新疆维吾尔族妇女人乳头瘤病毒感染自然史的转移概率研究  

Study on the transition probability of natural history of human papilloma virus infection in Uyghur women in Xinjiang

作　　者：闫小龙[1] 张眉[1] 王奎 张梅[1] 李小菊[1] 胡欣 李珂[1] 井明霞[1] YAN Xiao-long;ZHANG Mei;WANG Kui;ZHANG Mei;LI Xiao-ju;HU Xin;LI Ke;JING Ming-xia(Department of preventive medicine,Medical College of Shihezi University,Shihezi,Xinjiang 832002,China)

机构地区：[1]石河子大学医学院预防医学系,新疆石河子832002

出　　处：《现代预防医学》2021年第4期604-608,共5页Modern Preventive Medicine

基　　金：新疆生产建设兵团博士基金(2013BB015)。

摘　　要：目的探讨新疆维吾尔族妇女人乳头瘤病毒(human papillomavirus,HPV)感染病变进展、消退或持续感染的转移概率,为使用基于转移概率为基本参数的Markov模型进行卫生经济学评价提供参数。方法通过回顾性调查方法,收集2016-2018年3年期间内在新疆喀什地区泽普县妇幼保健院进行随访的维吾尔族妇女HPV感染自然史资料,以随访累计超过1年且资料完整的对象入选,共计调查895名研究对象。使用期望最大化算法(expectation maximization algorithm,EM)对HPV持续感染、发生进展及消退的年转移概率进行了估计。结果样本人群中HPV年感染率为4.71%,从健康状态转移至宫颈高级别病变的年进展率为0.10%。HPV感染患者的持续感染、发生进展及消退的年转移概率分别为43.53%、4.60%和51.87%。结论维吾尔族妇女HPV持续感染及进展到高级别病变的发生率较高,建议针对此类人群进行密切随访。根据数据特点,本研究应用EM法估算的维吾尔族妇女HPV感染自然史的转移概率,可为该地区卫生经济学评价中使用的Markov模型需要的参数提供参考依据。Objective To investigate the transition probability of progression, regression and persistence of HPV infection in Uyghur women in Xinjiang, in order to provide parameters for health economics evaluation using Markov model. Methods Through retrospective investigation, we collected natural history data of HPV infection of Uyghur women who were followed up in Zepu county Maternal and Child Health Care Hospital in Kashgar, Xinjiang during 2016 and 2018. A total of 895 subjects with complete data and more than 1 year of follow-up were included. The Expectation Maximization Algorithm(EM)method was used to estimate the annual transition probability of HPV persistence, progression and regression. Results The annual HPV infection rate was 4.71%, and the annual progression rate from healthy state to high-grade cervical lesions was0.10% in surveyed subjects. The annual transition probability of persistence, progression and regression in HPV infected population was 43.53%, 4.60% and 51.87%, respectively. Conclusion Uyghur women have a higher incidence of persistent HPV infection and progression to high-grade lesions, suggesting close follow-up for HPV positive individuals in the region.According to the characteristics of the data, the EM method is used to estimate the transition probability of HPV infection in Uyghur women, which can provide reference basis for the parameters required by the Markov model used in health economics evaluation.

关 键 词：人乳头瘤病毒 自然史 MARKOV模型 期望最大化算法 

分 类 号：R737.33[医药卫生—肿瘤]