A competing-risk approach to modelling length of stay in severe malaria patients in South-East Asia and the implications for planning of hospital services

Malaria burdens global health systems, and management with limited resources requires robust treatment guidelines and comprehensive planning. Expected length of stay (LOS) is useful in health-system planning, and factors influencing it can be targeted to reduce admission time and optimise service delivery.A secondary survival analysis of 1217 adult severe malaria patients from the South-East Asia Quinine Artesunate Malaria Trial, using a competing-risk approach.Median LOS was five days and time to discharge six days. 80% of patients were discharged, 70.2% within a week. 95.4% of deaths occurred within seven days. Compared to quinine, artesunate increased discharge incidence (subdistribution-hazard ratio 1.24 (1.09-1.40) p=0.001) and decreased incidence of death (0.60 (0.46-0.80) p<0.001). Cumulative incidence of discharge was decreased, and death increased, by low Glasgow coma scale (discharge: 1.08 (1.06-1.11) p<0.001, death: 0.85 (0.82-0.89) p<0.001), high blood urea-nitrogen (discharge: 0.99 (0.99-0.995) p<0.001, death: 1.00 (1.00-1.01) p=0.012), acidotic base-excess (discharge: 1.05 (1.03-1.06) p<0.001, death: 0.90 (0.88-0.93) p<0.001), and development of shock (discharge: 0.25 (0.13-0.47) p<0.001, death: 2.14 (1.46-3.12) p<0.001) or coma (discharge: 0.46 ( 0.32-0.65) p<0.001, death: 2.30 (1.58-3.36) p<0.001). Conventional Kaplan-Meier survival analysis overestimated cumulative incidence compared to competing-risk models.Clinical factors on admission and during hospitalisation influence LOS in severe malaria, offering targets to improve health and service efficiency. Artesunate has the potential to increase LOS, which should be accounted for in service-planning. Death in-hospital is a competing risk for discharge, and should be considered in LOS models to reduce overestimation of risk and misrepresentation of associations