【作者】 张晶；

【导师】 汪宁；

【作者基本信息】 中国医科大学 ， 流行病与卫生统计学， 2018， 博士

【摘要】 目的:自1980年在男男同性性行为者(men who have sex with men,MSM)中发现首例艾滋病毒(human immunodeficiency virus,HIV)感染者至今已近40年。截至2016年底,全球共有3670万存活HIV感染者(people living with HIV,PLHIV),2016年当年即有180万人新发生了HIV感染。在此背景下,世界卫生组织(world health organization,WHO)及联合国艾滋病规划署(united nations the joint program on AIDS,UNAIDS)于2013年12月提出2020年完成三个90%目标并于2030年终结全球HIV/AIDS疫情的目标。三个90%目标包括:截至2020年,90%的PLHIV知晓自己的感染状态、90%的确诊HIV感染者接受并持续进行抗逆转录病毒治疗(antiretroviral treatment,ART)治疗、90%接受了ART治疗者达到HIV病毒抑制。其中第一个90%目标是该目标的前提和瓶颈,需依靠HIV高危人群高比例参与HIV检测来实现。MSM人群已成为我国近年HIV疫情增长最快的HIV高危人群,而该人群参与HIV检测的比例不足50%,距离三个90%目标差距极大。与此同时,自愿咨询检测(voluntary counseling and testing,VCT)作为我国向HIV高危人群提供免费HIV检测的最主要途径,在应对我国近年MSM人群中快速增长和转变的疫情上遇到了挑战:首先,VCT依靠高危MSM主动求询的“守株待兔”的方式,对当下互联网时代我国MSM全面转变的高危行为特征和HIV疫情特点缺乏应对;其次,传统VCT限于技术及资源限制,无法“主动出击”向高危MSM提供预防干预服务;最后,VCT由于匿名原则,无法从既往进行的百万量级的免费HIV检测中识别出重复检测,从而无法获得HIV新发感染率这一HIV疫情评估的金指标,更无法实现持续干预。而随着互联网、智能手机及在线社交软件在我国的快速普及,电子科技技术在生物医学领域的广泛应用,为传统VCT拓展其疫情监测及预防干预的能力提供了契机。方法:本研究第一部分通过利用并整合基于微信的互联网技术构建电子队列。该电子队列策略包括:在MSM人群线上及线下热点活动场所投放招募二维码以获得代表性样本、采用在线自助调查技术获得研究样本的流行病学敏感信息,采用线下VCT实验室检测辅以线上HIV/梅毒自检试剂申领邮寄方式获得研究样本的HIV及梅毒感染率。从而掌握电子队列在目标人群中的接受率,线上及线下招募来源的高危MSM的社会学背景特征、高危行为等特征及梅毒、HIV感染率的差异,以及多来源代表性样本下的本地区高危MSM人群的HIV感染率及影响因素。本研究第二部分对基于VCT门诊求询MSM人群的电子队列进行前瞻性随访。通过身份二维码识别技术,在保障VCT保密原则的前提下,实现多现场、实时、匿名识别求询MSM身份,通过自助流行病学调查技术避免社会期望偏倚准确地获得求询者的社会背景学特征及HIV相关高危行为特征等敏感信息,通过梅毒及HIV实验室检测获得求询者的梅毒抗体阳性率、现症梅毒感染率及HIV新发感染率。最终获得该种队列维持技术下,本地区高危MSM人群及MSM亚人群的HIV新发感染率、队列保持率及影响因素。本研究第三部分对在线VCT功能对求询MSM的HIV预防干预效果的真实世界研究。通过设计并建立在线VCT的干预功能菜单,包括在线咨询、预防信息推送、预防干预电子游戏、检测结果在线查询及HIV感染风险自评估工具等;通过后台捕捉求询MSM对于干预菜单调用的网络行为轨迹客观评估在线VCT干预功能的使用情况及求询MSM的在线求询行为的特征。通过真实世界研究法,比较在线干预功能使用者及未使用者基线及末次随访的高危行为变化情况及在线干预功能的使用对于梅毒及HIV新发感染率的影响。结果:1.研究期间(2017年3月至2018年2月)累计招募2371名求询者,其中来自线下招募来源689人(29.0%),线上招募来源1682人(71.0%)。经筛选获得1128名符合纳入标准的MSM,其中993名MSM接受并同意参与电子队列,电子队列接受率为88.0%。相比于不接受者,接受者中更高比例地存在年龄≤24岁、在校大学生及有无保护肛交行为(condomless anal intercourse,CAI)的MSM。电子队列的基线HIV感染率为5.7%(57/993)、梅毒抗体阳性率为19.3%(191/993)。线上与线下招募来源MSM在HIV感染率方面无统计学差异(5.5%VS.5.9%,P=0.352)。流动人口(VS.一直在沈阳居住,Adjusted OR(aOR)=3.7,95%CI:1.8-7.8)、既往HIV检测频率每年≦1次(VS.每年≥2次,aOR=4.4,95%CI:1.5-12.8)、自评HIV感染风险为高风险(VS.无风险,aOR=4.4,95%CI:1.6-11.7),过去3个月中与男性固定性伴安全套使用频率为偶然使用(VS.一直使用,aOR=5.2,95%CI:1.8-15.0)及从不使用(VS.一直使用,aOR=6.5,95%CI:2.1-20.2)与基线HIV阳性显著相关(P<0.05)。2.对基于VCT求询MSM人群电子队列的731名高危MSM进行前瞻性随访,共观察202.7人年,观察到HIV新发感染病例18例,从而获得VCT求询者的电子队列的HIV新发感染率为8.9/100人年(100-person-year,100PY)。其中,年轻MSM(年龄≦24周岁)人群中的HIV新发感染为5.4/100PY,中老年MSM(年龄≧40周岁)人群中HIV新发感染率为13.1/100 PY,在沈阳居住≦2年的流动MSM人口的HIV新发感染率为22.0/100 PY,在校大学生MSM的HIV新发感染率为3.8/100 PY。流动人口(VS.一直在沈阳居住,aHR=0.3,95%CI:0.1-0.4),使用公园/酒吧/浴池等非互联网寻伴途径(VS.互联网等线上途径,aHR=3.8,95%CI:1.3-10.8),过去3个月与同性性伴安全套使用频率为偶尔使用(VS.一直使用,aHR=6.2,95%CI:1.7-22.9),梅毒抗体阳性(VS.梅毒抗体阴性,aHR=2.5,95%CI:0.9-6.6),随访中发生了现症梅毒感染(VS.随访中未发生现症梅毒感染,aHR=4.6,95%CI:1.7-12.5)与HIV新发感染显著相关(P<0.05)。研究期间共440名MSM参加过至少一次随访检测,自然队列保持率为60.2%(440/731),人均观察时间中位数为177天。招募来源为线上招募(VS.线下招募,aOR=2.4,95%CI:1.5-3.9),流动人口中在沈阳居住≥3年(VS.一直在沈阳居住,aOR=1.8,95%CI:1.2-2.8),既往HIV检测频率每年≧2次(VS.首次检测,aOR=4.5,95%CI:2.7-7.4),过去3个月和同性存在CAI(VS.无CAI,aOR=2.5,95%CI:1.8-3.5),过去3个月中男性固定性伴人数≧2人(VS.