Comparison of claims vs patient-reported adherence measures and associated outcomes among patients with nonvalvular atrial fibrillation using oral anticoagulant therapy

Judith J Stephenson,1 Mayura U Shinde,1 Winghan Jacqueline Kwong,2 An-Chen Fu,1 Hiangkiat Tan,1 William S Weintraub3 1HealthCore, Inc., Wilmington, DE, 2Daiichi Sankyo, Inc., Basking Ridge, NJ, 3Christiana Care Health System, Newark, DE, USA Objective: To compare oral anticoagulant (OAC) adherence among patients with nonvalvular atrial fibrillation (NVAF) using patient-reported and claims-based measures, and to evaluate the effect of OAC adherence on health care costs and patient satisfaction with OAC therapy.Methods: This was a hybrid US observational study consisting of a longitudinal cohort survey followed by linkage and analysis of respondents’ administrative claims data. Patients with NVAF receiving warfarin, dabigatran, rivaroxaban, or apixaban completed an initial survey and follow-up surveys at 4, 8, and 12 months. Patient-reported adherence was measured at each survey by Morisky Medication Adherence Scale (MMAS-8) and pharmacy claims-determined adherence by the proportion of days covered (PDC) for the 12-month period following the initial survey date; adherence was defined as MMAS-8 score =8 and PDC ≥80%. Patient satisfaction with OAC therapy was assessed by the Duke Anticoagulation Satisfaction Scale (DASS).Results: Overall, 675 patients completed at least the initial survey (warfarin, n=271; dabigatran, n=266; rivaroxaban, n=128; apixaban, n=10). Fewer than half (47.9%) were PDC adherent, 37.2% were MMAS-8 adherent, and 19.4% were adherent by both measures. Total medical costs of PDC-adherent patients were significantly lower vs PDC-nonadherent patients (US$640 vs $993 per-patient per-month, respectively, p<0.05). MMAS-8-adherent patients reported higher treatment satisfaction; total DASS score was significantly lower among MMAS-8-adherent than MMAS-8-nonadherent patients (37.3 vs 42.9, respectively, p<0.001).Conclusion: Using claims-based or patient-reported methods to measure OAC adherence may lead to different results whe