Objective: To develop a model to predict the probability of diabetes patients having inpatient hospitalization in the following year. Methods: A retrospective cohort study was conducted based on a population of 322 type II diabetes patients age>=19 and enrolled in a Medicaid managed care plan from the year 2003 to 2005. Models were developed by using medical/pharmacy utilization data to predict the probability of having the following events in the next year: (1) whether the patient had any inpatient hospitalization, (2) whether the patient had any micro/macro vascular inpatient hospitalization, (3) whether the patient had any metabolic related inpatient hospitalization and (4) whether the patient had any infectious inpatient hospitalization. Main predictors of interests are diabetes compliance and the use of statin. Covariates include diabetes treatment pattern, age, sex, co-morbidities among other variables. Logistic model is used to conduct the analysis. Results: The study population was 74% female with an average age of 49.2 (S.D=8.4). Non-adherence of diabetes drug (odds ratio=1.57, 95% CI: 1.09-2.24), use of statin (odds ratio=0.58, 95% CI: 0.39-0.85), and previous inpatient history (odds ratio=3.50, 95% CI: 2.42-5.07) were significant in predicting any inpatient hospitalization. Non-adherence of diabetes drugs was not a significant predictor for micro/macro vascular events (odds ratio=0.74, 95% CI: 0.33-1.67), but it was associated with significantly increased probabilities of having metabolic events (odds ratio=1.58, 95% CI: 1.03-2.43) and infectious events (odds ratio=2.70, 95% CI: 1.10-6.66). The use of statin was significant only for predicting metabolic events (odds ratio=0.49, 95% CI: 0.32-0.77). It was not significant for predicting vascular events (odds ratio=0.98, 95% CI: 0.45-2.15) and infectious events (odds ratio=1.40, 95% CI: 0.49-3.72). Conclusion: The modeling results show that improving compliance of diabetes drug and encouraging the use of statin could be associated with reducing inpatient hospitalizations in a short period of time.