Bayesian rank likelihood-based estimation: An application to low birth weight in Ethiopia

Area of Science:

• Public Health
• Biostatistics
• Demography

Background:

• Low birth weight (LBW) is a critical determinant of neonatal and infant mortality, especially in developing nations.
• Existing Ethiopian studies on LBW often suffer from small sample sizes and methodological limitations, potentially leading to biased results.
• Accurate nationwide estimates of LBW prevalence and its determinants are essential for effective public health strategies.

Purpose of the Study:

• To apply a novel Bayesian rank likelihood method within a latent trait model for nationwide LBW estimation in Ethiopia.
• To identify key risk factors associated with LBW across different regions of Ethiopia.
• To provide more accurate and reliable estimates compared to traditional statistical models.

Main Methods:

• Utilized data from the 2016 Ethiopian Demographic and Health Survey (EDHS), including 10,641 children aged 0-59 months.
• Employed a Bayesian rank likelihood approach within a latent trait model for parameter estimation.
• Evaluated model performance using metrics like root mean square error, mean absolute error, and probability coverage.

Main Results:

• The proposed model yielded superior estimates compared to classical methods.
• A significant prevalence of 40.92% for low birth weight was observed nationwide.
• LBW prevalence varied regionally, with higher concentrations in Afar, Somali, and SNNP regions, and lower variation in Addis Ababa, Dire Dawa, and Amhara.
• Maternal age, antenatal care visits, birth order, and maternal body mass index were significantly associated with LBW.

Conclusions:

• The Bayesian latent trait model provides more accurate LBW estimates than traditional methods in Ethiopia.
• Regional disparities in LBW prevalence necessitate geographically targeted interventions.
• Focusing on high-burden regions and addressing identified risk factors like maternal age and antenatal care is vital for reducing LBW and improving child health.