Analyzing the influence of energy poverty on newborn mortality through econometric and machine learning approaches

Energy poverty is a critical issue globally, especially in low and lower-middle-income countries. While its effects on human health and well-being have been extensively studied, there is limited research on how it specifically impacts newborn mortality. This study seeks to fill this gap by examining the relationship between energy poverty and newborn survival in these nations. By employing econometric and machine learning (ML) technique, the study provides a comprehensive examination of this issue. Our results based on 42 low and lower-middle-income countries during the period 2001–2020 show a consistent decrease in stillbirth, neonatal, and infant mortality rates with improved energy use. Specifically, energy poverty is associated with a 13.16 % increase in stillbirths, a 9.38 % rise in neonatal deaths, and an 8.56 % increase in infant mortality. This study also demonstrates the effectiveness of ML prediction modeling, achieving over 92 % accuracy in predicting stillbirth, neonatal mortality, and infant mortality, and revealing that energy poverty significantly increases these mortality rates. Furthermore, the mediation analysis shows that energy poverty affects newborn mortality through factors like low birthweight, air pollution, and human development. This understanding highlights the urgent need for targeted actions to reduce energy poverty in the affected countries. Policymakers are urged to promote the adoption of clean cooking technologies, alternative energy sources, and raise awareness regarding the adverse effects of energy poverty.