Methods & patient characteristics
- Administrative healthcare claims and laboratory results data from the HealthCore Integrated Research DatabaseSM were evaluated for COPD patients initiating first treatment with T/O versus LABA/ICS during January 2013–March 2019
- Date of first prescription was defined as the index date
- Patients were followed until discontinuation or switch of their index treatment, the end of health plan enrollment, or 1 year after the index date
- A Cox proportional hazard regression model was used to perform an as-treated analysis to assess risk of escalation to triple therapy, COPD exacerbation, pneumonia or an adverse outcome (i.e. one of the above)
- Potential imbalance of confounding factors between cohorts was handled using fine stratification and reweighting of the exposure propensity score (high-dimensional)
- Data were analyzed separately for subgroups based on circulating eosinophil levels (for those with available results) and exacerbation history
Abbreviations: ICS, inhaled corticosteroid; IR, incidence rate; LABA, long-acting β2-agonist; T/O, tiotropium/olodaterol.