The aplication of census socioeconomic and familial data to the study of morbidity from mental disorders.

Abstract

Descriptive data is given on rates of admission to psychiatric facilities according to type of family, family size and relationship to family head, as well as certain demographic variables. The emphasis is on problems related to the census-matching methodology employed in the study. The analysis of the data was complicated by at least 3 problems: failure to find matching Census schedules, under-enumeration of the population in the Census and large sampling variation in the rates. The latter 2 factors are not necessarily peculiar to a matching study alone. They might apply equally well to studies in which the numerators are obtained independently of the population denominator. Before any judgment can be made as to the best design to use in studying psychiatric morbidity rates, further investigation must be carried out with respect to these methodological problems.