The epidemiologic approach to studies of association between HLA and disease. II. Estimation of absolute risks, etiologic and preventive fraction.

It is the aim of the present report to review principles and measures which are fundamental in modern epidemiology and happen to be closely related to expressions used in studies of association between HLA and disease. These measures, the etiologic and the preventive fractions, may easily be estimated from the traditional 2 x 2 table. Furthermore, the practical procedures leading to estimates of the absolute frequency measures for individuals with and without a given HLA factor are presented; these estimates may also be obtained from the 2 x 2 table, supplemented with data on the disease frequency in the general population. The estimation procedures are simple in all the subtypes of the association study design, but especially in prevalence-based association studies some important limitations as to the interpretation and generalization should be kept in mind. A subsequent report will use the measures and principles, outlined here, in order to facilitate an extension of the perspectives of the results and establish a connection between this field of HLA research and clinical and genetic epidemiology.