Abstract Background Malaria transmission in Tanzania has declined significantly over the last two decades due to scaled-up control interventions. However, recent confirmation of artemisinin partial resistance (ART-R) in Kagera region in northwest Tanzania threatens the ongoing efforts to eliminate malaria in the country. This study was conducted according to the World Health Organization recommendation to generate evidence of malaria burden in areas with confirmed ART-R as the first step before developing a response strategy to the resistance. Methods We assessed the local burden of malaria in Kagera region by geospatial analysis, using data collected retrospectively from health facilities (HFs) and community surveys from 2015 to 2023 to identify malaria hotspots. Results From 2017 to 2023, a total of 8,124,363 suspected malaria cases were reported by HFs, and 2,983,717 (36.7%, 95% range across wards: 22.7 - 50.7%) tested positive by rapid diagnostic tests. Test positivity rates were similar among under-fives (33.1%, 95% range: 19.7–46.5%) and patients aged ≥5 years (33.7%, 95% range: 21.0–46.5%). Malaria prevalence was 10.0% (95% range across wards: 5.1–14.9%%, n=84,999/853,761) in pregnant women and 26.1% (95% range across wards: 11.7–40.6%, n=3,409/13,065) in school children. Despite high temporal variations, we identified hotspots and coldspots, including persistently high burden in 69/192 (35.9%) wards. Conclusion Malaria burden in Kagera exhibited high temporal and spatial heterogeneity, with school children showing the highest prevalence. This demographic pattern underlines the need for targeted interventions and provides evidence for developing an ART-R response for the region.