OBJECTIVES: Retrospective Drug Utilization Review (RDUR) can be effective in reducing drug-related problems and adverse health outcomes. This study examined whether RDUR intervention letters to prescribers impacted the recurrence of drug- prescribing programs (exceptions) and evaluated the subsequent effect on utilization of health care services. METHODS: Data used were from combined pharmacy and medical claims from January 2002 through November 2003. To assess the impact of intervention letters on the recurrence of exceptions, multiple regression models were developed using prescriber-level mean time (in days) between exceptions. In addition, the effect on medical costs and utilization was measured for a subset of data using a pre-post design with a control group. Pre-post periods were defined as 120 days before and after the date of intervention letters dated between May 1 and August 31, 2003. Controls were selected by matching to intervention cases using the propensity score methods. Sensitivity analysis was performed using varying time windows and bootstrap samples. Outcomes related to PMPM inpatient admissions, emergency room visits, and physician office visits were analyzed. RESULTS: Of 51,214 prescribers who had two or more exceptions during the 23-month time period, 6233 (12%) were randomly selected to receive intervention letters (ranging from one to 19). Model coefficients indicated that the time to exception was longer by 6.5 days (p <0.001) as prescribers received additional intervention letters, after adjusting for the number of exceptions, severity level, and average patient age. There were no significant differences in medication costs from pre to post time periods, or between groups (study vs. control). However, the study group had fewer PMPM inpatient admissions and emergency room visits. CONCLUSIONS: Retrospective drug utilization review processes can have a positive effect in delaying next exceptions for prescribers and reducing utilization of health care services.