District-level spatial clustering of multidrug-resistant/rifampicin-resistant tuberculosis notifications in Uganda, 2014–2023: a Local Moran’s I analysis

Multidrug-resistant and rifampicin-resistant tuberculosis (MDR/RR-TB) remains a major public health challenge in Uganda. Although geographic heterogeneity in MDR/RR-TB burden has been reported, little is known about the spatial dependence and clustering of MDR/RR-TB at district level. This study assessed spatial clustering dynamics of multidrug-resistant/rifampicin-resistant tuberculosis in Uganda between 2014 and 2023. We conducted a national ecological study using routinely collected MDR/RR-TB surveillance data obtained from Uganda’s District Health Information System 2 (DHIS2). Annual district-level MDR/RR-TB notification rates were calculated and spatially linked to administrative boundary data. Empirical Bayes smoothing was applied to stabilize rates. Local Moran’s I (LISA) was used to identify statistically significant hotspots, cold spots, and spatial outliers of MDR/RR-TB notifications. Statistical significance was assessed using 9,999 Monte Carlo permutations, with districts exhibiting p-values < 0.05 considered statistically significant. District-level analyses demonstrated statistically significant spatial clustering of MDR/RR-TB notifications during multiple years of the study period. Persistent High–High clusters were observed within and around urban and peri-urban districts, particularly the Kampala metropolitan area, including Kampala, Wakiso, and neighboring districts. Several districts consistently appeared as Low–Low clusters during the study period. Spatial outliers were also observed, suggesting localized heterogeneity in notification patterns. Temporal variation in clustering indicated evolving spatial dynamics over the ten-year period. MDR/RR-TB notifications in Uganda exhibit significant district-level spatial clustering. Persistent hotspots identified in Kampala, Wakiso, and Mukono districts highlight areas that may benefit from intensified surveillance, diagnostic strengthening, and targeted interventions. Routine incorporation of district-level spatial analysis into national tuberculosis surveillance systems could improve prioritization of resources and support more geographically targeted MDR/RR-TB control strategies.