The Impact of Different Factors on Infant Mortality: A Meta-Analysis on Developing Countries

Background
The infant mortality rate is viewed as perhaps the most grounded pointer to a nation's prosperity, as it reflects social, monetary, and ecological conditions. Estimation of 2017, among 6.3 million children who died in most cases from preventable causes, about 1.6 million of these deaths occurred in infancy (age 1–11 months). In this way, it is a significant region of research.
Objective
This study aimed to discover the prevalence of infant mortality and to assess how different
factors influence infant mortality in developing countries including Bangladesh by utilizing the
latest DHS data.
Methods
The most recent available DHS data of 24 different developing countries including Bangladesh from the year 2013 to 2018 is used to conduct this study. A logistic regression model is used to explore
the risk factors of infant deaths in Bangladesh; a random-effects meta-analysis is applied to
observe the differences in the influencing factors from study to study for developing countries. Furthermore, subgroup analysis is performed for making comparisons among different geographical areas (African vs. Asian) to evaluate the consistency of treatment.
Results
The prevalence of infant mortality is found highest in Sierra Leone (76deaths per 1000 live
births) among 24 developing countries. A logistic regression model reveals that for
Bangladesh, higher education level for fathers (OR: 0.344; CI: 0.147;0.807), middle-income
family (OR: 0.408; CI: 0.254;0.654), being 2nd born and above order infant (OR: 0.362; CI:
0.248,0.527), above 19 years age of mothers at delivery (OR: 0.477; 0.333;0.682),taking ANC
(OR: 0.271; CI: 0.192;0.382 for 1-4 visits), taking PNC (OR: 0.303; CI; 0.216;0.425) are
statistically significant determinants of lowering infant death. While carrying multiple fetuses
(OR: 6.634; CI: 3.247; 13.555) is exposed as a risk factor for infant mortality. These findings are
supported by meta-analyses for the developing countries as well. Additionally, meta-analyses
detect the size of the child, type of residence, and maternal education and working status as risks factor for infant mortality. Besides, subgroup analysis reveals that taking ANC and PNC is a better indicator for lowering infant mortality among Asian developing countries compared to the African developing countries.
Conclusion
In this study, parents' education level, wealth status, birth order number, maternal age at
delivery, number of fetuses, taking ANC, PNC, and size of the child are found to be significant factors for both the logistic and meta-analytical methods that affect the risk of infant mortality in developing countries except, type of place of residence, mother’s current working status has no impact on infant mortality for Bangladesh. So anticipation and control projects ought to be
taken in the field in regard to these hazard factors. Sub-group analysis uncovers that, taking
ANC visits are a better indicator for lowering infant mortality among Asian developing countries contrasted with the African developing countries. Therefore, women should be encouraged to have ANC visits during their pregnancy period. Close follow-up and monitoring should also be given for the women who carry multiple fetuses.

• Supervision