Sturmans F, Mulder P G, Valkenburg H A
Am J Epidemiol. 1977 Mar;105(3):281-9. doi: 10.1093/oxfordjournals.aje.a112384.
In epidemiology the concepts of relative and attributable risk are used to describe the statistical association between the incidence of a disease and the presence of possible risk factors. If the association is due to a cause-effect relationship, the attributable risk can be considered as an estimate for the reduction in incidence as a consequences of intervention. In order to get unbiased estimates these risks must be standardized for the influence of confounding variables. From data of the Framingham Study these risks, standardized for the confounding influence of age, are estimated for three risk factors related to the incidence of coronary heart disease (CHD)--hypertension, hypercholesterolemia and cigarette-smoking--both marginally and jointly. Under very optimistic assumptions a theoretical reduction in 12-year CHD incidence of a maximum of about 20% is estimated from the male Framingham sample by assumedly lowering systolic blood pressure and serum cholesterol. If only cigarette-smoking could be totally eliminated, the reduction is estimated at about 37%.
