Abstract
BACKGROUND: Current recommendations for neonatal bed requirements are largely assumption-based rather than data-driven. We aimed to estimate the number of beds per 1000 live births needed for the care of small and sick newborns. METHODS: We first extracted data from studies published between 2018 and May 2023. Then, due to considerable heterogeneity in the data, we performed a meta-analysis using a random effects model to estimate the number of neonatal admissions and the length of stay. We divided the total patient days (admission rate multiplied by the length of stay) by 365 to estimate the annual number of beds per 1000 live births. RESULTS: We include 54 included studies, of which 46 provided data on the incidence of neonatal admissions and 20 on length of stay. The pooled analysis indicated that the number of neonates requiring admission ranged from 126 to 143 per 1000 live births. Admission rates were higher in the African region (160.5; 95% confidence interval (CI) = 122.2-198.7), in low-income countries (175.3; 95% CI = 102.8 to 247.8), in tertiary care settings (147.5; 95% CI = 115.9-179.1), and in settings with a high neonatal mortality rate (149.4; 95% CI = 90.5-218.2). The pooled length of stay was estimated to be 6.4 days (95% CI = 5.7-7.1). The overall estimated number of beds needed for the care of small and sick newborns was 2.4 (95% CI = 2.0-2.8) per 1000 live births, with regional variations. CONCLUSIONS: This method estimates the required neonatal care beds using admission rates and hospital stay data, aiding healthcare planning. Refinements and local adaptations are needed for effective policy decisions. REGISTRATION: PROSPERO: CRD42023417847.