Empowering women to brighten health: Unlocking the energy-gender-health nexus through machine learning in Sub-Saharan Africa

Sustainably access to energy, combined with social development, involves indivisible interactions. This necessitates a contemporary yet rarely attained understanding of the interplay among three United Nation's Sustainable Development Goals (SDGs): clean energy (SDG7), gender equality (SDG5), and good health (SDG3), particularly in the least developed regions—Sub-Saharan Africa. Leveraging large-scale household survey dataset from over 44,000 households across 16 countries, we apply various econometric models and machine learning algorithms to systematically and robustly explore the complex interactions among energy poverty, gender equality, and women's health. By establishing the Multidimensional Energy Poverty Index (MEPI) and the Women's Empowerment Index (WEI), we reveal a leading role of gender equality in improving health outcomes impaired by energy poverty, with the light gradient boosting machine (lightGBM) model demonstrating the highest accuracy rate. Notably, our findings indicate that coordinated efforts towards SDG7 and SDG5 significantly advance SDG3. Furthermore, we provide three-dimensional visualizations fitted through the random forest (RF) model to transparency the nexus. Finally, we assess synergies among the three SDGs using the coupling coordination degree model, with results remaining robust, revealing a relatively high synergistic grade among them. These findings call for coherent, gender-responsive policies that prioritize efficient resource allocation to women, unlocking simultaneous energy-gender-health gains, thereby accelerating Africa's broader SDGs through a just energy transition.