无此类性伴,aOR=1.8,95%CI:1.1-2.8),过去3个月使用过助性药物(VS.未使用,aOR=1.8,95%CI:1.3-2.7),过去3个月存在酒后性行为(VS.无此类行为,aOR=1.7,95%CI:1.3-2.4),过去一年确诊了性传播性疾病(VS.无确诊性传播疾病,aOR=2.7,95%CI:0.9-8.2)与电子队列保持率显著相关(P<0.05)。3.研究期间,线上VCT干预功能菜单共发生了28818人次点击及功能调用。累计提供在线HIV人工及自助咨询解答2311人次。传统VCT非工作时间内(00:00-08:00,11:01-14:00,17:01-24:00)累计发生了14369人次干预功能调用,占总体功能调用人次数的49.9%(14369/28818)。电子队列随访的440名MSM中,283人使用过至少一种在线VCT干预功能,通过比较使用者基线与末次随访中的HIV相关高危性行为发生率发现,其与同性性伴近三个月CAI发生率由基线的56.2%下降为末次随访的39.9%,近三个月CAI发生率减少了16.3%(P<0.0001)。而在未使用者中未观察到具有统计学差异的高危行为减少。使用过在线VCT的干预类功能(HR=0.2,95%CI:0.1-1.1,P=0.065)及维持类功能(HR=0.8,95%CI:0.7-1.0,P=0.074)对梅毒新发感染率的降低存在边缘性相关。结论:1.电子队列策略可高效招募到大样本、多来源、具有代表性的高危MSM人群样本,在未来针对MSM人群的流行病学研究中具有应用前景。在此代表性样本基础上全面获得了本地区高危MSM人群的HIV及梅毒感染率。HIV检测频率不足及不清楚自身HIV感染风险显著增加了其HIV感染风险。2.基于VCT求询MSM电子队列在该人群及其亚人群中发现了较高的HIV新发感染率及队列保持率。该策略在维持及随访流动MSM人口方面存在低投入、高保持率的优势,亟待在本省内VCT进行推广应用。3.在线VCT干预功能在高危MSM人群中具有较高使用频次并拓展了传统VCT提供干预服务的时间及能力。在线干预功能的使用显著降低了求询MSM人群的CAI的发生率及梅毒发病密度,亟待推广应用。更多还原

【Abstract】 Objective: It has been nearly 40 years since the first human immunodeficiency virus(HIV)infection case has been found among men who have sex with men(MSM)in 1980.By the end of 2016,there were 36.7 million people living HIV(PLHIV)globally,There was 1.8 million new HIV infection occurred in 2016.In this context,the World Health Organization(WHO)and United Nations The Joint Programme on AIDS(UNAIDS)have announced the 90-90-90 target by 2020 and the goal to end the HIV/AIDS epidemic in 2030 globally in December 2013.The 90-90-90 target includes 90% of PLHIV be aware of their infection status,90% of those diagnosed with HIV received and sustained antiretroviral treatment(ART),and 90% received ART sustained HIV viral suppression.The first 90% target is the premise and the bottleneck.Its fulfillment needs to rely on HIV high-risk groups’ participation in HIV testing.MSM in China has beard the most rapidly growing of HIV epidemic in recent years.But less than 50% of this population has ever been tested for HIV,which is far from the 90-90-90 target.Meanwhile,voluntary counseling testing(VCT),as the most major way to provide free HIV testing to high-risk populations in China,has met challenges in response to the rapid growth and transformation of HIV epidemic among MSM populations: firstly,VCT rely on the inquiry from high-risk MSM,which failed to adapt to the current behavioral characteristics of the MSM;Secondly,the traditional VCT was limited by lack of technology and resources.It failed to "take the initiative" to reach out for the high-risk MSM and provide preventive intervention resources;Finally,VCT operates under the principle of anonymity testing.This leads to failure in identification of duplicate testing result from the same MSM,thus unable to obtain the most crucial marker of HIV epidemic surveillance,the HIV incidence rate.Not to mention the lognitude intervene and prevention.With the rapid popularization of the Internet,smartphones and online social software,and the wide application of electronic technology in Biomedicine,the traditional VCT has been given the innovative