Minnesota Heart Survey Research Articles

Sex differences in coronary heart disease mortality trends: the Minnesota Heart Survey, 1970-1988.

Recent U.S. national statistics indicate a slowing of the rate of decline in coronary heart disease (CHD) deaths among women. To examine recent sex-specific rates of mortality decline in the Minnesota Heart Survey, we computed the average annual percentage change in age-adjusted CHD death rate for the periods 1970-1978 and 1979-1988. We fit a log-linear regression model to the yearly CHD death rates, with separate sex-specific slopes estimated simultaneously for the two calendar periods. The average annual percentage decline in age-adjusted CHD death rate among men was slightly greater in the period 1979-1988 [4.8%; 95% confidence interval (CI) = 5.2-4.3] than in the period 1970-1978 (3.9%; 95% CI = 4.6-3.1). In contrast, among women, the rate of decline was less in the period 1979-1988 (3.6%; 95% CI = 4.8-2.3) than in 1970-1978 (4.8%; 95% CI = 5.6-4.0). We found a slowing of the rate of decline in out-of-hospital CHD deaths in both men and women. We observed a large increase in the rate of decline in the CHD death rate in hospital among men in the period 1979-1988 (8.3%; 95% CI = 9.3-7.2), compared with the period 1970-1978 (2.4%; 95% CI = 3.2-1.7). This increase did not occur among women (4.1%; 95% CI = 5.7-2.5 in 1970-1980 vs 3.8%; 95% CI = 4.9-2.7 in 1970-1978); this difference is a major factor in the steeper recent decline in overall CHD mortality among men compared with women.

Trends in Mortality, Morbidity, and Risk Factor Levels for Stroke From 1960 Through 1990

The Minnesota Heart Survey is a population-based study designed to monitor and explain trends in cardiovascular mortality, morbidity, and risk factors. Surveillance time-trends study. The following trends were examined among men and women aged 25 to 74 years living in Minneapolis-St Paul, Minn: (1) stroke mortality from 1960 through 1990; (2) risk factors in population-based surveys conducted in 1973 through 1974, 1980 through 1982, and 1985 through 1987; and (3) morbidity in a 50% sample of hospitalized discharges for acute-stroke in 1970, 1980, and 1985. Stroke mortality in Minneapolis-St Paul declined slowly from 1960 through 1972 (average fall, 2.4% per year), dropped sharply from 1972 through 1984 (average fall, 6.5% per year), but exhibited little change thereafter (average fall, 1.5% per year). The average level of cardiovascular disease risk factors fell from 1973-1974 to 1985-1987, with the exception of body mass index. In particular, hypertension diagnosis, treatment, and control levels improved substantially between 1973-1974 and 1980-1982, although there was little improvement after 1980-1982. While discharge rates for hospital-coded acute stroke declined substantially between 1970 and 1985 in both sexes, no clear trend was observed in definite stroke rates as validated using standard clinical criteria. Twenty-eight-day case fatality rates of definite stroke improved significantly from 1970 to 1985. The substantial decline in stroke mortality of more than 50% from 1960 through 1990 appears to have been attributable to both primary and secondary prevention. These data suggest that the long decline in stroke mortality and morbidity in Minneapolis-St Paul has plateaued, although improved detection of stroke with computed tomography prevents an unequivocal conclusion.

Trends in mortality, morbidity, and risk factor levels for stroke from 1960 through 1990. The Minnesota Heart Survey

Trends in mortality, morbidity, and risk factor levels for stroke from 1960 through 1990. The Minnesota Heart Survey

Menopause and serum cholesterol: differences between blacks and whites. The Minnesota Heart Survey.

