Current issues of ACP Journal Club are published in Annals of Internal Medicine


High cholesterol levels were a risk factor for mortality from coronary heart disease in men

ACP J Club. 1991 Mar-April;114:62. doi:10.7326/ACPJC-1991-114-2-062

Source Citation

Rubin SM, Sidney S, Black DM, et al. High blood cholesterol in elderly men and the excess risk for coronary heart disease. Ann Intern Med. 1990 Dec 15;113:916-20.



To determine whether high blood cholesterol levels are a risk factor for mortality from coronary heart disease in elderly white men.


A cohort of elderly men followed for 10 years.


A health maintenance organization in California, USA.


Participants were drawn from 2941 white men between the ages of 60 and 79 years, who had at least 3 multiphasic health check-ups between 1964 and 1972 and had no self-reported history of coronary heart disease at the third examination. Cholesterol measurements made at the third check-up were available for 2746 men, who constituted the study group. All patients were followed from the third check-up through 1980, for a mean of 10.1 years.

Assessment of risk factors

Because serum cholesterol levels were measured by 3 different techniques (SMA and Autoanalyzer [Technicon, U.S.A.] and Autochemist [AGA Corporation, Sweden]), they were classified by method-specific quartile levels. Information on age, systolic blood pressure, and smoking status was collected.

Main outcome measure

Mortality from coronary heart disease, defined by death certificate International Classification of Diseases (8th edition) codes 410 to 414, was ascertained through the California Automated Mortality Linkage System.

Main results

260 deaths from coronary heart disease occurred during follow-up. Cardiac heart disease mortality rose with increasing age and cholesterol level. Comparing mortality from coronary heart disease in the highest quartile of cholesterol with mortality from coronary heart disease in the other 3 quartiles, the relative risk was similar in each of the 5-year age groups (1.4 at 60 to 64 y [95% CI 0.9 to 2.2] to 1.7 at 75 to 79 y [CI 0.8 to 3.6]). However, excess risk for mortality from coronary heart disease associated with the highest quartile of cholesterol increased from 2.2 deaths per 1000 person-years at age 60 (CI -1.2 to 5.5) to 11.3 deaths per 1000 person-years after age 75 (CI -6.6 to 29.2). After adjustment for age, blood pressure, and smoking, men in the highest quartile of serum cholesterol had a 1.38-fold (CI 1.04 to 1.83) higher coronary mortality than men in the other quartiles.


High blood cholesterol was an important risk factor for mortality from coronary heart disease in elderly men.

Sources of funding: Henry J. Kaiser Family Foundation and the Kaiser Foundation Research Institute.

Address for article reprint: Ms. S.M. Rubin, University of California at San Francisco, Prevention Sciences Group, Box 0886, San Francisco, CA 94143, USA.


This study is limited by being restricted to elderly white men, preventing extrapolation to women or to men of other races. Causes of death were determined from death certificates, which are not noted for their accuracy. The precision of the risk estimates is limited by the relatively few deaths caused by coronary heart disease in each age stratum. Nevertheless, the study supports the view that elevated serum cholesterol levels remain an important risk factor for mortality from coronary heart disease in elderly white men.

The investigators stress the differences between relative risk and excess risk. Calculating risk allows comparison of cohorts with different exposures. Relative risk (the incidence in an exposed cohort divided by the incidence in an unexposed cohort) points out how many times more likely exposed persons are to develop a disease compared with those unexposed. Excess risk (the incidence of disease in an exposed cohort minus the incidence in an unexposed cohort) indicates the excess incidence of a disease caused by exposure. Excess risk helps define the effect of an association between an exposure and a disease and provides a more meaningful expression of risk in most clinical situations. The study does not, however, provide information on a clinical approach. Currently, it is not known whether lowering serum cholesterol in the elderly will decrease the risk for mortality from coronary heart disease. As the investigators state, a randomized trial of reducing cholesterol in the elderly is needed to provide clinical guidance.

George J. Caranasos, MD
University of FloridaGainesville, Florida, USA