应用反向计算法估计2010-2023年云南省德宏傣族景颇族自治州HIV新发感染病例  

Application of the back-calculation method for estimating new HIV infections in Dehong Dai and Jingpo Autonomous Prefecture,Yunnan Province,2010-2023

作　　者：肖民扬 杨跃诚 贾曼红[1] 汤后林[3] 施玉华[1] 付丽茹[1] 张祖样 唐仁海 王晓雯[1] 陈方方[3] Xiao Minyang;Yang Yuecheng;Jia Manhong;Tang Houlin;Shi Yuhua;Fu Liru;Zhang Zuyang;Tang Renhai;Wang Xiaowen;Chen Fangfang(Yunnan Provincial Center for Disease Control and Prevention,Kunming 650500,China;Dehong Prefecture Center for Disease Control and Prevention,Mangshi 678400,China;National Center for AIDS/STD Control and Prevention,Chinese Center for Disease Control and Prevention,Beijing 102206,China)

机构地区：[1]云南省疾病预防控制中心,昆明650500 [2]德宏傣族景颇族自治州疾病预防控制中心,芒市678400 [3]中国疾病预防控制中心性病艾滋病预防控制中心,北京102206

出　　处：《中华流行病学杂志》2025年第4期669-675,共7页Chinese Journal of Epidemiology

基　　金：国家科技重大专项(2018ZX10715006);云南省兴滇英才支持计划。

摘　　要：目的HIV新发感染是评估艾滋病疫情动态变化的核心指标。本研究旨在通过反向计算法,结合诊断延迟分布和贝叶斯理论,估算云南省德宏傣族景颇族自治州(德宏州)HIV新发感染者人数,并比较两种方法估计结果的差异。方法资料来源于中国疾病预防控制信息系统。基于CD4^(+)T淋巴细胞(CD4)计数消除模型,利用德宏州2010-2023年诊断的HIV感染者的抗病毒治疗前首次CD4计数数据,反向计算HIV感染者的感染日期,并确定各年度的HIV新发感染者中已确诊人数。分别运用诊断延迟分布和贝叶斯理论估计HIV新发感染者的确诊概率,以推算2010-2023年德宏州的HIV新发感染者病例数。结果2010-2023年德宏州共报告非母婴传播途径感染的≥15岁HIV感染者5693例。排除无CD4计数结果和首次CD4计数异常值(≥2000个/μl)后,最终纳入5329例。通过CD4计数反向计算,在2010-2023年历年HIV新发感染者中,已确诊人数分别为479、427、337、305、256、219、194、193、131、166、120、71、42和47例。基于诊断延迟分布,使用寿命表分析显示,累计确诊概率由1年内的0.301逐年上升至14年内的0.913,由此估计新发感染人数由2010年的577例下降至2023年的168例,累计HIV新发感染人数为4412(95%CI:4350~4480)例。而基于贝叶斯理论,累计确诊概率由1年内的0.413上升至14年内的0.946,新发感染估计数则由2010年的557例降至2023年的122例,累计新发感染人数为3814(95%CI:3787~3837)例。结论应用两种方法估算的2010-2023年德宏州HIV新发感染者人数基本一致,考虑当地持续开展艾滋病扩大检测,贝叶斯理论的估计结果可能更接近实际情况。本研究为德宏州的艾滋病防控策略制定和优化提供参考依据,有助于推动该地区向2030年消除艾滋病目标迈进。Objective New HIV infections serve as a crucial indicator for assessing the dynamic changes in the HIV epidemic.This study aims to estimate the number of new HIV infections in Dehong Dai and Jingpo Autonomous Prefecture of Yunnan Province(Dehong),using a back-calculation method that integrates diagnosis delay approaches and Bayesian theory.Additionally,it compares the differences between these two estimation methods.Methods Data were obtained from the Chinese Information System for Disease Control and Prevention.Based on CD4^(+)T lymphocytes(CD4)counts depletion model,the first CD4 count prior to antiretroviral therapy of HIV-infected individuals diagnosed in Dehong from 2010 to 2023 was utilized to retroactively determine the infection date of HIV-infected individuals and ascertain the annual number of new HIV infections who had been diagnosed.Subsequently,the diagnosis delay distribution method and Bayesian theory were leveraged to assess the diagnosis probability of newly infected individuals,thereby projecting the number of new HIV infections in the region over the specified period.Results During 2010-2023,a total of 5693 individuals aged 15 and above,excluding mother-to-child transmission,were diagnosed with HIV in Dehong.After excluding 364 cases due to missing CD4 count results or abnormal first CD4 counts(≥2000 cells/μl),5329 HIV-infected individuals were included in the final analysis.Through CD4 counts back-calculation from 2010 to 2023,the annual number of new infections diagnosed was 479,427,337,305,256,219,194,193,131,166,120,71,42 and 47.When using the diagnosis delay distribution method and life table analysis,the cumulative diagnosis probability rose from 0.301 within one year to 0.913 within 14 years,leading to a reduction in the number of estimated new infections from 577 in 2010 to 168 in 2023,with a total estimate of 4412(95%CI:4350-4480).Alternatively,based on Bayesian theory,the diagnosis probability increased from 0.413 within one year to 0.946 within 14 years,leading to a reduction in t

关 键 词：艾滋病病毒 新发感染 CD4^(+)T淋巴细胞 反向计算法 

分 类 号：R512.91[医药卫生—内科学]