The relation between menopause and serum total and high-density-lipoprotein cholesterol was examined by the Minnesota Heart Survey in a cross-sectional, population-based study of 344 black women and 474 white women aged 35-54 years from the Twin Cities metropolitan area in 1985-1986. Analysis of covariance was used to examine differences in serum total and high-density-lipoprotein cholesterol in black women and white women by menopausal status, adjusting for the effects of age, educational level, cigarette smoking, body mass index, exercise, alcohol consumption, diabetes mellitus, sex hormone, beta blocker, and diuretic use. Among whites, adjusted serum total cholesterol was 13 mg/dl higher in postmenopausal than in premenopausal women (p less than 0.002). Black postmenopausal women had slightly higher serum total cholesterol than did their premenopausal counterparts (5.4 mg/dl). However, this was not statistically significant. An interaction term in a linear regression model confirmed a racial difference in the total cholesterol association with menopause (p less than 0.02). The higher total cholesterol levels observed in white postmenopausal women were mainly among those with natural menopause (20.7 mg/dl higher than premenopausal, p less than 0.0003) and those with a hysterectomy and at least one intact ovary (11.0 mg/dl higher, p = 0.05). Among black women, only the subgroup with a hysterectomy and a bilateral oophorectomy had a significantly higher serum total cholesterol (19.9 mg/dl higher than premenopausal, p less than 0.05). There was no significant association between high-density-lipoprotein cholesterol and any type of menopause in either black women or white women. Our findings may reflect a true physiologic difference in the relation between menopause and serum total cholesterol between American blacks and whites. The lack of a significant association between menopause and high-density-lipoprotein cholesterol in either race raises the possibility that menopause may not affect atherosclerosis risk via reduced high-density-lipoprotein cholesterol.

Community trends in CPR training and use: The Minnesota heart survey

To examine community changes in self-reported CPR training and use from 1980-82 to 1985-87 using data obtained from the Minnesota Heart Survey. A comparative investigation of CPR training among blacks and whites in 1985-86 also was completed. Data were obtained in 1980-81, 1981-82, 1985-86, and 1986-87 from four population-based samples drawn from the seven-county Minneapolis-St Paul metropolitan area. To increase sample sizes and to compare prevalences of CPR training and use in the early 1980s with prevalences in the mid-1980s, the four Minnesota Heart Survey surveys were combined into two time periods, 1980-82 and 1985-87. A separate survey of black individuals was conducted in 1985, and these data were used in the comparisons between blacks and whites in 1985-86. The prevalence of whites trained in CPR increased significantly between 1980-82 and 1985-87 in both nonhealth professionals (18.5% vs 30.9%) and health professionals (71.9% vs 86.8%). No significant change was observed between the two periods in the percentage of nonhealth professionals who had ever used their CPR skills (9.7% vs 10.7%), whereas use among health professionals increased significantly (40.2% vs 53.4%). Training within the prior two or three years decreased from 1980-82 to 1985-87 among nonhealth professionals, but increases in recent training were observed among health professionals. There were no significant differences between black and white nonhealth professionals in the prevalence of CPR training. Black trainees, however, reported a higher percentage of ever using CPR skills than white trainees (15.4% vs 9.8%, respectively). Black trainees also had higher rates of recent CPR training than white trainees. No differences were observed between black and white health professionals regarding CPR training and use, or recency of certification. These results suggest that the percentage of individuals trained in CPR is increasing. Improvement is needed, however, in the rates of recent certification among nonhealth professionals.

Trends in survival of hospitalized myocardial infarction patients between 1970 and 1985. The Minnesota Heart Survey.

The Minnesota Heart Survey is a population-based study designed to monitor and explain trends in cardiovascular mortality, morbidity, and risk factors. As part of this effort, a 50% sample of patients hospitalized for myocardial infarction (MI) in the seven-county Twin Cities (Minneapolis and St. Paul) metropolitan area was reviewed in 1970, 1980, and 1985. Those with a validated definite MI were followed for 4-year mortality. The purpose was to determine whether the improved survival observed between 1970 and 1980 was extended to the 1980-1985 period. Crude 28-day mortality in men changed from 18% in 1970 to 12% in 1980 to 13% in 1985; in women it changed from 27% in 1970 to 22% in 1980 to 18% in 1985. After adjustment for severity factors (e.g., age, previous MI, and admission heart rate and systolic blood pressure), 28-day mortality was significantly lower in 1980 than in 1970 in men (RR, 0.66; 95% CI, 0.47, 0.92) and in women (RR, 0.69; 95% CI, 0.46, 1.04), but no change occurred from from 1980 to 1985 (p greater than 0.25). After adjustment for severity indicators, 4-year survival was better in 1980 than in 1970 for men (RR, 0.67; 95% CI, 0.54, 0.83) and for women (RR, 0.72; 95% CI, 0.54, 0.98), but there was no significant change from 1980 to 1985 (p greater than 0.25). These results suggest that improvements in survival among hospitalized MI patients contributed to the overall decline in coronary heart disease mortality in the Twin Cities area between 1970 and 1980 but not between 1980 and 1985.

