OBJECTIVES: To analyze the risk of re-hospitalization for COPD following different treatment regimens. METHODS: A retrospective cohort study with data from the Dutch PHARMO system, including medication and hospital admission records of 1.6 million inhabitants of 34 Dutch cities was performed. Patients > 40 yrs with a first admission for COPD during 1992-2002 were identified. Treatment was assessed in the period of 90 days following first discharge. Cox proportional hazard analysis was performed to analyze the time to re-hospitalization for COPD for different treatment regimes. Hazard rate (HR) and 95% confidence intervals (CI) were adjusted for potential risk factors including gender, age, year of hospitalization and disease severity. The maximum follow-up time was two years. RESULTS: From the total cohort of 3855 patients, 1118 (29%) were re-hospitalized within two years after the first hospitalization. The mean period to re-hospitalization was 230 days. Compared to ICS as a reference group, an increased risk for re-hospitalization was found in the following treatment groups: theophylline (HR 1.83, CI 1.12-2.99), and ICS combined with anticholinergics and beta-agonists (HR 1.50, CI 1.02-2.20). CONCLUSIONS: This study found re-hospitalizations occurred in approximately one-third (1/3) of patients. Differences in COPD re-hospitalization risks between treatment groups were observed. Anti-inflammatory therapy alone demonstrated a significant protective effect compared to theophylline therapy alone or a combination therapy of ICS, anticholinergics and beta-agonists.