general conclusions regarding the relative roles of heredity