This article reports an examination of factors that might affect the interpretation of computed relative risks of medical illness obtained when estimates of morbidity derived from a specific subpopulation sample are compared with estimates obtained from official statitical bureau surveys of the...