OBJECTIVES: To compare the risk of hospital and/or Emergency Department (ED) visits [events], based on various treatment regimens in a population of Medicaid patients with Chronic Obstructive Pulmonary Disease (COPD). METHODS: Retrospective observational cohort design. Texas Medicaid patients aged 40 to 65 years enrolled from 1998 to 2001 with a primary diagnosis of COPD were identified. Five index therapy cohorts were assessed 1) ipratropium / Combivent (IPR); 2) salmeterol (SAL); 3) inhaled corticosteroid (ICS); 4) ICS + IPR; and 5) ICS + SAL. Subjects were followed until they had a COPD-related event, or for 12-months, whichever came first. Cox proportional hazard analysis compared therapy cohorts, assessing time to first COP-related event, adjusting for gender, age, ethnicity, presence of co-morbid respiratory diseases, number of other co-morbidities, pre-index use of short-acting beta agonists (SABAs), oral steroids and theophylline, and pre-index events. RESULTS: A total of 4447 patients were identified, 2435 (55%) on IPR, 1088 (24%) on ICS alone, 410 (9%) on ICS + IPR, 299 (7%) on SAL alone and 215 (5%) on ICS + SAL. Compared to IPR alone, the use of ICS alone (HR 0.82, 95% CI: 0.69, 0.96), SAL alone (HR 0.62, 95% CI: 0.46, 0.84), and ICS+SAL (HR 0.64, 95% CI: 0.45, 0.91) were associated with a lower event risk when controlling for the other factors. Factors that were significant predictors of an increase risk of an event were: being white, having a co-morbid respiratory disease, using pre-index SABAs, and having pre-index events. CONCLUSION: The results of this analysis suggest that the use of inhaled corticosteroids and salmeterol were associated with a significant decrease in the risk of COPD-related hospital/ED events compared to the use of IPR alone in a population of Medicaid patients. This was an observational retrospective analysis and these data should be confirmed by prospective studies.