Attributable risk percent in case-control studies.

The difference between the disease rates of two groups, one exposed and one not exposed to a suspected aetiological factor, is a useful measure of the risk associated with the factor. This measure is of most interest when the association is considered causal, and has been termed the attributable risk (MacMahon and Pugh, 1970). Expressed as a percentage of the total disease rate in a group (the attributable risk percent), the measure describes the percentage of the group's total risk which is in excess of the risk among persons not exposed to the suspect factor. Both measures (attributable risk and attributable risk percent) may be computed either for persons exposed to the suspect factor or for the population as a whole. Both measures are readily obtained when estimates of absolute disease rates are available for the groups to be compared. However, in many case-control studies subjects are not referable to an enumerated population, and, while relative risks can be obtained, absolute disease rates can not. If an estimate of the disease rate in the population is available from some other source, disease rates in exposed and nonexposed persons, and hence the attributable risks, may still be estimated (MacMahon and Pugh, 1970). It appears not to have been recognized, however, that even in the absence of knowledge of the disease rate in the population, estimates of attributable risk per cent can be obtained-both for exposed persons and for the population as a whole. When estimates of absolute disease frequency are available the attributable risk percent among the exposed is determined as: