Burden and outcome of neonatal diseases and conditions at a rural district hospital setting in Kenya

Although the common causes of neonatal morbidity and mortality are known, there is paucity of information on many aspects that are required for effective management of neonates presenting for care in health facilities in low-income settings. In this thesis, I set to bridge this knowledge gap by looking at the neonatal admissions burden, assessing the ability of clinical signs to identify very ill neonates likely to die, examining the causes of infections among neonates, and describing the long-term post admission outcomes. I found that neonates may comprise close to 20% of all under five years old children admitted to rural health facilities in low income settings and contribute nearly 60% of inpatient childhood deaths. Importantly, simple clinical signs are useful in identifying neonates at risk of death during admissions and could help target care in many conditions including invasive bacterial infections and hypoxemia. Crucially, up to 40% of all neonates admitted with severe illness (injuries during birth, bacterial infections & prematurity) suffer long-term neurological deficits post admission. However, data on long-term outcome is very poor in low-income countries. Urgent interagency approaches to improving data are therefore needed. In summary, in this thesis I demonstrate that it may be necessary to remodel our approach on how best to prevent neonatal morbidity and mortality in low-income countries. I suggest a more robust five-prong model that build upon the existing three-delay model