Neonatal admissions in a conflict-affected setting: an eight-year multi-center retrospective cohort study from Taiz, Yemen (2018–2025)

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Abstract Background Neonatal mortality in conflict-affected regions is a critical and under-documented public health issue. This study aimed to characterize admission patterns, diagnoses, and outcomes of neonates in Taiz, Yemen, over an eight-year period. Methods We conducted a multi-center retrospective cohort study. All neonates ( N = 1,936) admitted to five major hospitals in Taiz governorate between January 1, 2018, and December 31, 2025, were included. Data were extracted from hospital registries. Primary outcomes were in-hospital mortality and discharge against medical advice (DAMA). Multivariable logistic regression was used to identify independent predictors of these outcomes. Findings The cohort was 57.2% male. Most admissions (71.3%) occurred within 24 h of birth. The predominant diagnoses were birth asphyxia (27.8%), prematurity (24.3%), and neonatal sepsis (15.6%). Overall in-hospital mortality was 12.1% (235/1,936), with the highest rates among neonates with congenital anomalies (34.3%). However, in absolute terms, three preventable conditions—prematurity, birth asphyxia, and neonatal sepsis—accounted for 79.1% of all deaths. DAMA occurred in 19.8% (383/1,936) of admissions. Independent predictors of mortality were congenital anomalies (aOR 3.12, 95%CI 1.48–6.58) and admission at 8–27 days (aOR 1.89, 95%CI 1.16–3.08). Independent predictors of DAMA were admission at 1–7 days (aOR 1.58, 95%CI 1.21–2.06) and respiratory distress (aOR 1.86, 95%CI 1.15–3.01). Interpretation This study demonstrates a high burden of preventable neonatal conditions in conflict-affected Taiz. The findings highlight critical gaps in perinatal care and underscore the need for urgent, context-specific interventions, including scaling up neonatal resuscitation training, strengthening referral systems, and addressing the high DAMA rate through targeted family support. The results likely underestimate the true community burden of neonatal mortality, as they only capture facility-born infants.