To characterize the treatment received by patients with an Upper Respiratory Infection (URI) in the South Carolina Medicaid population during 1996. Literature suggests that antibiotics are commonly prescribed to patients presenting with symptoms of an URI even though less than 20% are bacterial in origin. Overuse of antibiotics may lead to antibiotic resistance and unnecessary costs. METHODS: Patients with a diagnosis for a cold (ICD-9-CM 460) or URI (ICD-9-CM 465) between 01/01/96 and 12/31/96 were identified. The study design allowed for each subject to have multiple diagnoses (episodes). For each episode, subjects with prescriptions for antibiotics or symptomatic cold medications within 7 days after the diagnosis were identified. Any episode within 14 days of a claim for a bacterial infection, sinusitis or otitis was removed from the study, to rule out the possibility that the antibiotic was related to another infection. RESULTS: A total of 22,791 patients were eligible for inclusion in the study, incurring 27,847 episodes (Range 1-7 episodes). 9468 (34%) of the episodes were treated with an antibiotic, accounting for $166,702.33. This rate is lower than that seen in previous studies, however our inclusion were more stringent than was used previously. If 20% of URI's are bacterial, approximately 3900 of these antibiotics were unnecessary. A 25% reduction in the excess prescriptions would provide a cost savings of $17,169.75 for South Carolina Medicaid. CONCLUSION: In spite of comprehensive documentation arguing against using antibiotics for treatment of URI's, it appears they still remain a popular choice for physicians, with 34% of subjects in this study receiving antibiotics. This represents an area for cost savings for the South Carolina Medicaid Program.