ORIGINAL PAPER Impact of heat waves on nonaccidental deaths in Jinan, China, and associated risk factors Jun Zhang 1 & Shouqin Liu 1 & Jing Han 1 & Lin Zhou 1 & Yueling Liu 1 & Liu Yang 1 & Ji Zhang 1 & Ying Zhang 2,3 Received: 29 April 2015 /Revised: 22 December 2015 /Accepted: 24 December 2015 # ISB 2016 Abstract An ecological study and a case-crossover analysis were conducted to evaluate the impact of heat waves on nonaccidental deaths, and to identify contributing factors of population vulnerability to heat-related deaths in Jinan, China. Daily death data and meteorological data were collected for summer months (June to August) of 2012–2013. Excess mor- tality was calculated and multivariate linear regression models were used to assess the increased risk of heat waves on deaths. Univariate and multivariate logistic regression models were performed to estimate the odd ratios (ORs) of risk factors and their 95 % confidence intervals (CIs). Overall, heat waves were related to 24.88 % excess deaths of total nonaccidental deaths and 31.33 % excess deaths of circulatory diseases, with an OR of 16.07 (95 % CI 8.80–23.33) for total nonaccidental deaths and 12.46 (95 % CI 7.39–17.53) for deaths of circula- tory diseases. The case-crossover analysis indicated that older people were more likely to die during heat waves (OR = 1.233, 95 % CI 1.076–1.413) and more deaths occurred outside a hospital during heat waves (OR = 1.142, 95 % CI 1.006– 1.296). In conclusion, heat waves have caused excess deaths and significantly increased the risk of circulatory deaths. The risk factors identified in our study have implications for public health interventions to reduce heat-related mortality during extreme heat events. Keywords Heat wave . Risk factors . Death . Vulnerability . Case crossover Introduction Extreme heat and heat waves (HW) have caused hundreds of thousands of deaths over the past decades (Macpherson 2014; Steffen et al. 2014; UNISDR 2015). HW can not only cause direct death from heat stroke but also lead to excess deaths of other diseases (e.g., circulatory diseases and respiratory dis- eases). It has been projected that the frequency, duration, and intensity of HW are likely to increase in the future, which may result in a greater impact on human health (Bittner et al. 2014; Christopher et al. 2014; Steffen et al. 2014). An average of 400 deaths are counted as directly related to heat in the USA each year, with the highest death rates occurring in persons aged 65 years or more (Basu 2009). The global estimate for annual increases in heat-related deaths approximately around 90,000 in 2030 and more than 255,000 in 2050 (Hales et al. 2014). The effects of HW on human health depend on the charac- teristics of HW (including intensity, duration, and frequency), exposure access and time to HW and vulnerability of popula- tion (Christopher et al. 2014; Lin et al. 2011). Higher mortality risk from HWs was reported in the USA when HWs were more intense, with a longer duration, or occurring earlier in summer times (Anderson and Bell 2011). A study in 66 com- munities of China (Jinan not included) has suggested that a * Jun Zhang cdczhangjun@163.com * Ji Zhang zhangji1967@163.com * Ying Zhang ying.zhang@sydney.edu.au 1 Jinan municipal center for disease control and prevention, 2 Weiliu Road, Huaiyin District, Jinan 250021, China 2 Sydney School of Public Health/China Studies Center, The University of Sydney, Sydney NSW 2006, Australia 3 School of Public Health/Climate and Health Research Center, Shandong University, Shandong, China Int J Biometeorol DOI 10.1007/s00484-015-1130-7