Genetic markers associated with high density lipoprotein cholesterol levels in a biracial population sample.

An explanation for the consistently documented finding of higher levels of high density lipoprotein-cholesterol in black men relative to white men was sought by comparing the frequency of restriction fragment length polymorphism markers present in blacks and in whites at the gene loci coding for the two major apolipoprotein constituents of high density lipoprotein, apolipoproteins AI and AII. The measurements were made in population-based samples of 45 to 54-year-old black (n = 190) and white (n = 370) subjects from the Minnesota Heart Survey for whom lipoprotein levels were available. The mean high density lipoprotein-cholesterol level for black men in the sample (47 +/- 1.5 mg/dl) was higher (P less than 0.05) than that for white men (42 +/- 0.9 mg/dl), while levels in women were not different between races. While the SacI and MspI markers at the apolipoprotein AI-CIII-AIV gene locus showed similar frequencies in blacks compared to whites, the degree of the linkage disequilibrium previously noted between these markers in white subjects was altered in blacks and the minor allele of the PstI marker at this locus was virtually absent in the black subjects (P less than 0.005 vs whites). For black men, there were significant associations of the M2 allele and the S2M2 haplotype at the apolipoprotein AI locus with lower high density lipoprotein-cholesterol levels. The results are consistent with the hypothesis that DNA sequence variations in the vicinity of the apolipoprotein AI-CIII-AIV gene locus are associated with the difference in high density lipoprotein-cholesterol levels between blacks and whites.

Trends in Prevalence of Diabetes Mellitus in Patients with Myocardial Infarction and Effect of Diabetes on Survival: The Minnesota Heart Survey

The purpose of this study was to document trends in the prevalence of diabetes among men and women hospitalized for myocardial infarction (MI) and to determine the effect of diabetes on in-hospital case fatality rates and long-term survival. The Minnesota Heart Survey is a population-based surveillance system that has monitored trends in coronary heart disease morbidity since 1970. As part of this effort, a 50% random sample of acute MI discharge records in Minneapolis-St. Paul metropolitan area hospitals was abstracted in 1970, 1980, and 1985. The prevalence of diabetes among MI patients was compared over time, and the data indicated a significant increase between 1970 and 1985 in both men (8.2 vs. 16.8%, P less than 0.001) and women (16.0 vs. 25.8%, P = 0.01). Diabetic individuals had an odds ratio of in-hospital death after an MI 1.5 times that of nondiabetic individuals (P less than 0.01) after controlling for the effects of sex, age, and year of MI. Among discharged MI survivors, the risk of death was 40% higher (P less than 0.01) in diabetic individuals than nondiabetic individuals after 6 yr of follow-up. Compared with nondiabetic individuals, diabetic individuals appeared more likely to have cardiac (pump) failure with acute MI. Our findings suggest that the risk of coronary heart disease morbidity and mortality attributable to diabetes may be increasing over time. Therefore, clinicians need to take extra care in the management of MIs in diabetic individuals, and public health efforts to reduce diabetes prevalence are warranted.

Secular trends in Q wave and non-Q wave acute myocardial infarction. The Minnesota Heart Survey.

The Minnesota Heart Survey examined trends of Q wave and non-Q wave acute myocardial infarction (AMI) using a 50% random sample of all hospital discharges of patients with AMI or another acute coronary disease from 35 of 36 hospitals in 1970 and 30 of 31 hospitals in 1980 in the Minneapolis-St. Paul metropolitan area. A total of 1,901 and 1,864 potential AMI cases were abstracted in 1970 and 1980, respectively. Electrocardiograms were coded according to the Minnesota code. AMIs were validated by computerized algorithm based on chest pain, enzymes, electrocardiograms, and autopsy. This study shows that with the use of a consistent, standard diagnostic algorithm, attack rates for Q wave AMI did not change significantly between 1970 and 1980 and that attack rates for non-Q wave AMI decreased significantly during the same decade. However, when the more sensitive cardiac enzymes creatine phosphokinase and creatine phosphokinase-MB were considered, attack rates of both Q wave and non-Q wave AMIs increased. This research documents four important trends for community AMI rates that are at variance with those reported by others. There was a decline in non-Q wave AMI attack rates from 1970 to 1980; women had outcomes equal to or worse than those for men for both case-fatality and 7-year survival rates; patients with non-Q wave AMIs had worse in-hospital prognoses than those with Q wave AMIs; and 7-year survival rates were worse for Q wave AMI in 1980. These findings demonstrate the need for standard diagnostic criteria for Q wave and non-Q wave AMI if trends are to be monitored. In the future, as new trials of operative and nonoperative therapies of AMI are undertaken, these considerations will increase in importance.

