Extreme environmental temperatures and motorcycle crashes: a time-series analysis

Zare Sakhvidi, Mohammad Javad; Yang, Jun; Mohammadi, Danial; FallahZadeh, Hussein; Mehrparvar, Amir Houshang; Stevenson, Mark; Basagaña Flores, Xavier; Gasparrini, Antoni; Dadvand, Payam

doi:http://dx.doi.org/10.1007/s11356-022-21151-8

Extreme environmental temperatures and motorcycle crashes: a time-series analysis

License

Citation

Zare Sakhvidi MJ, Yang J, Mohammadi D, FallahZadeh H, Mehrparvar A, Stevenson M, Basagaña X, Gasparrini A, Dadvand P. Extreme environmental temperatures and motorcycle crashes: a time-series analysis. Environ Sci Pollut Res Int. 2022 Oct;29(50):76251-62. DOI: 10.1007/s11356-022-21151-8

Abstract

Extreme temperature could affect traffic crashes by influencing road safety, vehicle performance, and drivers' behavior and abilities. Studies evaluating the impacts of extreme temperatures on the risk of traffic crashes have mainly overlooked the potential role of vehicle air conditioners. The aim of this study, therefore, was to evaluate the effect of exposure to extreme cold and hot temperatures on seeking medical attention due to motorcycle crashes. The study was conducted in Iran by using medical attendance for motorcycle crashes from March 2011 to June 2017. Data on daily minimum, mean and maximum temperature (°C), relative humidity (%), wind velocity (km/h), and precipitation (mm/day) were collected. We developed semi-parametric generalized additive models following a quasi-Poisson distribution with the distributed nonlinear lag model to estimate the immediate and lagged associations (reported as relative risk [RR], and 95% confidence interval [CI]). Between March 2011 and June 2017, 36,079 medical attendances due to motorcycle road traffic crashes were recorded (15.8 ± 5.92 victims per day). In this time period, the recorded temperature ranged from -11.2 to 45.4 °C (average: 25.5 ± 11.0 °C). We found an increased risk of medical attendance for motorcycle crashes (based on maximum daily temperature) at both extremely cold (1st percentile) and hot (99th percentile) temperatures and also hot (75th percentile) temperatures, mainly during lags 0 to 3 days (e.g., RR: 1.12 [95% CI: 1.05: 1.20]; RR: 1.08 [95% CI: 1.01: 1.16]; RR: 1.20 [95% CI: 1.09: 1.32] at lag0 for extremely cold, hot, and extremely hot conditions, respectively). The risk estimates for extremely hot temperatures were larger than hot and extremely cold temperatures. We estimated that 11.01% (95% CI: 7.77:14.06) of the medical attendance for motorcycle crashes is estimated to be attributable to non-optimal temperature (using mean temperature as exposure variable). Our findings have important public health messaging, given the considerable burden associated with road traffic injury, particularly in low- and middle-income countries.