opportunity to expand its ability in monitor and prevention.Methods: in the first chapter of this study,we buid the digital prospective cohort by using and integrating the current existing Internet technologies.The strategy of the digital cohort includes: using and distribute the recruitment two-dimensional code(QR code)in MSM online and offline hotspot to obtain a representative samples,using the online self-administrated survey technology to obtain demographpic and behavioral information of the participants,using both the offline VCT laboratory testing and the online HIV/AIDS Syphilis self-test reagent to obtain the HIV and syphilis status of the participants.The goal of the study is to obtain the acceptance rate of the digital cohort strategy among MSM,the disparities of behavior and HIV infection rate between online and offline recruited participants,and finally obtained the HIV prevalence and correlates from this multiple-source representative samples of the local high risk MSM.In the second chapter of this study,we prospectively follow up the digital cohort of MSM based on VCT clinics.Using the real-time identification QR code to recognize the participants and ensure the VCT confidentiality principle.Using the self-administrated epidemiological survey technology to avoid social expectations bias and accurately obtain the participants’ demographic and HIV-related high-risk behaviral characteristics,behavioral characteristics.Syphilis and HIV laboratory test were used to detect the current syphilis infection and the HIV sero coversion.Finally,to obtain the HIV incidence,the cohort retention rate and correlates among high risk MSM and its subpopulation based on this digital cohort of VCT stragetegy.In the third chapter of this study,we explore the application and effect on HIV prevention targeted at MSM population of the online VCT by the real-world study design.We design and establishe the online VCT intervention function menu,including on-line consultation,the prevention information post and distribution,the test result on-line inquery and the HIV infection risk assessment tool,etc.By the real-time online footprint of the inquiry MSM caught by the backstage,we objectively evaluate the usage of the intervention menue and its effect on reduction of HIV related high risk behavior and incidence of HIV and syphilis.Results: 1.During the study period(March 2017 to February 2018),2,371 participants were recruited,including 689(29%)recruited from offline recruitment source and 1682(71%)recruited from online recruitment source.During screening,there was 1128 qualified for the study criteria,and among which 993 MSM participants concented to accept the digital cohort strategy and enrolled to the study,which yield to an acceptance rate of 88%(993/128).MSM who accepted compared to those who do not,had a higher proportion of being younger than 24-year-old,currently being a college student,more condomless anal intercourse(CAI).The syphilis antibody positive rate is 19.3%(191/993).The HIV prevalence was 5.7%(57/993)at baseline,with