Abstract
<title>Abstract</title> <p> Background Severe malaria remains a leading cause of hospitalization and death among children under five years of age in sub-Saharan Africa. However, evidence on survival patterns and independent prognostic factors using time-to-event analysis remains scarce in the Democratic Republic of the Congo. This study aimed to describe survival patterns and identify prognostic factors associated with in-hospital mortality among children hospitalized with severe malaria in Mbujimayi. Methods A multicenter retrospective cohort study was conducted in ten General Referral Hospitals in Mbujimayi, Democratic Republic of the Congo, between January and December 2025. A total of 674 children aged 0–59 months meeting the World Health Organization criteria for severe malaria were included. Survival probabilities were estimated using the Kaplan–Meier method and compared with the Log-rank test. Independent prognostic factors were identified using multivariable Cox proportional hazards regression. Results Among the 674 enrolled children, 112 died, yielding an in-hospital case fatality rate of 16.6%. Most deaths occurred early during hospitalization, with 42.0% occurring on the first day and 82.1% within the first three days. The estimated mean survival time was 22.4 days (95% CI: 21.5–23.3). Multivariable Cox regression identified age below 24 months (adjusted HR = 1.564, 95% CI: 1.071–2.285; <italic>p</italic> = 0.021), blood transfusion (adjusted HR = 2.200, 95% CI: 1.494–3.239; <italic>p</italic> < 0.001), coma (adjusted HR = 2.993, 95% CI: 1.783–5.025; <italic>p</italic> < 0.001), and acute pulmonary edema (adjusted HR = 2.862, 95% CI: 1.618–5.065; <italic>p</italic> < 0.001) as independent predictors of in-hospital mortality. Conclusions Mortality among children hospitalized with severe malaria occurred predominantly during the first 72 hours after admission. Early identification of children younger than 24 months and those presenting with coma or acute pulmonary edema, together with prompt intensive supportive care, may substantially improve survival in resource-limited settings. </p>