Health insurance and differential healthcare utilization in Suryabinayak municipality, Nepal: a survey-weighted analysis using Andersen’s Behavioral Model

Nepal’s National Health Insurance Program (NHIP) was introduced to reduce financial barriers and improve healthcare utilization, yet the evidence on service-specific utilization patterns in rapidly urbanizing municipalities remains limited. The objective of this study was to estimate the association between NHIP enrollment and service-disaggregated healthcare utilization in a municipality in Nepal. A community-based comparative cross-sectional survey was conducted in five out of 10 randomly selected administrative wards of Suryabinayak municipality, Bhaktapur, from June to July 2025. We selected 301 households (150 insured and 151 uninsured) using a two-stage stratified random sampling. Healthcare utilization outcomes included overall service use, outpatient visits, diagnostic services, emergency care, inpatient admissions, and preventive services during the past six months. Data were analyzed using survey-weighted multivariable logistic regression, with confounders identified using a directed acyclic graph; robustness was assessed using Firth’s penalized regression, inverse-probability-of-treatment weighting, and sensitivity analyses. Overall healthcare utilization was significantly higher among insured than uninsured households (74.7% vs. 53.6%; p < 0.001), as were diagnostic, outpatient, emergency, and inpatient services (all p < 0.05); preventive service use did not differ ( p = 0.617). In adjusted models, NHIP enrollment was associated with higher odds of overall utilization (aOR 2.69, 95% CI 1.56–4.63), outpatient (aOR 2.13), diagnostic (aOR 2.57), emergency (aOR 2.81), and admission services (aOR 2.21), robust across sensitivity analyses. NHIP was not associated with preventive service use overall (aOR 1.37), but this varied by socioeconomic status: a positive association among wealthier households (aOR 3.40) was absent among poorer households (aOR 0.40). Concentration index analysis confirmed pro-rich inequality in utilization; NHIP reduced this inequality for curative services but not for preventive care or hospital admission. Chronic illness (aOR 3.90) and higher socioeconomic status (aOR 3.39) were the strongest independent predictors of overall utilization. NHIP enrollment was associated with increased utilization of curative and diagnostic services but showed no overall association with preventive care use, although this association varied by socioeconomic status. Findings suggest a need for equity-focused enrollment strategies and benefit-package reforms that specifically target preventive care access among disadvantaged households.