Differences in leisure-time physical activity levels between blacks and whites in population-based samples: The Minnesota heart survey

Energy expenditure in leisure-time physical activity (LTPA) was measured using the Minnesota LTPA Questionnaire in 35- to 74-year-old black and white residents of Minneapolis-St. Paul, Minnesota. Estimates of the geometric mean LTPA energy expenditure were 129 and 204 kcal per day for black and white men (p less than .05) and 91 and 123 kcal per day for black and white women (p less than .05). The percentage of individuals expending 2000 kcal or more per week in LTPA was significantly lower in black men than white men (25 vs. 35%; p = .01) but was not different in black versus white women (18 vs. 17%). Although black men and women reported greater occupational physical activity than their white counterparts, LTPA and job activity were unrelated in all race and sex groups. In both races, LTPA energy expenditure declined with age. LTPA increased with level of formal education, and the largest LTPA difference between blacks and whites was observed in those who had a high-school diploma or less. Blacks had lower participation rates than whites in most of the individually assessed physical activities. Additional research is needed on the determinants and promoters of LTPA in population subgroups.

ACCURACY OF DEATH CERTIFICATE DIAGNOSIS OF INTRACRANIAL HEMORRHAGE AND NONHEMORRHAGIC STROKE

The validity of death certificate diagnosis of stroke and its type as the underlying cause of death was investigated in a sample of in-hospital deaths of possible stroke cases from the Minnesota Heart Survey. The 228 in-hospital deaths in 1970 and the 180 deaths in 1980 had a stroke diagnosis either on hospital discharge records or as the underlying cause of death on the death certificate. Relative to a standardized physician diagnosis, positive predictive values for the death certificate diagnosis in 1970 were 96% for all types of stroke, 59% for intracranial hemorrhage, and 87% for nonhemorrhagic stroke. The respective values in 1980 were 100%, 82%, and 97%. An increase in positive predictive values, particularly for intracranial hemorrhage between 1970 and 1980, was attributed to the increased use of computerized tomography. Sensitivity for the death certificate diagnosis in 1970 was 63% for all types of stroke, 66% for intracranial hemorrhage, and 45% for nonhemorrhagic stroke. The respective sensitivities in 1980 were 70%, 76%, and 58%. The lower sensitivity for nonhemorrhagic stroke as compared with hemorrhagic stroke was due in part to 1) frequent reporting of nonhemorrhagic stroke as a contributing cause of death rather than the underlying cause of death and 2) time from stroke onset to death. Specificity among these possible strokes was high in both years. The low sensitivity of death certificate diagnosis of stroke may reduce estimated relative risks in epidemiologic studies. Nevertheless, since the advent of widespread use of computerized tomography, a death certificate diagnosis of intracranial hemorrhage versus nonhemorrhagic stroke appears to be sufficiently accurate for use in epidemiologic studies of stroke etiology.

CONTINUED DECLINE IN CARDIOVASCULAR DISEASE RISK FACTORS: RESULTS OF THE MINNESOTA HEART SURVEY, 1980–1982 AND 1985–1987

Systematic, simultaneous surveillance of cardiovascular disease risk factors, morbidity, and mortality is ongoing in the Minnesota Heart Survey. Risk factors for cardiovascular disease were measured in population-based surveys of Twin Cities metropolitan area residents aged 25-74 years in 1980-1982 and 1985-1987. During this period, age-adjusted, average systolic blood pressure declined nonsignificantly in men and significantly in women, and average, age-adjusted diastolic blood pressure declined 1.1 mmHg in men and 0.9 mmHg in women. Between 1980-1982 and 1985-1987, serum total cholesterol declined significantly, 5.2 and 5.8 mg/dl in men and women, respectively. Serum high density lipoprotein cholesterol decreased 1.6 mg/dl in men and 0.9 mg/dl in women. The prevalence of cigarette smoking declined by 14% in men and 18% in women. In contrast, the average body mass index increased 0.6 kg/m2 in men and 1.2 kg/m2 in women. Systematic hospital and death certificate surveillance found that mortality rates for coronary heart disease in the Twin Cities metropolitan area declined 20.1% in men and 12.9% in women from 1981 to 1986. Despite difficulties in interpretation of ecologic studies, it appears likely that improvements in population risk factor levels played a role in the decline in disease rates and could influence future mortality trends in this population.

