血吸虫病患者数推算两种方法结果比较探讨

Comparison of estimated number of schistosomiasis cases by using two types of samples

目的得到较为准确的武汉市血吸虫病患者推算数。方法以"历史累计新发病人数"为容量变量,采取分层整群结合容量比例概率法(PPS)抽样,从武汉市265个血吸虫病4类村,抽取46个样本村,加上全部3类村(44个),总计90个血吸虫病患者数推算抽样村(简称市抽样村);湖北省按地区分层抽样,分配武汉市22个血吸虫病患者数推算抽样村(简称省抽样村),对以上抽样村开展人群查病,血检阳性者进行粪检,计算各区血吸虫病感染率和推算数,汇总得到武汉市血吸虫病患者推算总数,比较两种方法推算患者数的差异。结果市抽样村和省抽样村推算武汉市血吸虫病患者数分别为2 811人(其中3类村464人,4类村1 994人)和2 344人(其中3类村482人,4类村1 509人),两类样本推算3类村和4类村患者数,其构成比差异有统计学意义(χ2=18.68,P=0.000)。结论以"历史累计新发病数"指导PPS抽样,可以使抽样检出患者的效率提高,得到的血吸虫病患者数推算结果较为准确,其对血吸虫病低流行区人群查病工作也具有一定指导意义。

Objective To obtain more accurate estimated number of schistosomiasis eases in Wuhan. Methods Setting ＂historical cumulative number of new cases＂ as the volumetric variable, adopting stratified cluster sampling and probability proportional to size sampling （PPS）. Collecting 46 villages as samples from 265 ＂Type Four Villages＂ of schistosomiasis eases in Wuhan, and with all ＂Type Three Villages＂ （ total 44）, the total number of villages of schistosomiasis for sampling was 90 （referred as ＂City Sample Village＂ ）. For Hubei Province, adopting stratified cluster sampling, allocating in Wuhan with 22 sample villages of schistosomiasis （referred as ＂Provincial Sample Village＂）. For the above villagers ,feces exami- nation was performed for those who were positive in blood tests;schistosomiasis infection rate and number in each district were calculated,and the reasoning number of schistosomiasis eases in Wuhan was summed up, and the results of the two types of samples were compared. Results There were 2 811 schistosomiasis cases in total in ＂ City Sample Village＂ （ 464 in ＂Type Three Villages＂, and 1 994 in ＂Type Four Villages＂ ）, and there were 2 344 schistosomiasis cases in total in ＂Provincial Sample Vinage＂ （482 in ＂Type Three Villages＂ ,and 1 509 in ＂Type Four Villages＂ ）. The constituent ratio of schistosomiasis eases in ＂Type Three Villages＂ and ＂Type Four Villages＂ of the sampling results had statistical significance （X2 = 18.68,P = 0. 000）. Conclusions Based on ＂accumulated eases of patients＂ , using probability proportional method to size sampling could be more efficient, and the estimated number of schistosomiasis eases could be more accurate. It had certain degree of guiding significance in the survey of schistosomiasis in low endemic area.