Informing the design of a trial of kangaroo mother care initiated before stabilisation amongst small and sick newborns in a sub-Saharan African context using mixed methods

An estimated 2.5 million neonates die every year, with preterm birth being the leading cause. Sub-Saharan Africa and southern Asia account for 78% of neonatal deaths. The WHO recommends kangaroo mother care (KMC) for stabilised newborns ≤2000g; however, most deaths occur before stabilisation. An evidence gap exists regarding KMC for this population. The overall aim of this PhD was to inform the design of a trial of KMC initiated before stabilisation in a sub-Saharan African context. The first part focused on assessing facility readiness and quantifying neonatal mortality risk. Cascade models were developed and used to assess 23 East African facilities. A logistic model was derived and validated using data from 187 UK hospitals and one Gambian hospital. The final model, including three parameters, demonstrated very good performance. The score requires further validation in low-resource contexts, but has potential to improve neonatal resource allocation. The second part of this PhD focused on evaluating the feasibility of initiating KMC before stabilisation and designing the trial. This study showed it was feasible to monitor and provide care in the KMC position, and found the intervention was acceptable to parents and providers. Launched in 2020, the OMWaNA trial will determine the mortality impact of this intervention within 7 days relative to standard care at four Ugandan hospitals. Process and economic evaluations will explore causal pathways for clinical effects, estimate incremental cost and costeffectiveness, and examine barriers and facilitators to inform uptake and sustainability. This PhD has developed a cascade model to assess facility readiness, validated a score to assess individual risk, and demonstrated the feasibility of initiating KMC before stabilisation. These studies have informed the design of a trial evaluating the mortality impact of this intervention in Uganda. The findings are expected to have broad applicability to low-resource hospitals and important policy implications.