Factors Associated with Outcomes of Inpatient Severe Malaria Cases in the Ashanti Region, Ghana: An Analytic Cross-sectional Study using Routine Surveillance Data, 2018 to 2022

Abstract Background Malaria remains a major global health threat, with 249 million cases and 609,000 deaths reported in 2022. The Ashanti Region of Ghana bears a disproportionate burden, with severe malaria accounting for 24% of hospital admissions in 2021, above the national average of 19%. Despite intensified control efforts, inpatient mortality patterns remain poorly understood. This study identifies key determinants of severe malaria mortality among hospitalized patients in the Ashanti Region. Methods We analyzed inpatient surveillance data from the District Health Information Management System 2 (DHIMS2) for severe malaria admissions from 2018 to 2022. Descriptive statistics, bivariate analyses with robust survey estimation (accounting for design effects), and multivariable Firth penalized logistic regression were used to identify mortality predictors. Survey-adjusted logistic regression served as a sensitivity analysis to validate findings. Results Among 54,544 severe malaria admissions, females comprised 51.1% and children under five 39.4%. The case fatality rate was 0.4% (200 deaths). Mortality was significantly associated with age, occupation, insurance status, facility ownership, admitting department, length of stay, and comorbidities. Males had 1.4 times higher mortality odds than females. Compared to children under five, patients aged 5–17 years had 44% lower odds of mortality (aOR = 0.56, 95% CI: 0.33–0.94). Active NHIS membership had lower mortality odds by 67% (aOR=0.33, 95% CI: 0.25–0.45) compared to inactive membership. Admissions to faith-based facilities showed lower mortality odds (aOR=0.38, 95% CI: 0.23–0.65) than government facilities, while medical wards had higher odds (aOR=2.38, 95% CI: 1.48–3.84) than paediatric wards. Stays of 3–5 days were associated with lower mortality odds (aOR=0.67, 95% CI: 0.47–0.97) compared to stays <3 days. Those with comorbidities had twice the mortality odds versus those without. Sensitivity analysis confirmed consistent direction and significance. Conclusion Age, comorbidities, insurance coverage, facility type, and admission practices strongly influence severe malaria mortality in Ashanti. Strengthening NHIS enrollment, extending inpatient monitoring beyond three days, and adopting best practices from paediatric and faith-based facilities could improve survival. Integrating comorbidity screening and management into malaria protocols is essential to reducing preventable deaths.