Epidemiological studies of diabetes mellitus in Denmark. I. A case finding method based on the National Service Conscript Registry.

To provide unbiased material for epidemiological studies of diabetes mellitus a case finding method based on the Danish National Service Conscript Registry and death certificates is presented and discussed. Eight Danish total male birth cohorts, totalling 320 162 persons, have been observed as to the occurrence of diabetes mellitus during the first twenty years of life, and it is concluded that the conscript registry represents an ideal basis for case identification, if the individual cases are further elucidated. The material is estimated to be about 95% complete. The main source of error originates from omissions when the registries are scrutinized. The cumulative incidence rates for the birth cohorts range from 1.3 to 3.2 per thousand; in the total material 2.4 per thousand.