Abstract
IMPORTANCE: Although the immediate impact of neonatal illness is well recognized, its wider and longer term outcomes on childhood mortality and the role of specific illnesses across childhood are unclear. OBJECTIVE: To investigate how many deaths in childhood are associated with neonatal illness and the underlying conditions of the children who died. DESIGN, SETTING, AND PARTICIPANTS: This population-based cohort study of children who died before age 10 years in England between April 1, 2019, and March 31, 2021, used data from the National Child Mortality Database. Data analysis was performed from September 2022 to May 2023. EXPOSURE: Children who received care in a neonatal unit after birth plus those who died in the first day of life, before admission to a neonatal unit, were considered to have likely neonatal illness. MAIN OUTCOMES AND MEASURES: The primary outcome was the relative risk (RR) of dying, stratified by likely neonatal illness and specific neonatal conditions. Comparisons were made using the χ2 or likelihood ratio test, as appropriate. RESULTS: A total of 4829 children were included (median [IQR] age at death, 28 [2-274] days; 2606 boys [54.8%]; 2690 White children [64.0%]). Overall, 3456 children who died (71.6%) had evidence of likely neonatal illness. Children with neonatal illness were more likely to die before their tenth birthday than those without evidence of neonatal illness (RR, 13.82; 95% CI, 13.00-14.71). The estimated population-attributable risk fraction for neonatal illness among all deaths before age 10 years was 66.4% (95% CI, 64.9%-67.9%). Children with preceding neonatal illness who died were more likely to have underlying behavioral or developmental disorders (odds ratio [OR], 3.31; 95% CI, 2.47-4.42), chronic neurological disease (OR, 3.00; 95% CI, 2.51-3.58), and chronic respiratory disease (OR, 3.01; 95% CI, 2.43-3.73) than children without neonatal illness. CONCLUSIONS AND RELEVANCE: In this cohort study, most children who died before age 10 years had some evidence of neonatal illness, and they died of a range of causes, including infections and sudden, unexpected, unexplained death. These findings suggest that improvements to perinatal morbidity, an area with an existing evidence base for improvement, may have important impacts on child health across the next decade.