Climate change increasingly threatens maternal, neonatal, and child health, yet evidence on the effect of temperature extremes on neonatal mortality in sub-Saharan Africa remains limited. This study quantified the association between maximum temperature and neonatal mortality and identified high-burden communities across 261 administrative districts in Ghana.

We conducted a longitudinal district-level panel analysis using monthly health and climate data from 261 districts between 2008 and 2025. Extreme temperature effects on neonatal mortality were estimated with Poisson, negative binomial, and quasi-Poisson models, using the number of deliveries as the offset. Counterfactual risk estimates were generated under a median maximum-temperature scenario to assess the contribution of above-median heat exposure to neonatal deaths.

A one-unit increase in combined maximum temperature across lags of 0–3 months was associated with a 9% increase in neonatal mortality rates (incidence rate ratio, 1.09; 95% CI, 1.04–1.13). Temperatures above 28.2 °C were linked to significantly higher neonatal mortality than median temperatures. Between 2008 and 2025, an estimated 652 neonatal deaths (95% CI, 516–788) were attributable to exposure above the median maximum temperature. Temperature-related risks varied substantially across districts and regions, with the greatest burden concentrated in Ghana’s five northern regions, where median monthly temperatures consistently exceed the national average.

Higher maximum temperatures were associated with increased neonatal mortality in Ghana. Targeted heat-adaptive interventions, including enhanced neonatal monitoring during hot months and early warning systems for pregnant women and nursing mothers, should be prioritized in high-exposure districts to effectively reduce preventable heat-related neonatal deaths nationwide.