Abstract
BACKGROUND AND AIMS: Adherence to medication is key for effective treatment, especially for chronic illnesses. Poor adherence leads to more hospital visits, worse health outcomes, and higher healthcare costs. Many tools exist to measure adherence, but they often lack cultural sensitivity or only fit specific diseases. This study aimed to create and validate a culturally relevant tool for assessing medication adherence in Bengali-speaking patients with chronic illnesses. METHODS: This was a cross‑sectional scale‑validation study. We used a multi‑step approach comprising item generation, expert review, translation, pre‑testing, and psychometric validation. An initial pool of 27 items was refined to a final 9‑item Bangladesh Medication Adherence Scale (BMAS). Data from 300 patients at Bangladesh Medical University (BMU) were analyzed using IBM SPSS Statistics, version 29.0 for exploratory factor analysis and descriptive statistics, and SmartPLS 4 for confirmatory factor analysis and structural equation modeling. The 9‑item scale was then cross‑validated in an independent sample of 155 patients with hypertension from a district hospital. RESULTS: Exploratory factor analysis revealed a three-factor structure that explained 63.8% of the variance. Confirmatory factor analysis showed a strong model fit, with most factor loadings above 0.70. The scale had high internal consistency (Cronbach's α = 0.889) and strong composite reliability (ρC = 0.898). Convergent and discriminant validity were confirmed by the Fornell-Larcker criterion and Heterotrait-Monotrait (HTMT) analysis. Nomological validity was supported by links to health locus of control and quality of life measures. CONCLUSION: These findings indicate that the BMAS is a reliable and valid instrument for assessing medication adherence among chronically ill patients in Bangladesh. Its use in clinical practice and public health research could support better adherence monitoring and inform interventions to improve treatment outcomes.