黑龙江省黑河市肾综合征出血热流行因素相关分析及回归模型的建立

Correlation analysis and regression model of epidemic factors of hemorrhagic fever with renal syndrome in Heihe city,Heilongjiang province

目的探讨黑龙江省黑河市气象因素、主要宿主鼠密度与肾综合征出血热(HFRS)发病率的相关性,建立回归模型。方法收集黑河市1984-2007年的HFRS病例数据、气象数据(最高气温、最低气温、平均气温、平均相对湿度和降雨量)、主要宿主鼠密度(黑线姬鼠和褐家鼠)数据,采用Spearman等级相关和多元逐步回归方法进行统计分析。结果分析结果表明:黑河市前6个月的平均气温、平均相对湿度和降雨量等因素变量与本月HFRS发病率具有显著的相关性,其中3个月前、4个月前和5个月前(即上季度)的气象因素与本月HFRS发病率的相关系数最高,呈正相关;主要宿主鼠密度与HFRS发病率呈显著的正相关,以1个月前、2个月前和3个月前为最高;按照决定系数最大的择优原则建立的回归模型为:y=-11.359+0.08x6+0.27n1+0.221n2+0.175x2(其中:y代表月发病率,x6代表4个月前降雨量,n1代表1个月前宿主鼠密度,n2代表2个月前主要宿主鼠密度,x2代表3个月前平均相对湿度),模型的决定系数(R2)为0.709,差异有统计学意义(F=34.703,P=0.000)。结论回归模型的拟合效果很好,表明可以在该地区综合运用气象和宿主鼠的数据进行HFRS流行趋势的预测分析。

Objective The study was conducted to investigate the correlation between meteorological factors and rat （ the main host） density and the incidence rate of hemorrhagic fever with renal syndrome （HFRS） in Heihe city, Heilongjiang province, and establish the regression model. Methods HFRS cases, meteorological data （maximum temperature, minimum temperature, mean temperature, mean relative humidity and rainfall ）, rat density （Apodemus agrarius and Rattus norvegicus） in Heihe city from 1984 to 2007 were collected. The statistical analysis was performed using Spearman rank correlation and multiple-stepwise regression. Results The analysis showed significant correlation between such variables as mean temperature, mean relative humidity and rainfall in the past six months and the incidence rate of HFRS in the current month in Heihe city, of which the meteorological factors three months ago, four months ago and five months ago （the last quarter） were shown to be of the highest positive correlation coefficient with the incidence rate of HFRS. The rat （main host） density was significant/y and positively correlated with the incidence rate of HFRS, and the highest correlation was observed in the data one months ago, two months ago and three months ago. According to the merit-based largest coefficient of determination, the regression model was established as follows： y = - 11. 359 ＋ 0.08x6 ＋ 0.27n1 ＋ 0.221n2 ＋ 0. 175x2 （where： y was the incidence rate of HFRS; x6 was the rainfall four months ago; nl was the host density one month ago; n2 was the host density two months ago; x2 was the mean relative humidity three months ago）. The coefficient of determination （Rz ） was 0. 709, the difference being statistically significant （F = 34. 703, P = 0. 000）. Conclusion The regression model was fitting favorably, indicating that it is applicable to perform predictive analysis of the epidemic trend of HFRS based on the combination of meteorological factors and host density data in the region.