Title: Corruption in health sector procurement and its impact on essential medical supply availability
Authors: Dr. Twinomujuni Rosebell
Asiimwe Isaac Kazaara
Ahumuza Audrey
Volume: 10
Issue: 7
Pages: 192-201
Publication Date: 2026/07/28
Abstract:
Background: Corruption in public health procurement is widely recognized as a structural driver of medicine stockouts and diminished health system performance in low- and middle-income countries, yet facility-level quantitative evidence quantifying this relationship remains limited in the Ugandan context. Objective: This study examined the influence of corruption in health sector procurement on the availability of essential medical supplies across public and private-not-for-profit (PNFP) health facilities in Uganda, and specifically sought to determine the prevalence of corrupt procurement practices, assess the association between facility-level corruption exposure and essential medicine availability, and quantify the effect of corruption exposure on medicine availability while accounting for facility- and district-level covariates. Methods: A cross-sectional facility survey was conducted among 180 health facilities sampled through a two-stage stratified procedure across 15 districts. Data were collected on a composite Corruption Index, an Essential Medicines Availability Index (EMAI), specific corrupt practices, and facility- and district-level covariates. Analysis proceeded from univariate description, through bivariate tests (one-way ANOVA, Pearson correlation, chi-square, Kruskal-Wallis), to a linear mixed-effects regression with a random intercept for district to account for facility clustering. Results: Supplier favoritism (30.6%) and bribery solicitation (26.1%) were the most commonly reported corrupt practices, while ghost deliveries (12.2%) were least common but most severe. Mean EMAI declined monotonically across corruption exposure tertiles, from 90.9% (Low) to 83.5% (High) (F(2,177) = 12.39, p < 0.001; ?² = 0.123), and corruption index correlated negatively with EMAI (r = -0.417, p < 0.001). In the mixed-effects model, a one-unit increase in the corruption index was associated with a 0.275 percentage-point reduction in EMAI (95% CI: -0.381 to -0.169, p < 0.001), controlling for facility type, ownership, funding adequacy, supervision frequency, and procurement lead time. The intraclass correlation coefficient was low (ICC = 0.017), indicating that corruption's effect on medicine availability operated predominantly at the facility rather than the district level. Conclusion: Corruption in procurement processes was independently and significantly associated with reduced availability of essential medicines, even after accounting for legitimate systemic and facility-level determinants. Strengthening facility-level procurement transparency, supervision, and accountability mechanisms is likely to yield greater gains in medicine availability than district-level interventions alone.