Symptom Burden and GOLD Classification in Medicare Advantage Patients with COPD Initiating Umeclidinium/Vilanterol or Fluticasone Propionate/Salmeterol Therapy

Chad Moretz,1 Beth Hahn,1 John White,2 Alyssa Goolsby Hunter,2 Breanna Essoi,2 Caitlin Elliott,2 Riju Ray3 1US Value Evidence & Outcomes, GlaxoSmithKline, Research Triangle Park, NC, USA; 2Optum, Eden Prairie, MN, USA; 3US Medical Affairs, GlaxoSmithKline, Research Triangle Park, NCCorrespondence: Beth HahnGSK, 5 Moore Drive, Research Triangle Park, Durham, NC, USATel +1 919 274 0660Email beth.a.hahn@gsk.comBackground: Long-acting muscarinic antagonist/long-acting β2-agonist (LAMA/LABA) provide greater improvements in lung function and symptoms than inhaled corticosteroid (ICS)/LABA in patients with chronic obstructive pulmonary disease (COPD). This study evaluated symptom burden and Global Initiative for Obstructive Lung Disease (GOLD) categorization among patients who recently initiated umeclidinium/vilanterol (UMEC/VI; LAMA/LABA) or fluticasone propionate/salmeterol (FP/SAL; ICS/LABA) single-inhaler dual therapy.Methods: COPD-diagnosed Medicare Advantage enrollees aged ≥ 65 years were identified from the Optum Research Database (ORD). Eligible patients had ≥ 1 pharmacy claim for UMEC/VI or FP/SAL in the 6-month period before sample identification, with no evidence of triple therapy (ICS/LAMA/LABA), asthma, or lung cancer. Symptom burden was assessed via cross-sectional surveys using the COPD Assessment Test (CAT) and modified Medical Research Council (mMRC) dyspnea scale. Patients were classified into GOLD categories using patient-reported symptoms and claims-based exacerbation history. Treatment groups were balanced on potential confounders using inverse probability of treatment weighting (IPTW). CAT and mMRC scores were analyzed with generalized linear regression models using IPTW propensity scores.Results: The final analytic sample included 789 respondents (UMEC/VI: N=392; FP/SAL: N=397). Approximately 66% patients were classified as GOLD B when assessing symptoms with CAT and mMRC together, or CAT alone; more patients were classified as GOLD A (∼ 40%) than GOLD B (∼ 36%) using mMRC alone. Proportions of patients in each GOLD group were similar between treatment cohorts. Post-IPTW multivariable modeling showed similar symptom burden between treatment groups.Conclusion: After controlling for baseline characteristics, symptom burden was similar between patients receiving UMEC/VI or FP/SAL. GOLD classification using mMRC produced more conservative results compared with CAT, potentially underestimating patient symptoms. Many patients receiving FP/SAL were classified as GOLD A or B, despite GOLD recommending non-ICS-containing therapy in these patients. These findings support the need for routine assessment of symptoms in patients with COPD.Keywords: COPD, fluticasone propionate/salmeterol, umeclidinium/vilanterol, CAT, mMRC, GOLD group