Geodetector analysis of individual and joint impacts of natural and human factors on maternal and child health at the provincial scale

Abstract

This ecological study examined the individual and joint impacts of natural–human factors on the spatial patterns of maternal and child health status in China at the provincial scale in 2020. We considered natural factors (forest coverage, average temperature, and total sulfur dioxide and particulate matter emissions) and human factors (economic development, urbanization, healthcare access, and education level). We combined maternal, infant, and under-five mortality rates into a composite maternal and child health index using the entropy method. The spatial autocorrelation analysis of this index highlighted distinct health patterns across provinces, whereas the geodetector method assessed the effects of natural–human factors on the patterns. A notable east–central–west stepwise decline in health status was observed. Global Moran’s I showed positive spatial clustering, with high–high clustering areas in the Yangtze River Delta and low–low clustering areas in western regions. Factor detection identified eight significant natural–human factors impacting maternal and child health, with total sulfur dioxide emission density having the greatest impact. The interaction between average schooling years and total sulfur dioxide emission notably affected maternal and child health patterns. The study concludes that natural–human factors critically affect the spatial distribution of maternal and child health.