Objective: To investigate the effect of humidex combined with mean temperature and relative humidity on the incidence of bacillary dysentery in Hefei. Methods: Daily counts of bacillary dysentery cases and weather data in Hefei were collected from January 1, 2006 to December 31, 2013. Then, the humidex was calculated from temperature and relative humidity. A Poisson generalized linear regression combined with distributed lag non-linear model was applied to analyze the relationship between humidex and the incidence of bacillary dysentery, after adjusting for long-term and seasonal trends, day of week and other weather confounders. Stratified analyses by gender, age and address were also conducted. Results: The risk of bacillary dysentery increased with the rise of humidex. The adverse effect of high humidex (90 percentile of humidex) appeared in 2-days lag and it was the largest at 4-days lag (RR=1.063, 95%CI: 1.037-1.090). Subgroup analyses indicated that all groups were affected by high humidex at lag 2-5 days. Conclusion: High humidex could significantly increase the risk of bacillary dysentery, and the lagged effects were observed.
目的： 探讨温度和湿度的共同作用指标——温湿指数对细菌性痢疾发病的影响。 方法： 收集2006年1月1日至2013年12月31日合肥市细菌性痢疾每日发病资料及同期的气象资料，并计算出温湿指数，采用分布滞后非线性模型分析温湿指数与细菌性痢疾发病的关系，调整季节性和长期趋势、星期几效应和其他气象因素的影响。同时，按性别、年龄和居住地进行亚组分析。 结果： 细菌性痢疾的发病风险随着温湿指数增大而增加。高温湿指数（P(90)）对细菌性痢疾发病影响的危害效应于滞后2 d产生，在滞后4 d达到最大（RR=1.063，95%CI：1.037~1.090）。亚组分析发现在滞后2~5 d，高温湿指数对各组人群均有影响。 结论： 高温湿指数能增加细菌性痢疾发病的风险，并存在滞后效应。.
Keywords: Bacillary dysentery; Distributed lag nonlinear model; Humidex.