787Multistate modelling to investigate the impact of recurrent malaria episodes on hospital admissions and mortality

Abstract Background Inadequate prevention and treatment of malaria can lead to reinfections and recurrent episodes, and for vivax malaria, further recurrences from the dormant liver stage. This study quantified the impact of recurrent malaria episodes on morbidity and mortality. Methods Routinely collected data were available from 68,381 malaria patients presenting to the primary referral hospital in Papua, Indonesia. A multi-state modelling framework, with Cox regression for transition rates, was employed to determine the risks of re-presentation to hospital, receiving in-patient treatment, and early (≤14 days post treatment)/late death following multiple malaria episodes. Results The risk of re-presentation to hospital increased from 34.7% (95%CI: 34.4%–35.1%) at first episode to 58.6% (57.5%–59.6%) following the third episode. Infection with vivax malaria increased the rate of re-presentation to hospital by 1.48-fold (Hazard Ratio 1.48; 95%CI 1.44–1.51) and late hospital in-patient admission by 1.17-fold (1.11–1.22), compared to falciparum. Falciparum malaria caused a higher overall rate of early death (1.54 (1.25–1.92)), however, after multiple episodes, there was a trend towards a greater rate of early death for vivax infection (1.91 (0.73–4.97)). Conclusions Recurrent episodes of malaria can cause substantial morbidity and mortality, highlighting the importance of prevention and effective treatments for both falciparum and vivax malaria. Key messages To achieve elimination of malaria in South-East Asia, where prevalence of vivax malaria is high, we must prioritise the radical cure of vivax to eliminate the liver-stage of this species that causes relapses of infection.