The main objectives of this paper were to develop and subsequently test a Bayesian discrimination model for the purpose of identifying both the personal and the health care system characteristics predictive of hospitalization for the treatment of diabetes or commonly observed cormorbidities associated with the disease. METHOD: First, a Bayesian classification framework was proposed to discriminate patients into two groups - the high risk and the low risk. The model was then tested by using a logit regression technique in order to estimate the probability of one or more hospitalization events among diabetes patients. The study used claims data extracted from the Hawaii Medical Service Association (HMSA) Private Business Claims (PBS) files for the calendar year 1995. Patients under 65 years were identified by paid claims with ICD-9-CM diagnosis codes of 250.xx which gave a sample size of 6,841. RESULTS: Age, gender, various pharmacotherapy variables, presence of hypertension, hyperlipidemia, coronary heart failure, multiple cardiovascular diseases, any combination of commonly observed comorbidities, dialysis services, and annual eye examination were highly predictive of one or more hospitalization events. The model showed a predictive power of almost 90%. CONCLUSION: Multivariate discriminant analysis using a logit regression model successfully (1) identified important explanatory variables predictive of hospitalization, (2) assigned patients into one of two mutually exclusive classes, and (3) offered a benchmark for a comprehensive disease management strategy for more involved diabetic patients.