A single cholesterol measurement underestimates the risk of coronary heart disease. An empirical example from the Lipid Research Clinics Mortality Follow-up Study.

In prospective epidemiologic studies of coronary heart disease, a single measurement of cholesterol is made to assess its relationship to the risk of coronary disease. Statistical theory states that if this measurement is subject to within-individual variability, the strength of the relationship will be underestimated. This is empirically shown for the example of plasma cholesterol. For the Lipid Research Clinics Follow-up Study population (comprising 2170 white men over 30 years of age), the age-adjusted coronary heart disease mortality regression coefficient increases from .453 to .496 if the average of two cholesterol measurements is used instead of a single measurement. Since the correlation between the two repeated cholesterol measurements is .815, an increase in the regression coefficient up to .556 would be expected if the true cholesterol values were available. Thus, epidemiologic studies have substantially underestimated the strength of the relationship between cholesterol levels and the risk of coronary disease by calculating the relationship on the basis of a single cholesterol determination.