The authors compared the epidemiology of childhood insulin-dependent diabetes mellitus in Estonia during 1980-1989 between native Estonians and an immigrant group that consisted mainly of Russians. The average annual incidence of diabetes mellitus was significantly higher in Estonians (11.8 per 100,000 children aged less than 15 years; 95% confidence interval (CI) 10.4-13.3) than in non-Estonians (7.6 per 100,000 children aged less than 15; 95% CI 6.2-9.4). This difference appeared in both sexes. The highest incidence in both Estonians and non-Estonians was recorded in 1982, when the incidence in the immigrant population was twice as high as the baseline level. These data indicate that immigrant populations need not acquire the same risk of insulin-dependent diabetes as the native population.