| Objective 1)To analyze climatic trend in 15 Chinese capital cities during 1951-2013;2)To examine the death burden of CVD attributable to ambient temperature,and to assess the extent to which the seasonal variation of CVD deaths caused by temperature;3)To examine the health impact of short-term variation of temperature,including DTR and TV,on CVD,and to test the effect modification of this burden according to gender,age and education level;4)To detect the definitions for extreme temperature events based on their health impacts,and to identify the vulnerable subpopulations.5)To quantify the future temperature-related CVD excess death burden under 16 GCM models and four RCP scenarios.Methods 1)Mann-Kendall test and Sen's slope method were applied to assess the long-term trends of temperature variables in 15 Chinese capital cities during 1951-2013;2)We obtained daily data on temperature and CVD mortality from 15 Chinese mega-cities during 2007-2013,including 1,936,116 CVD deaths.A quasi-Poisson regression combined with distributed lag non-linear model(DLNM)was used to estimate the temperature-mortality association for each city,adjusting for the potential confounding variables.Then,a multivariate meta-analysis was used to derive the overall effect estimates of temperature at national level.Attributable fraction of deaths were calculated for cold and heat(i.e.temperature below and above minimum-mortality temperatures,MMT),respectively.Furthermore,the seasonal attributable fraction(AF)was calculated by dividing the total estimates of excess deaths caused by ambient temperature in four seasons by the corresponding total deaths.3)DLNM models were used to examine the city-specific DTR/TV effect on CVD death.The combined effects of DTR/TV on CVD death were then obtained using a meta-analysis,based on restricted maximum likelihood.4)15 types of heat wave and cold spell definitions through combining five heat thresholds(90.0th,92.5th,95.0th,97.5th and 99.0th percentiles of daily maximum temperature)and five cold thresholds(2.5th,7.5th,5.0th,7.5th and 10.0th percentiles of daily minimum temperature)definitions,and both three duration days(?2,?3 and?4)were compared,respectively.The pooled effects were then computed using the meta-analysis.5)We adopted DLNM model to fit the city-specific relationship between temperature and mortality.We then estimated the future temperature-related deaths using the downscaled temperature projections from the 16 general circulation models(GCMs)and four Representative Concentration Pathways(RCPs).Results 1)In the past six decades,the annual mean,maximum and minimum temperature rose significantly at the rate of 0.0401?/year,0.019?/year and 0.050?/year,while the DTR decreased significant at rate of 0.022?/year;the changing patterns were the fastest during winter but the slowest during summer;the magnitudes were positively associated with latitudes;2)Significant and non-linear associations between temperature and mortality were observed.The MMT centered at 78th of temperature at the national level.In total,17.1%(95%empirical CI:14.4-19.1%)of CVD mortality(330,352 deaths)was attributable to ambient temperature,with substantial differences among cities,from 10.1%in Shanghai to 23.7%in Guangzhou.Most of the attributable deaths were due to cold,with a fraction of 15.8%(13.1-17.9%),compared to 1.3%(1.0-1.6%)for heat.Furthermore,180,895(95%empirical CI:74,383-262,499)CVD related to non-optimum temperature occurred in winter,compared with 34,622(9,315-57,653)in summer.Temperature-related AFs were 16.10%(2.34-27.52%)in spring,10.68%(-0.57-20.07%)in summer,8.17%(2.20-13.61%)in autumn and 31.42%(12.92-45.59%)in winter.3)The DTR impacts were generally limited to four days,with harvesting effects followed during lag 4-8 days.An increase of 1? in DTR during hot and cold temperature was associated with pooled estimate of-0.76%(-1.15--0.36%)and 0.79%(0.35-1.23%)increase in CVD deaths at lag 0-10 days,respectively.Estimates were higher among females,the elderly and those with lower education level,but the statistical difference was only observed among age group(P=0.079)and education level(P=0.006).For the TV,the effect estimates generally increased by exposure days.An increase of 1 ? in TV at 0-7 days was associated with pooled estimate of 0.97?(0.55-1.39%)increases in CVD death.The impacts of TV at 0-7 days was significantly correlated with latitudes(rs=-0.586,P=0.022).Effects of TV among individual characteristics shared similar trends of those from DTR-CVD association.4)Heat wave definition using daily maximum temperature>92.5th percentile during May-September with duration? 3 days produced the best model fit,and cold spell using daily minimum temperature?7.5th percentile during November-March with duration?4 days.Impact of heat wave was immediately,but harvesting effects were followed after lag 4 day;while impact of cold spell could last for 10 days.The national pooled relative risk was 1.07(1.02,1.12)at lag 0-2 days for heat wave and 1.11(1.07,1.15)at lag 0-10 days for cold spell.Health effects of heat waves were positively correlated with latitudes,while cold spells were negatively related to latitudes.The elderly,females and those with lower education level suffered more from extreme temperature events.The statistical difference among gender and age group were significant for heat wave(P=0.083,P=0.053),and only significant difference among age group for cold spell(P=0.010).5)Annual heat-related excess CVD deaths varied greatly across different GCM models in all cities.On average,the annual heat-related excess CVD mortality will increase by 6.0%,7.3%,4.9%and 9.5%in 2030s,and 8.0%,16.3%,17.9%and 30.3%in 2080s under RCP2.6,RCP4.5,RCP6.0 and RCP8.5,respectively.Compared with moderate hot temperature,higher increasing rates were observed among future heat wave,with average magnitude of 88.6%in 2080s under RCP8.5.The combined effects of future heat and cold will increase in Changchun,Shenyang and Jinan,while will decrease in the rest cities.Conclusions 1)Climate change varied by different regions and climates,the policy on reducing impacts of climate change should be based on local climate and its long-term trend.2)This study emphasizes how cold weather is responsible for most part of the temperature-related CVD death burden.Temperature-related CVD death was significantly higher during winter season.Our results may have important implications for the development of policies to reduce CVD mortality from extreme temperatures.3)DTR and TV have considerable effects on CVD death risk in various cities in China.The elderly and those with low education level were the vulnerable subpopulations to the short-term variation of ambient temperature.Public health interventions should be targeted to minimize public health burden due to the unstable weather patterns.4)The findings may have important implications for the development of heat alert systems and protecting vulnerable subpopulations from extreme temperature events in China.5)This study confirmed that future temperature warming will lead to increased burdens of heat-related mortality in China,particularly for the future heat wave.The magnitudes of the future temperature-related burden varied greatly by regions.Some preventive and intervention strategies should be efficiently developed in response to climate change,... |
PDF Full Text Request