Type A behavior and its association with cardiovascular disease prevalence in Blacks and Whites: The Minnesota heart survey

Population-based surveys were conducted in 1985 and 1986 to measure the prevalence of coronary heart disease (CHD) history and risk factors in Black and White adults. Type A behavior was measured by the Jenkins Activity Survey (JAS). JAS scores were associated with age (negatively), education (positively), and sex (men greater than women) but were largely unrelated to CHD risk factors. Blacks had significantly lower age- and education-adjusted Type A and component scores than Whites, more so for men than women. Univariate analysis indicated that a history of angina and/or heart attack was positively associated with the Type A score in both Blacks and Whites. Following adjustment for known cardiovascular risk factors, Type A score remained positively and significantly associated with CHD prevalence. These findings are consistent with other cross-sectional studies and suggest that Type A behavior, as measured by the JAS, may increase the risk of CHD in both Blacks and Whites. Follow-up of these cohorts may help to clarify the complex relationship of Type A behavior to the risk of CHD.

The effect of the number of electrocardiograms analyzed on cardiovascular disease surveillance: The Minnesota heart survey (MHS)

One method used to control costs in community cardiovascular disease surveillance is to limit the number of electrocardiograms (ECGs) used to validate acute myocardial infarction (AMI). The Minnesota Heart Survey investigated the impact of decreasing the maximum number of ECGs analyzed on classification of ECG pattern and final AMI diagnosis (definite, probable, none). A 50% sample of all 1980 acute CHD hospital discharge records (ICD-9 code 410 or 411) from 30 of 31 Twin Cities hospitals were abstracted. Comparing results using all available ECGs in the record (maximum of 12) with those obtained using up to 4 ECGs showed little differences in the ECG classification or final AMI diagnosis.

Hypercholesterolemia prevalence, awareness, and treatment in blacks and whites: The Minnesota heart survey

Two cross-sectional population-based surveys were conducted in 1985 and 1986 to describe cardiovascular risk factors in blacks and whites in the Twin Cities. A total of 1,254 blacks and 2,934 whites ages 35–74 years participated. The surveys consisted of a home interview followed by survey center visit during which nonfasting serum total cholesterol level was measured and medication use during the past year was reviewed. Age-adjusted mean values for serum total cholesterol were significantly higher among white than black participants for both men (207 vs 193 mg/dl, P < 0.001) and women (206 vs 202 mg/dl, P < 0.05). Blacks had significantly higher serum HDL cholesterol levels than whites (men, 49 vs 41 mg/dl, P < 0.001; women, 56 vs 54 mg/dl, P < 0.01). The age-adjusted prevalence of hypercholesterolemia (serum total cholesterol ⩾240 mg/dl on the day of survey and/or current use of cholesterol lowering medication) was significantly higher among white than black men (18.3% vs 12.2%, P < 0.01). No significant race differences were noted for women (whites, 19.7% vs blacks, 16.6%). Among hypercholesterolemic men, 66% of whites and 80% of blacks were unaware of their condition; among women, 72% of whites and 79% of blacks were unaware. Among individuals told by a physician they had “high blood fats,” 2.9% of whites and no blacks were using medication for elevated blood cholesterol levels, while 70% of whites and 63% of blacks reported being advised to follow a low-fat-low-cholesterol diet. These data emphasize the need for education programs for physicians and patients regarding detection and control of hypercholesterolemia.

Trends in CHD Mortality, Morbidity and Risk Factor Levels from 1960 to 1986: The Minnesota Heart Survey

