Association between rainfall and pre-hospital traffic injury characteristics in Southern China: A multi-city time-series analysis in an E-bike dominated region

Objective: Precipitation acts as a significant environmental determinant of road safety, modifying traffic dynamics and injury mechanisms. From an environmental and public health perspective, understanding how rainfall interacts with specific transport modes—particularly in regions dominated by electric two-wheelers (E-bikes)—is essential for building climate-resilient emergency response systems. This study aimed to systematically quantify the association between rainfall intensity and pre-hospital traffic injury characteristics (severity and body region) in Southern China, providing empirical evidence to optimize healthcare resource allocation under adverse weather conditions. Methods: We conducted a retrospective multi-city time-series analysis in four geographically diverse cities (Baise, Beihai, Wuzhou, and Guilin) from November 2022 to November 2024. Daily pre-hospital trauma records were matched with satellite-derived meteorological data. A two-stage analytic strategy was employed: city-specific associations were first estimated using distributed lag non-linear models (DLNM) controlling for temperature, humidity, wind speed, and COVID-19 transitional periods, followed by a multivariate random-effects meta-analysis to pool regional estimates and assess spatial heterogeneity. Results: Among 23,572 trauma records, 5,470 traffic-related injuries were identified. Rainfall was associated with a significant increase in the daily volume of traffic injuries (Crude Risk Ratio [RR] = 1.30, 95 % CI: 1.10–1.54 [width = 0.44]). Analysis of injury specificity revealed that limb injuries maintained a robust positive association even after adjusting for meteorological confounders (Adjusted RR = 1.46, 95 % CI: 1.14–1.88 [width = 0.74]), whereas head and torso injuries showed no significant change. Regarding severity, rainfall was associated with a surge in the absolute volume of minor injuries (Crude RR = 1.33) but had no statistically significant impact on severe injuries (RR ∼ 1.0). Crucially, this stable absolute count of severe cases amidst the influx of minor trauma resulted in a significant ''dilution effect,'' where the relative proportion of severe cases decreased from ∼31.7 % to ∼5.9 % as rainfall intensified (Spearman’s ρ = -0.153, P < 0.01). Significant spatial heterogeneity was observed (I² = 34.3 %), with risks diverging between mountainous and tourism-oriented cities. Conclusions: In this subtropical region, precipitation acts as a specific environmental modifier, driving a surge in minor limb trauma that reflects the unique vulnerability of micromobility systems under wet conditions. These findings serve as a sentinel warning for global emerging economies, highlighting that weather-related trauma is a transnational environmental health priority requiring adaptive public health strategies and urban planning to mitigate the risks of a wetter, more mobile future.