Abstract
Optimising medication adherence is essential for effective heart failure (HF) management, yet nonadherence remains common, particularly among hospitalised and advanced-stage patients. This study evaluated the internal consistency reliability and score association of the Medication Adherence Report Scales, MARS-5 and MARS-10, self-report questionnaires in clinical pharmacist-led adherence assessments in hospitalised HF patients. Tools were administered during structured pharmacist-led interviews. To complement quantitative findings, four clinical cases were presented to illustrate the clinical relevance of adherence assessment in real--world HF management. Results showed a strong association between MARS-5 and MARS-10 (n = 70) responses (unweighted Cohen’s kappa 0.820 (95 % CI 0.683–0.957; p < 0.001) for categorizing patients as nonadherent or adherent and Pearson’s r coefficient of 0.899 (95 % CI 0.847–1.000; p < 0.001) for continuous score correlation), supporting score association and flexible use in clinical settings, with MARS-5 reliably identifying nonadherence (defined as score < 20 for MARS-5 and ≤ 8 for MARS-10), with potentially reduced respondent burden due to fewer items. Serious clinical complications were documented in nonadherent patients (41.43 % by MARS-5 and 35.71 % by MARS-10), illustrated through selected cases including stent thrombosis, embolic stroke, graft dysfunction, and deterioration in glycaemic control. These findings indicate the potential of MARS-5 as a practical, time-efficient tool for routine adherence assessment in acute settings. Case analyses underscore the critical role of the clinical pharmacist in proactively identifying nonadherence and enabling timely, targeted interventions to mitigate risk and improve patient outcomes in HF care.