Tiotropium (Spiriva is an inhaled, once-daily anticholinergic medication for chronic obstructive pulmonary disease (COPD). We conducted a population-based cohort study to examine the risk of cardiovascular and respiratory hospitalizations and mortality with tiotropium. Using the Danish healthcare registries, we identified persons >/=40 years old in three counties who were hospitalized for COPD from 1/1/1977 to 12/31/2003. Respiratory and cardiovascular medications were assessed from dispensing records. Cox regression was used to compute incidence rate ratios (RR) and 95% confidence intervals (CI) for hospitalization and death between 1/1/2002 and 12/31/2003, associated with periods of tiotropium use compared to non-use, controlling for age, gender, time since COPD, concomitant respiratory and cardiovascular medications, prior hospitalizations and Charlson comorbidity index. Among persons with COPD (10,603), 75% were >/=60 years old. Follow-up was >/=18 months for 64%. Among those exposed to tiotropium compared to periods of non-use, the RR for total and cause-specific hospitalization endpoints were not elevated except for COPD hospitalization (RR = 1.52, 95% CI: 1.29, 1.79). Mortality endpoints included total mortality (RR = 0.77, 95% CI: 0.65, 0.91), respiratory mortality (RR = 0.79, 95% CI: 0.60, 1.04), sudden death (RR = 0.71, 95% CI: 0.21, 2.34), cardiac arrest (RR = 0.74, 95% CI: 0.42, 1.32), heart failure (RR = 0.84, 95% CI: 0.41, 1.75), and myocardial infarction (RR = 1.25, 95% CI: 0.49, 3.17). Compared to periods of non-use, tiotropium was associated with reduced respiratory and overall mortality and was not associated with increased cardiac mortality. An increase in COPD hospitalization is inconsistent with clinical trial data and suggests preferential prescribing due to disease severity.