OBJECTIVES: To understand the potential for drug-drug interactions (DDIs) among patients with diabetes and/or hypertension and multiple other comorbidities. METHODS: Patients were selected from the 2005 MarketScan databases who had hypertension and/or diabetes, a chronic disease score in the top 10% of the cohort, and 12 months of continuous enrollment. Concomitant medications were identified from the 12 month follow-up and tested against the DRUG-REAX system, which is used by pharmacists to check for potential DDIs and determine their clinical significance. The system includes a severity rating and documentation quality classification for the interaction potential. A DDI was counted when drugs in potentially interacting combinations with excellent or good documentation were dispensed within 30 days of each other during the time period. RESULTS: A total of 98,844 patients met the study criteria with 79,830 (80.7%) of them having at least one potential DDI. These patients filled an average of 28 prescriptions per year and almost 98% were over age 64. Among these patients, 306,649 unique, potential DDIs were identified with the severity rating distributed as follows: Contraindicated – 0.8%; Major – 29.9%; Moderate – 61.1%; Minor 8.3%. Potential DDIs included: potassium chloride in combination with anticholinergics (contraindicated), potassium chloride with lisinopril (major), furosemid with digoxin (moderate), and warfarin with levofloxacin (moderate). CONCLUSIONS: Potential DDIs were common in this elderly population with multiple comorbidities. While some drug combinations with potential DDIs may be clinically appropriate, they require ongoing monitoring to ensure patient safety. The large number of potential DDIs identified with these data warrants future research into the prevalence of appropriate monitoring when potentially interacting drug combinations are prescribed. This study demonstrates the utility of using prescription claims databases for identifying specific sub-populations of patients at high risk for potential DDIs and targeting appropriate areas for intervention.