Associations of compound hot extremes and heat waves with first-ever stroke morbidity in the context of climate change

Climate change has increased the frequency and intensity of abnormal weather, with current daytime and nighttime temperatures being higher than their historical counterparts. Previous studies have focused on exploring the health hazards of absolute heat (above the optimum temperature, often calculated on the basis of short periods of current data). However, the health hazards of climate change–induced relative heat (above the extremes of historical counterparts, often calculated on the basis of 30 years of temperature data or more) are unclear. Therefore, this study aims to explore the associations of different types (daytime only, nighttime only and combined daytime–nighttime) of heat and heat waves with stroke morbidity in consideration of human climate adaptation. The data of patients with stroke were obtained from Shenzhen, China, for the period of 2003–2018. Daytime and nighttime heat thresholds for specific calendar days in the study period were defined on the basis of hourly temperatures for long-term counterparts, which were the day and 7 d before and after each calendar day in the historical baseline (1973–2002). The associations of different types of heat and heat waves defined by bivariate heat thresholds with stroke morbidity were explored by using distributed lag nonlinear models. Relevant vulnerable populations and sensitive disease subtypes were identified through stratified analyses. Compound hot extremes and heat waves (combined daytime and nighttime heat and heat waves) were associated with stroke morbidity, with relative risks (RRs) of 1.279 (95% confidence interval (CI): 1.078, 1.519) and 1.500 (95% CI: 1.142, 1.969), respectively, and attributable fractions (AFs) of 1.658% (95% CI: 0.548%, 2.594%) and 0.970% (95% CI: 0.362%, 1.432%), respectively. Associations between heat and heat waves during daytime only and nighttime only with stroke morbidity were statistically insignificant. Males, females and adults aged under and over 65 years were vulnerable to compound hot extremes and heat waves, and the differences between subgroups were statistically insignificant. Ischaemic stroke was the subtype sensitive to compound hot extremes and heat waves with RRs of 1.338 (95% CI: 1.101, 1.626) and 1.553 (95% CI: 1.138, 2.119), respectively, and AFs of 1.956% (95% CI: 0.709%, 2.982%) and 1.064% (95% CI: 0.363%, 1.578%), respectively, whereas haemorrhagic stroke had statistically insignificant associations. Compound hot extremes and heat waves may lead to an increased risk of stroke morbidity in the context of climate change. Governments should emphasise the forecasting and warning of compound hot weather with temperatures higher than the extremes of long-term historical counterparts to reduce associated disease burdens.