Abstract 12761: Association of Medical Non-Adherence to Health Status Outcomes in the ISCHEMIA Trial

Abstract

Introduction: In the ISCHEMIA trial, participants managed invasively (INV; guideline-directed medical therapy [GDMT] + angiography ± revascularization) had better health status outcomes than those managed conservatively (CON; GDMT alone). It is unknown whether adherence to prescribed medications affects health status outcomes in chronic coronary disease (CCD). Hypothesis: Patient-reported medication non-adherence more adversely impacts health status outcomes in those treated with a CON, but not an INV, strategy. Methods: Self-reported adherence was assessed at randomization with a modified Morisky-Green-Levine medication adherence measure and patients were dichotomized into adherent (never skipping meds) and non-adherent (occasionally skipping meds). Twelve-month health status outcomes were assessed with the 7-item Seattle Angina Questionnaire (SAQ) Summary Score (SS; range 0-100, higher = better) using adjusted mixed effects cumulative probability models within a Bayesian framework to estimate differences between adherent and non-adherent patients, stratified by study arm. Results: Among 4480 randomized patients in the SAQ cohort, 1245 (27.8%) reported non-adherence at baseline. These patients had worse SAQ scores at baseline (CON: 72.9 ± 19.3 vs. 75.6 ± 18.4; INV: 71.0 ± 19.8 vs. 74.2 ± 18.7) and follow-up (CON: 81.2 ± 18.3 vs. 85.4 ± 15.0; INV: 84.3 ± 17.4 vs. 88.2 ± 13.9). The Figure shows similar posterior distributions of better SAQ scores with adherence in both arms (Mean difference = 1.63 (95% CrI 0.33, 2.89) in CON; 1.93 (95% CrI 0.80, 3.06) in INV). Conclusions: Medication non-adherence—as was reported in over 1 in 4 patients—was associated with worse 12-month health status, an effect that was similar in CON and INV strategies. While causation cannot be established, improving adherence may improve health status outcomes in CCD, regardless of treatment strategy.