我国HIV传播耐药监测中一类重要问题的处理（英文）

Tackling the Bottleneck in the Surveillance of Transmitted HIV Drug Resistance in China

对于HIV传播耐药,WHO给出了截断序贯抽样方法来判别耐药水平,要求的最大样本量是47.但是,中国当前HIV流行水平较低,最大样本量很多时候无法达到,导致不能给出耐药水平的准确判断.为了解决这个问题,本文借助贝叶斯估计,给出了一种方法加于改进.借助于该方法,最大样本147被减小到38.做为该方法的应用,对于WHO抽样方法无发判断的新疆和湖南地区,我们分别给出了低水平耐药和中水平讨药的判定.同时,对于WHO抽样方法能够判断的地区,我们的方法也给出了相同的结论.本文的方法可以应用到样本量不够充足的地区,是WHO抽样方法的改进.

For transmitted HIV drug resistance,the WHO gives a truncated sequential sampling method to classify the prevalence,which asks for a maximum sample size 47.However,in China,where HIV prevalence is relatively low,sometimes it is impossible to reach this maximum sample size,in this case perhaps a clear prevalence result could not be obtained.In order to solve this problem,with the help of Bayes estimate we provide a supplemental method to classify the prevalence of transmitted HIV drug resistance for a target area in China.By our method,the maximum sample size required by the WHO’s sampling method may be reduced to 38 from 47,and as its applications,two provinces in China where the WHO’s method can not give clear classification results may be classified as low prevalence for Xinjiang and moderate prevalence for Hunan.Moreover,for some areas in China where the WHO’s method can give clear classification results,the same results can also be obtained by our method.As a summary,our method is suitable for the surveillance of transmitted HIV drug resistance and could be applied to other countries and regions,especially for those without enough specimens.It is an improvement of the WHO’s sampling method.