Cofactors of neonatal mortality and hospitalization in western Kenya

Although global neonatal mortality has been declining over the past several decades, it remains a major contributor to infant and childhood mortality rates. Determining the biological, psychosocial, and behavioral cofactors of neonatal mortality and hospitalization can help identify high-risk populations and behaviors associated with adverse neonatal health outcomes and will be relevant to research and programmatic interventions offering perinatal support.Methods: The study cohort included 2451 maternal-infant pairs recruited from six clinics in western Kenya, who were control participants in the Mobile WACh NEO trial. Participants were enrolled during the third trimester of pregnancy and had follow-up visits at two and six weeks following delivery. Univariate and multivariate Cox regression models were used to test associations. Results: Cumulative neonatal mortality was 15.1/1,000 live births, while cumulative neonatal hospitalization was 19.2/1,000 live births. Higher gestational age and birthweight were associated with decreased risk of mortality in multivariate models (Hazard Ratio [HR] = 0.83, 95% confidence interval [CI] 0.74, 0.93; HR = 0.41, 95% CI 0.20, 0.83), while male sex and primigravida were associated with increased mortality in univariate models (HR = 2.24, 95% CI 1.05, 4.76; HR = 1.95, 95% CI 1.02, 3.72). Lower maternal age was marginally associated with increased risk of mortality in the univariate model (HR = 0.94, 95% CI 0.88, 1.00). Higher gestational age and birthweight were associated with decreased risk of hospitalization in the multivariate models (HR = 0.81, 95% CI 0.73, 0.91; HR = 0.39, 95% CI 0.21, 0.72), while primigravida and higher maternal education were associated with increased risk of hospitalization in univariate models (HR=1.82, 95% CI 1.02, 3.22; HR=1.96, 95% CI 1.02, 3.78). Among univariate models of infant care practices, appropriate cord care was associated with increased risk of neonatal mortality (HR = 3.505, 95% CI = 0.949, 12.946), while early initiation of breastfeeding was associated with a lower risk of hospitalization (HR = 0.160, 95% CI 0.04, 0.60). Conclusion: Biological and social cofactors were associated with increased neonatal mortality and hospitalization, consistent with published literature. Our findings underscore the need for interventions to prevent preterm or low birthweight outcomes or to enhance care of these infants. Maternal education and breastfeeding practices are modifiable factors that may be addressed to improve neonatal outcomes. Associations between cord care practices and neonatal outcomes may have been due to infant illness at delivery. Lack of association between psychosocial factors and adverse infant outcomes suggests that structural factors may exert more influence than individual-level factors.