Causes of mortality and severe morbidity in young children in Sierra Leone: a prospective cohort study.

backgroundAccurate data on morbidity and mortality in children younger than 2 years remain scarce in many low-income and middle-income countries. We aimed to examine the causes and risk factors for severe morbidity and mortality in young children in Sierra Leone.

methodsWe conducted a prospective cohort study nested within a randomised controlled trial (NCT04235816). The study was done in 14 primary health facilities in three districts in Sierra Leone. We enrolled infants aged 6-10 weeks, eligible for pentavalent vaccine dose 1 through the EPI programme, weighing at least 2·5 kg, resident in the study area, without known contraindications to macrolides or sulfadoxine-pyrimethamine, between March 17, 2021, and April 25, 2024. The objective was to assess causes of hospital admission and death through to Dec 31, 2024, when approximately 80% of children had completed 18-month follow-up. Hospital admissions, deaths, and immunisation status were captured through facility surveillance, household visits, telephone follow-ups, and community informants. Causes of death were classified using ICD-10 and the WHO 2016 verbal autopsy tool (InSilicoVA). Incidence rates were expressed per 1000 child-months at risk. Morbidity was analysed using negative binomial regression; in-hospital mortality was analysed using penalised logistic regression, and post-discharge mortality was analysed using Cox proportional hazards regression.

findingsData on severe morbidity and mortality were collected from 20 560 children, of whom 10 644 (52%) were male and 9916 (48%) were female; the most common ethnicity was Temne (14 048 [68%] of 20 551 children with available data). Median follow-up was 16·6 months (IQR 16·0-16·7). There were 2688 hospital admissions (excluding readmissions and in-hospital deaths), yielding an incidence of 8·67 per 1000 child-months at risk (95% CI 8·35-9·00). Malaria (1459 [54·3%] of 2688 admissions), lower respiratory tract infections including pneumonia (449 [16·7%]), diarrhoea (259 [9·6%]), undernutrition (198 [7·4%]), and sepsis (128 [4·8%]) were the leading causes of admission. The leading causes of in-hospital deaths were malaria (122 [50·6%] of 241), lower respiratory tract infections (41 [17·0%]), sepsis (33 [13·7%]), undernutrition (17 [7·1%]), and diarrhoea (nine [3·7%]). The leading causes of community deaths for which a verbal autopsy was performed were malaria (103 [24·2%] of 425), lower respiratory tract infections (76 [17·9%]), other infectious diseases (73 [17·2%]), and diarrhoea (54 [12·7%]). Post-discharge mortality was 33·8 per 1000 child-months at risk (95% CI 27·08-42·20). Being underweight was associated with higher in-hospital mortality compared with not being underweight (adjusted odds ratio [OR] 1·60 [95% CI 1·21-2·12]). Full immunisation was associated with lower overall mortality (adjusted hazard ratio [HR] 0·65 [95% CI 0·55-0·76]), lower in-hospital mortality (adjusted OR 0·71 [95% CI 0·54-0·95]), and lower post-discharge mortality (adjusted HR 0·56 [95% CI 0·33-0·92]).

interpretationThis study shows that despite major investments and implementation efforts, mortality in infants and young children in Sierra Leone remains high and predominantly caused by preventable infections. These findings highlight the substantial unfinished agenda in global health and the urgent need for more effective tools and strategies.

fundingGates Foundation and La Caixa Foundation.