基于分布滞后非线性模型评估气温对北京市居民缺血性脑卒中入院人数的影响
Effects of temperature on ischemic stroke admissions in Beijing residents based on distributed lag non-linear model
-
摘要:
背景 缺血性脑卒中是我国面临的重大公共卫生问题之一。近年来气温与人群缺血性脑卒中的关系受到广泛关注，但目前的研究结果尚存在差异及不一致性。
目的 评估北京市日均气温对居民缺血性脑卒中入院的影响，为相关部门制定脑卒中综合防治策略和措施提供科学依据。
方法 收集2013年1月11日-2017年12月31日北京市的气象数据、空气污染物数据以及缺血性脑卒中入院数据，描述气象因素、空气污染物与居民缺血性脑卒中入院人数的基本情况。采用Spearman秩相关分析空气污染物及气象因素的相关性，采用分布滞后非线性模型（DLNM）分析日均气温与缺血性脑卒中入院的关系，绘制三维关联图与不同温度滞后效应分布图，并按照性别、年龄（< 65岁和≥ 65岁）进行分层分析。
结果 2013-2017年北京市居民缺血性脑卒中入院人数共380 127人，每日入院人数中位数（
P 25，P 75）为209（138~261）。北京市日均气温与缺血性脑卒中入院人数之间呈非线性关系，并存在一定的滞后效应。低温对缺血性脑卒中入院人数的效应在第10天开始出现，约第13天达最高值后开始下降。当日均气温为-3℃，累积滞后21 d时，RR 值最大，为1.18（95%CI ：1.05~1.32）。当日均气温为1℃时，在累积滞后14 d时，缺血性脑卒中入院人数的RR 值为1.08（95%CI ：1.02~1.15），在累积滞后21 d时，缺血性脑卒中入院人数的RR 值为1.14（95%CI ：1.06~1.24）。分层分析结果显示低温对男性、小于65岁人群缺血性脑卒中入院人数的影响更敏感。低温累积滞后21 d时，对于男性人群，缺血性脑卒中入院人数的RR 值为1.20（95%CI ：1.05~1.36）；对于年龄小于65岁人群，缺血性脑卒中入院人数的RR 值为1.29（95%CI ：1.12~1.50）。此外，尚未发现高温对缺血性脑卒中入院人数的影响。结论 低温会增加北京市居民缺血性脑卒中的入院风险，且存在累积滞后效应。居民应注重低温对健康的影响，在低温天气中加强对缺血性脑卒中的防范。
Abstract:Background Ischemic stroke is one of the major public health issues in China. In recent years, the relationship between temperature and ischemic stroke in general population has been widely concerned. However, the current research results are inconsistent.
Objective This study is conducted to explore the influence of daily temperature on ischemic stroke admissions of residents in Beijing and provide scientific basis for relevant departments to establish comprehensive prevention and treatment strategies against ischemic stroke.
Methods Meteorological data, air pollution data, and ischemic stroke admission data from 11 January 2013 to 31 December 2017 in Beijing were collected and described. The correlation between air pollutants and meteorological factors were analyzed by Spearman rank correlation analysis. The association between daily temperature and hospital admissions for ischemic stroke was estimated using distributed lag nonlinear model (DLNM). Three-dimensional graph for ischemic stroke and lag-response curves for different temperatures were plotted. The association stratified by sex or age (< 65 and ≥ 65 years) was also examined.
Results In total, there were 380 127 ischemic stroke admissions in Beijing from 2013 to 2017, the median number of daily admissions was 209 (interquartile range:138-261). A nonlinear relationship was found between daily mean temperature and ischemic stroke admissions, and the effect was lagged. The effect of temperature on admissions of ischemic stroke appeared since day 10, and declined after reaching the peak since day 13. The largest
RR at -3℃ was 1.18 (95%CI :1.05-1.32) at lag 0-21 days. TheRRs at 1℃ were 1.08 (95%CI :1.02-1.15) at lag 0-14 days and 1.14 (95%CI :1.06-1.24) at lag 0-21 days, respectively. The results of stratified analysis presented that men and visitors less than 65 years of age were more sensitive to ischemic stroke at low temperature. At lag 0-21 days, theRRs at -3℃ were 1.20 (95%CI :1.05-1.36) for men and 1.20 (95%CI :1.12-1.50) for visitors less than 65 years of age. In addition, the association between hot temperature and ischemic stroke admission was not found.Conclusion Low temperature could increase the risk of ischemic stroke admissions of residents in Bieijing and there is a cumulative lag effect. The study findings suggest that residents pay attention to the health impact of low temperature and strengthen the prevention against ischemic stroke in cold weather.
下载:
