Cyclosporin and other immunosuppressive agents have been proposed as a preventive treatment against the development of insulin-dependent diabetes mellitus (IDDM) in relatives at increased risk for the disease, based on the understanding that its etiology is an ongoing process of autoimmune beta-cell destruction. We used an epidemiological approach to evaluate several recent trials of cyclosporin in newly diagnosed IDDM patients to determine the degree of benefit that is to be expected. We assessed these and other studies to estimate the potential adverse effects of such treatment, were it to be used in the future, either in newly diagnosed subjects or healthy high-risk relatives. Standard sample-size calculations were used to quantify the number of study subjects necessary to allow adequate statistical power to test the positive and negative effects of a future treatment (alpha = 0.05, beta = 0.20). The estimates were based on the data available from published studies of cyclosporin treatment. The importance of conducting an adequate trial of such a therapy, both from an ethical and a practical viewpoint, is discussed. Five small immunosuppression trials were evaluated. Remission rates in treated subjects exceeded those in control subjects by 15-59%. Variability in defining remission may account for the differences in rates across the studies. Estimates of the major beneficial and adverse effects of cyclosporin were derived from these trials and studies of patients who have undergone long-term immunosuppression. Indicators of kidney damage associated with cyclosporin treatment were reported in 5-47% of treated subjects.(ABSTRACT TRUNCATED AT 250 WORDS)