Epidemiology of malaria and district-level factors associated with malaria elimination in Sumatra region, Indonesia: a retrospective analysis of surveillance data

Abstract Background As transmission declines, malaria becomes more spatially and demographically concentrated. This study investigated individual-level malaria characteristics and district-level factors associated with elimination in all 154 districts in Sumatra, Indonesia. Methods We analysed individual-level data of 13,386 reported malaria cases from 2019 to 2021. P. falciparum and P. vivax were confirmed by microscopy or rapid diagnostic test, while probable P. knowlesi was defined as microscopically identified P. knowlesi . Cases were classified as imported if they reported travel to endemic region within 2–4 weeks. Multivariable logistic regression with province as a random effect was used to examine individual-level factors associated with imported cases, and district-level socioeconomic and environmental factors associated with elimination. Spatial clustering of indigenous cases was assessed using Global Moran’s I and SaTScan. Results Among reported cases, 71.8% (9,617) were P. vivax , 18.4% (2,467) P. falciparum , and 2.8% (369) probable P. knowlesi . Of 11,390 investigated cases, 8,748 (76.8%) were indigenous, 1,741 (15.3%) imported, and 901 (7.9%) unclassified. Imported cases were associated with older age (adjusted odds ratio 1.79, 95%CI 1.42–2.25 for 15–24 years old, and 1.97, 1.59–2.46 for 25–64 years old, compared to < 15 years old), males (1.70, 1.43–2.03), mobile and migrant population-related occupations (1.46, 1.22–1.76), diagnosed at hospitals or clinics (2.32, 1.85–2.91), infected with P. falciparum (1.48, 1.22–1.80), and hospitalised (15.29, 12.66–18.46). At the district level, a higher human development index (1.15, 1.01–1.34), lower historical annual parasite incidence (0.46, 0.29–0.69), and fewer endemic neighbours (0.78, 0.67–0.89) were associated with a higher likelihood of elimination. Spatial analysis revealed clustering tendencies, with localised clusters for P. falciparum , P. vivax , and probable P. knowlesi . Conclusions Dynamics of imported malaria, heterogeneous clusters of local transmission, and lower district-level human development index challenged elimination efforts. This highlights the need for strengthened surveillance and targeted interventions.