TOPIC: Obstructive Lung Diseases TYPE: Original Investigations PURPOSE: This retrospective cohort study compared time to a) first hospitalization for community-acquired pneumonia and b) first COPD exacerbation in patients initiating maintenance therapy with the long-acting dual bronchodilator therapy tiotropium/olodaterol (T/O) vs triple therapy (TT) with long-acting muscarinic antagonist (LAMA)/long-acting beta agonist (LABA)/inhaled corticosteroids (ICS). METHODS: Administrative data from the US HealthCore Integrated Research Database (Jan 2013–Mar 2019) were used to compare patients for up to 1 year. Eligible patients were aged ≥40 years, diagnosed with COPD (not asthma), and had ≥1 inhaler prescription of fixed-dose T/O, or fixed or free TT (first prescription=index date). Propensity scores were calculated. Within the area of overlap, fine stratification and reweighting were used to balance measured covariates between the cohorts. Residual imbalances (standardized difference >0.10 [10%]) in the reweighted pseudo-population were adjusted for in the outcome models. A sensitivity analysis excluded patients with an exacerbation ≤30 days pre-index to minimize confounding e.g. due to higher TT prescription in frequent exacerbators. Further sensitivity analyses excluded patients with any prior use of maintenance therapy, and substituted the propensity score fine stratification/reweighting method with a 1:1 greedy match (propensity score caliper: 0.05). In all analyses, a Cox proportional hazard regression model was used to assess outcomes (as-treated analysis). RESULTS: The study population had 2,864 T/O and 24,326 TT users (after reweighting: 2,785 T/O and 15,465 TT users; mean age, 65.6 years; female, 47.1% [T/O] vs 49.7% [TT]). There were no differences in baseline characteristics, except for calendar year and season of cohort entry. No difference in prior exacerbations (any) was observed (std diff: <0.1); however, a higher proportion of TT users had severe exacerbations ≤1 year pre-index (32.7% vs 17.3%) and any exacerbations ≤30 days pre-index (25.8% vs 10.8%) vs T/O users. For T/O vs TT, adjusted hazard ratio (aHR) for exacerbation was 0.85 (95% CI: 0.76–0.96) (adjusted for calendar year, season of cohort entry). Excluding exacerbators ≤30 days pre-index balanced the cohorts (std diff: <0.1) in terms of prior exacerbations (any, moderate or severe). In this population, the aHR for exacerbation was 1.04 (95% CI: 0.91–1.19) (adjusted for calendar year, LABA/ICS use, season of cohort entry). Results were similar when applying 1:1 matching (aHR 1.02; 95% CI 0.84–1.24) and when restricting to treatment-naïve patients (no LABA, LAMA, ICS use prior to T/O or TT) (aHR 1.08; 95% CI 0.81–1.44). T/O users had a numerically lower risk of pneumonia than TT users (aHR 0.83; 95% CI 0.64–1.07). CONCLUSIONS: This study showed no benefit of TT vs T/O in reducing exacerbation risk in patients with no recent exacerbations, including treatment-naïve patients. In line with previous data, T/O was associated with a numerically lower risk of pneumonia vs TT. CLINICAL IMPLICATIONS: In an infrequently exacerbating population, prescribing T/O in place of TT appears not to increase exacerbation risk and could reduce pneumonia risk. DISCLOSURES: Employee relationship with Boehringer Ingelheim Please note: $20001 - $100000 by Alberto de la Hoz, source=Web Response, value=Salary Consultant relationship with HealthCore, Inc Please note: 2010-2018 Added 05/04/2021 by Daina Esposito, source=Web Response, value=Consulting fee Employee relationship with Moderna Please note: Since Feb 2021 Added 05/04/2021 by Daina Esposito, source=Web Response, value=Salary Owner/Founder relationship with Ciconia Inc Please note: 2018-2021 Added 05/04/2021 by Daina Esposito, source=Web Response, value=Ownership interest Employee relationship with Anthem Inc (HealthCore Inc) Please note: April 2019-present Added 05/04/2021 by Xintong He, source=Web Response, value=Salary Employee relationship with Boehringer-Ingelheim Please note: 2016- Added 05/03/2021 by Jukka Montonen, source=Web Response, value=Salary Advisory Committee Member relationship with GlaxoSmithKline Please note: Last 24 months Added 04/29/2021 by Jennifer Quint, source=Web Response, value=Honoraria Advisory Committee Member relationship with Boehringer Ingelheim Please note: Last 24 months Added 04/29/2021 by Jennifer Quint, source=Web Response, value=Honoraria Advisory Committee Member relationship with AstraZeneca Please note: Last 24 months Added 04/29/2021 by Jennifer Quint, source=Web Response, value=Honoraria Advisory Committee Member relationship with Bayer Please note: Last 24 months Added 04/29/2021 by Jennifer Quint, source=Web Response, value=Honoraria