Effect of breastfeeding on neonatal mortality : A systemic review and meta-analysis of observational studies in Ethiopia

Objective: This systematic review and meta analysis was conducted to determine the pooled effect size and examine the inconsistencies among the studies conducted in Ethiopia. Design: We conducted a systemic review and meta analysis of identified observational studies. Data Source: An electronic bibliographic search was conducted in PUB Med, MEDLINE, EMBASE, Cochran Library data base and Google scholar from September 05, 2024 until November 07, 2024. Eligibility Criteria for selected studies: We included observational studies conducted in Ethiopia and published in English language. Intervention : Exclusive breastfeeding of neonates. Primary outcome measure : The primary outcome was neonatal mortality. Data extraction and synthesis: data extraction was independently conducted by two reviewer and. investigate research bias using Risk of Bias in Non- Randomized Studies of Intervention tool. This study adhere with the guidelines of Meta-Analysis of Epidemiological observational studies and Preferred Report Items for Systemic Review and Meta-Analysis. Random effect model was employed and statistical heterogeneity was assessed using the Q and I ² statistic. Result: Only 10 studies were included based on established criteria. The overall pooled effect measure demonstrated a statistically significant association [OR=0.24, 95%CI(0.03-0.45), P=0.02]. Heterogeneity, assessed using the Q statistical test, revealed [Q=97.46 , degree of freedom = 9 and P = 0.00] while, the I ² statistic ^{test showed [I 2 = 96%]. Publication bias of Begg’s funnel plot and Egger’s test , indicated an absence of publication bias [OR = 0.07, 95%CI ( -6.285 - 6.424) , P = 0.98]. Sensitivity analysis were deemed implausible. Certainty evidence of the studies was found to be moderate. Conclusion: Result of meta-analysis indicated that breastfeeding serves as a protective intervention to reduce risk of neonatal mortality in Ethiopia. Additionally, confirmed presence of heterogeneity among the studies. PROSPPRO protocol registration number : CRD42024599421}