Purpose. Chronic therapy with long-acting bronchodilators (LB) is recommended to treat moderate-to-severe COPD. Although the benefits of adding inhaled corticosteroid (ICS) to LB are still unclear, patients who experience repeated exacerbations are suggested to add ICS to their LB treatment. The objective of this study is to analyze whether adding ICS to LB therapy reduces mortality. Methods. We identified a cohort of patients discharged from hospital with COPD diagnosis between 2006 and 2009. The first prescription for LB or ICS following discharge was defined as the index prescription. Only new users were included (no use of any study drug in the 6 months before treatment). A 4-day time window was used to classify patients into “LB alone” or “LB plus ICS” initiators. We used propensity score to balance the study groups. Sensitivity analyses were performed in patients with recent out-of-hospital exacerbations. Results. Among the 18615 adults enrolled, 12207 initiated “LB plus ICS” therapy and 6408 “LB alone.” Crude mortality rates were 110 and 143 cases per 1000 person-years in the “LB plus ICS” and “LB alone” groups, respectively. The adjusted hazard ratio (HR) was 0.83 (95% CI: 0.72–0.97; p-value: 0.024). When analyzing patients with recent out-of-hospital exacerbations, the benefit of the combination therapy was more pronounced, HR = 0.63 (95% CI: 0.44–0.90; p-value: 0.012). Discussion. Our findings showed a beneficial effect on mortality of adding inhaled corticosteroids to long-acting bronchodilators. The advantage was much more pronounced in patients with frequent exacerbations.