OBJECTIVES: Roflumilast is indicated for patients with moderate to severe Chronic Obstructive Pulmonary Disease (COPD). Information on the prevalence and risk of diarrhea, the most reported adverse event in clinical trials, among patients initiating roflumilast in real-world settings is limited.

METHODS: We selected patients aged ≥40 years with COPD who were new users of roflumilast and the active comparator group, triple therapy (inhaled corticosteroids (ICS), long acting beta agonists (LABA) and long acting muscarinic antagonists (LAMA)) from March 1, 2011 to December 31, 2017 in the Truven Health MarketScan Research Database. The study outcome was a primary diagnosis claim for non-infectious diarrhea within 14 days after treatment initiation. We estimated adjusted hazard ratios (HR) and 95% confidence intervals (CI) for diarrhea risk using Cox proportional hazards models. Follow-up continued to the earliest of the occurrence of the outcome of interest (diarrhea), treatment discontinuation, end of 14-day follow-up, disenrollment from health plan or end of study period. We used high-dimensional propensity score (hdPS) matching and calculated the E-value to quantify unmeasured confounding.

RESULTS: The final cohort included 4,725 new users of roflumilast and 17,540 matched sample of triple therapy users with a mean (SD) age of 67.9 years (10.9) and 50.7% female. Roflumilast users had higher risk of diarrhea over the 14-day period after drug initiation relative to the comparator group [HR=2.22, 95%CI (1.01, 4.89)]. The corresponding E-values for the HR and lower bound of the CI were 3.87 and 1.1 respectively. A similarly higher risk of the outcome occurred in the roflumilast group at 30 days [HR=2.06, 95%CI (1.22, 3.47)] and at 90 days [HR=1.40, 95% CI (1.01, 1.95)].

CONCLUSIONS: We observed a consistently higher risk of diarrhea in patients initiating roflumilast in the real-world settings and is the highest within the first 30 days post-initiation.