Schwartz Joel, Samet Jonathan M, Patz Jonathan A
Department of Environmental Health, Harvard School of Public Health, Boston, Massachusetts 00215, USA.
Epidemiology. 2004 Nov;15(6):755-61. doi: 10.1097/01.ede.0000134875.15919.0f.
We estimated the effects of temperature and humidity on hospital admissions for heart disease (International Classification of Diseases, 9th revision [ICD-9] codes 390-429) and myocardial infarction (ICD-9 code 410) of persons age 65 and older in 12 U.S. cities with a wide range of climates. To account for possible delayed effects and harvesting, we examined the impact of weather up to 20 days before each admission.
Poisson regression models were fitted in each city, with regression splines used to control for season and barometric pressure. We also controlled day of the week. We estimated the effect and the lag structure of both temperature and humidity based on a distributed lag model.
For cities in both hot and cold climates, we found that hospital admissions for all heart disease increased monotonically with average temperature on the same day as and the day before admission. The effect of very high temperatures had a temporal pattern consistent with harvesting: several days after an episode of high temperature, there were fewer admissions. In contrast, a protective effect of cold temperature persisted without rebound. The effects of either hot or cold temperature disappeared within 10 days of exposure. There was no evidence for a humidity effect. Similar but smaller effects of temperature were seen for admissions for myocardial infarction specifically.
The effects of temperature on hospital admissions predominantly occur within a few days after exposure, and much of the effect of hot temperatures is short-term displacement of events.
我们评估了温度和湿度对美国12个气候差异较大城市中65岁及以上人群因心脏病（国际疾病分类第九版[ICD - 9]编码390 - 429）和心肌梗死（ICD - 9编码410）住院情况的影响。为了考虑可能的延迟效应和收获效应，我们研究了每次入院前长达20天的天气影响。
在每个城市拟合泊松回归模型，使用回归样条来控制季节和气压。我们还控制了星期几。基于分布滞后模型，我们估计了温度和湿度的效应及滞后结构。
对于炎热和寒冷气候的城市，我们发现所有心脏病的住院人数在入院当天及前一天随平均温度单调增加。极高温度的影响呈现出与收获效应一致的时间模式：高温事件发生几天后，入院人数减少。相比之下，低温的保护作用持续存在且无反弹。高温或低温的影响在暴露后10天内消失。没有证据表明湿度有影响。心肌梗死入院情况的温度效应与之相似但较小。
温度对住院人数的影响主要发生在暴露后的几天内，高温的大部分影响是事件的短期转移。
