A multi-dimensional index of rurality in Chile and its association with infant and neonatal mortality

OBJECTIVE: To construct a multi-dimensional Index of Relative Rurality (IRR) for Chile and assess its association with infant and neonatal mortality. MATERIALS AND METHODS: A cross-sectional analysis was conducted using data from 343 municipalities. The IRR was calculated using four standardized components-population size, population density, urban population proportion, and travel time to service-rich areas-to reflect the complexity of rurality in Chile. Associations between the IRR and mortality outcomes were evaluated using negative binomial regression and generalized additive models. RESULTS: A rurality threshold was identified at approximately IRR= 0.404, beyond which the risk of infant and neonatal mortality increased notably. Municipalities with IRR values >0.50 showed elevated risks of infant mortality (RR: 1.18; 95%CI: 1.01,1.38) and neonatal mortality (RR: 1.27; 95%CI: 1.06,1.50), compared to more urban areas. While the official classification defined nearly half of municipalities as rural, the IRR captured a smaller, more specific subset of municipalities with higher mortality risk-particularly in southern macrozones. CONCLUSIONS: The IRR offers a more precise and data-driven approach to characterizing rurality and identifying health disparities. Its continuous design improves the detection of high-risk populations and challenges the overgeneralization of current classifications. These findings support the use of the IRR to inform targeted public health interventions and equitable policy planning.