The Minnesota Heart Survey (MHS) assessed population trends in coronary heart disease (CHD) mortality, morbidity and risk factor levels in the Twin Cities metropolitan area to explain the decline in CHD deaths. Age-adjusted CHD mortality rates declined in Twin Cities residents aged 30 to 74 from 1968 to 1986 by 52% in men and 58% in women. Much of the decline in CHD mortality was attributable to decreased out-of-hospital deaths. Attack rates based on hospitalized definite myocardial infarction (MI) did not change from 1970 to 1985, and hospital case-fatality rates declined consistently from 1970 to 1985 (from 21% to 11% in men, from 27% to 17% in women). Dramatic increases were noted in the percentage of patients undergoing cardiac procedures (coronary artery bypass, angioplasty, and thrombolytic agents). Trends in CHD rates were associated with improved population levels of risk factors. Average serum total cholesterol levels decreased by 8 mg/dl and 11 mg/dl between 1973 and 1987 in men and women, respectively. Cigarette smoking prevalence declined from approximately 40% to 30%. Average systolic blood pressure apparently declined about 1 mm Hg and diastolic blood pressure by 2 mm Hg from 1973 to 1987. Body mass index (wt/ht2) increased by 1 unit in men and 1.7 units in women from 1973 to 1987. These data suggest that a combination of primary prevention and improved medical care of acute CHD contributed to the decline in CHD mortality.

The effects of diagnostic criteria on trends in coronary heart disease morbidity: The Minnesota heart survey

The Minnesota Heart Survey assessed attack rates of MI in Twin Cities residents ages 30–74 years in 1970 and 1980. The age-adjusted attack rate per 100,000 of definite MI was similar in 1970 (174.2) and 1980 (179.9) p > 0.05, using ECG, chest pain, and blood enzyme concentrations of aspartate transaminase and/or lactic dehydrogenase as criteria. The attack rate of definite MI also remained constant when autopsy findings were included in the algorithm, 197.0 in 1970 and 191.4 in 1980 ( p > 0.05). Adding creatine phosphokinase (CPK) and CPK-MB isoenzyme to the algorithm increased the rate of definite MIs from 209.0 in 1970 to 277.0 in 1980 ( p < 0.001). Interpretation of long-term trends in coronary heart disease morbidity is highly dependent upon variables used to validate cases. Care must be taken to maintain consistent criteria to avoid bias due to improvements in diagnostic techniques over time which increase sensitivity for detection of cardiac ischemia.

Trends in CHD Mortality, Morbidity and Risk Factor Levels from 1960 to 1986: The Minnesota Heart Survey

Trends in CHD Mortality, Morbidity and Risk Factor Levels from 1960 to 1986: The Minnesota Heart Survey

Educational attainment and nutrient consumption patterns: The Minnesota Heart Survey

To explore whether dietary patterns differ according to educational attainment, we examined the associations of nutrient intake with educational level in a sample of 825 male and 893 female Twin Cities (Minneapolis and St. Paul) residents aged 25 to 74 years who were surveyed between 1980 and 1982. Of particular interest was the pattern of macronutrient consumption, since the proportion of fat and type of fat in the diet is thought to be related to the development of chronic disease. Estimates of nutrient intake were based on 24-hour dietary recalls, while information on education was obtained through interview. Educational attainment was positively associated with nutrient consumption patterns that may decrease risk of chronic disease for both men and women. For women, the proportion of dietary fat was less (p=.03) and carbohydrate greater (p=.004) with increasing education. For men, the Keys dietary score decreased as education increased (p=.05) (high scores indicate greater blood cholesterol-raising effect of the diet). Inverse associations of average total serum cholesterol level with education in both men (p=.02) and women (p=.06) support these dietary associations. These results suggest that educational attainment is related positively to eating patterns that may carry a decreased risk of cardiovascular and other chronic disease. Whether or not this association is causal, it would appear to point to a need for more effective nutrition education strategies for those with less education. J Am Diet Assoc 88:1230, 1988.

Use of aspirin for prevention of cardiovascular disease—1981–1982 to 1985–1986: The Minnesota Heart Survey

The efficacy of aspirin to prevent ischemic cardiovascular disease has received considerable attention recently. To determine the prevalence of aspirin use for cardiovascular disease prevention, the Minnesota Heart Survey examined population-based samples of Twin Cities' adults in 1981 to 82 and 1985 to 86. Over the 4-year period, reported use of aspirin for cardiovascular disease prophylaxis increased from 0.6% to 2.4% in women ( p < 0.05) and from 1.7% to 3.3% in men ( p = 0.10). Prophylactic aspirin use was more common in older than in younger adults, in whites than in blacks, in those with a history of cardiovascular disease or hypercholesterolemia, and in health professionals and nonsmokers. Some individuals were apparently taking aspirin for “primary” prevention, although it has not yet been approved for that reason. Use of aspirin for cardiovascular disease prevention has been increasing and is likely to increase more rapidly over the next few years. This could favorably impact on mortality rates of cardiovascular disease, but untoward side effects of aspirin may be expected to increase as well.