Abstract
BACKGROUND: Hypertension and diabetes impose significant health and economic burdens in low-and middle-income countries like Bangladesh. This study aimed to estimate both direct and indirect costs associated with hypertension and diabetes care in Dinajpur, a rural district of Bangladesh, and identify factors influencing these costs. METHODS: This cross-sectional study used baseline data from a community survey conducted as part of an ongoing implementation research project. A multistage cluster sampling approach was used to randomly select adults aged 40 years and above from 45 wards across three subdistricts of Dinajpur. The analysis included 832 individuals who reported being on medication for hypertension (n = 635) and/or diabetes (n = 335). Data were collected through structured questionnaires, capturing direct (medical and non-medical) and indirect costs (productivity losses). Descriptive statistics, Wilcoxon rank-sum, and Kruskal-Wallis tests were used for univariate analyses. Multivariate linear regression models with log-transformed cost data were used to identify cost determinants. RESULTS: The average monthly total cost per patient was BDT 1,308 (USD 10.8) for hypertension and BDT 2,064 (USD 17.1) for diabetes. For both conditions, direct medical costs accounted for around 80% of total costs (60% for medicines), direct non-medical costs around 11% (mostly food and travel), and indirect costs approximately 9%. Direct costs were lower at public facilities compared to private dispensaries (Hypertension: GMR 0.46, 95% CI 0.33–0.64; Diabetes: GMR 0.15, 95% CI 0.10–0.24), while higher costs were observed for private clinics and NGO facilities. Level of education was associated with higher direct costs, particularly among patients with primary or secondary or higher education. Comorbidities were also associated with higher direct costs: in hypertension, cardiovascular disease (GMR 1.78, 95% CI 1.35–2.35) and high cholesterol (GMR 2.16, 95% CI 1.49–3.14) increased direct costs, with similar associations for diabetes costs. Indirect costs, reflecting productivity losses, were higher for private clinics, public facilities, and NGO facilities compared to private dispensaries. CONCLUSIONS: In the sample taken from a rural region from Bangladesh, hypertension and diabetes care entails a considerable financial burden, driven largely by medicine costs and reliance on private healthcare providers. Improved access to essential services and financial protection strategies are needed to reduce out-of-pocket expenditures. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12889-026-26456-8.