Cholesterol and coronary heart disease. The importance of patient-specific attributable risk.

The value of treating hypercholesterolemia remains controversial despite extensive investigation, in part because of the research orientation of much of the relevant literature. Most published studies report relative measures of risks such as the risk ratio. However, clinicians are concerned with the magnitude of risk, or attributable risk, the difference in risk in those with highest and lowest serum cholesterol levels. Since maximum risk modification is rarely achieved, the attributable risk often overstates the potential benefit of risk factor reduction. A variant of the attributable risk, the practical attributable risk, gives a more realistic estimate of potential benefit. Since clinicians treat individuals and not populations, these measures of risk are most useful if reported for subgroups of patients (eg, men and women, as well as those of varying age). To illustrate these concepts, we review the literature on hypercholesterolemia, report the risk ratio, attributable risk, and practical attributable risk for important patient populations, and discuss the implications for clinical practice.