成都市不同温度段对呼吸和心脑血管疾病死亡影响的归因风险评估

Short-term Risk Assessment on Respiratory and Cardiovascular Mortality due to Different Temperature Ranges of Chengdu

目的评估成都市气温暴露造成的人群死亡归因风险。方法本研究收集成都市2014-2016年人群呼吸和心脑血管疾病逐日死亡资料,以及同期的气象资料和大气污染物资料,利用时间序列分析方法和分布滞后非线性模型,在控制污染物等混杂因素的基础上,探明气温对当地呼吸和心脑血管疾病死亡的影响,并定量估算了由不同温度段造成的归因死亡人数和人群归因分值(AFs)。结果成都市气温对呼吸和心脑血管疾病死亡的累积暴露-反应关系近似呈"L"型,22.2℃为最适温度(MMT),其对应死亡风险最小;经计算,由气温直接造成的呼吸和心脑血管疾病年均超额死亡人数分别为3121例和4043例,对应AFs分别为15.69%和15.81%,其中归因于冷效应(<MMT)的AFs分别为11.52%和13.76%,归因于热效应(≥MMT)的AFs分别为4.17%和2.05%;在此基础上,以2.5th和97.5th温度百分位进一步将气温分成强低温、弱低温、弱高温和强高温四段,发现由弱低温造成呼吸和心脑血管疾病死亡的AFs为10.07%和12.16%。结论气温对人群健康的影响以冷效应为主,而冷效应中以弱低温效应最为凸显。虽然弱低温健康效应的相对危险度(RR)小于强低温,但由于弱低温发生频次高,进而导致其造成的死亡人数最多,在今后的研究中应该加强弱低温不利健康效应的研究。

Objective To evaluated the attribute fractions of mortality caused by temperature in Chengdu.Methods The daily data on respiratory and cardiovascular death numbers,meteorological factors,and air pollutants ofChengdu were collected during 2014-2016.A time series study with a distributed lag non-linear model(DLNM)was adopted to estimate the association between ambient temperature and respiratory mortality,as well as temperature and cardiovascular mortality,when the confounding factors such as pollutantswerecontrolled.Then the attribute numbers(ANs)and attribute fractions(AFs)of respiratory and cardiovascular mortality that were attributable to different temperature sections were calculated,respectively.Results The overall cumulative exposure-response curves between temperature and respiratory and cardiovascular mortality both were L-shaped at lag 21 days.And 22.2℃was the optimum temperature,with the minimum-mortality temperature(MMT)in Chengdu.The ANs of respiratory and cardiovascular mortality that caused by temperature were 3121and 4043 cases per year,corresponding tol5.69%and 15.81%of AFs.Cold effects(<MMT)were responsible for 11.52%and 13.76%of respiratory and cardiovascular mortality;while heat effects(mMMT)were responsible for4.17%and 2.05%of respiratory and cardiovascular mortality,respectively.On the basis,Temperature was further refine to four sections,including strong low-temperature,moderate cold,moderate heat and strong high-temperature,using three temperature cutoffs(the 2.5th temperature percentiles,MMT,and the 97.5 th temperature percentiles).We found that moderate cold account for 10.07%and 12.16%of respiratory and cardiovascular mortality that were caused by temperature.Conclusion Cold effects afford most of temperature-related mortality,while moderate coldwas responsible for most of mortalitythat was caused by cold effects.The findingexposesthat moderate cold playan important role in temperature-relatedrespiratory and cardiovascular mortality.Although moderate cold had a lower RR than extreme cold,it occurred on more days than did extreme cold.Therefore,the health burden caused by moderately cold temperature requires more attention in the future.