no significant difference between the online and offline recruited participants(5.5% VS.5.9%).Being migrants(VS.Resident population,adjusted or(a OR)=3.7,95%ci:1.8-7.8),had HIV testing frequency≦1 times per year(vs.2 times per year,a=4.4,95%CI:1.5-12.8),self-rated HIV infection risk as high(vs.risk-free,a OR=4.4,95%CI:1.6-11.7),inconsistent condom use with male sexual partners in the past 3 months(vs.cosistent using,a OR= 5.2,95%CI:1.8-15.0,p=0.002)and never use condom(vs.cosistent using,a OR=6.5,95%CI:2.1-20.2)were significantly correlated with HIV positive at baseline(p<0.05).2.During the study period,the digital cohort had prospectively followed up 731 HIV-negative high-risk MSM,yielding a total of 202.7 person-year observation,and observed 18 cases of HIV sero-coversion,resulting in a HIV incidence of 8.9/100 person-year(PY).Among the subgroup of MSM population,we report a 5.4/100 PY HIV incidence among young MSM(age ≦ 24 years old),a 13.1/100 PY HIV incidence among middle-aged and senior MSM(age ≧ 40 years old),a 22.0/100 PY HIV incidence among migrants MSM(lived in Shenyang≦2 years,a 3.8/100 PY HIV incidence of students MSM on campus.Being the migrants for more than 3 years(vs.has always been living in Shenyang,a HR=0.3,95%CI: 0.1-0.4),using the park/bar/bath and other non-Internet as major crusing venue(vs.Internet and other online venues,a HR=3.8,95%CI:1.3-10.8),inconsistent condom use with male sexual partners in the past-3-month(vs.always used,a HR=6.2,95%CI:1.7-22.9),syphilis infection history(vs.no syphilis infection,a HR= 2.5,95%CI:0.9-6.6),current syphilis infection(VS.a HR=4.6,95%CI:1.7-12.5)were significantly correlates of HIV seroconversion(p<0.05).During the study,440 MSM participated had attented at least one follow-up,which yields in a natural cohort retention rate as 60.2%(440/731.The median observation time was 177 days.Online recruited(vs.offline recruitment,a OR=2.4,95%CI:1.5-3.9),being migrants in Shenyang for more than 3 years(vs.has always been living in Shenyang,a OR=1.8,95%CI:1.2-2.8),HIV testing frequency≧2 times per year(vs.first time testers,a OR=4.5,95%CI:2.7-7.4),had CAI in the past-3-month(vs.no CAI,ao OR=2.5,95%CI:1.8-3.5),had more than 2 regular male sexual partners in the past-3-month(vs.no such partners,a OR=1.8,95%CI:1.1-2.8),recreational drug using behavior in the past-3-month(vs.no use,a OR=1.8,95%CI:1.3-2.7),had sexual intercourse under influence of alcohol in the past-3-month(vs.no such behaviour,a OR= 1.7,95%CI:1.3-2.4),has been diagnosed with sexual transmitted infections(STIs)in the past year(vs.undiagnosed of STIs,a OR=2.7,95%CI:0.9-8.2)were positive correlated with higher retention rates(p< 0.05).3.During the study,there were 28,818 person-times usage of the online VCT intervention function menu.We provided 2,311 person-times of counseling and prevention advices.There were 49.9%(14,369/28,818)of the usage took place during the none-working hours of traditional VCT(00:00-08:00,11:01-14:00,17:01-24:00),Among the 440 MSM followed up by the digital cohort,283 of them had enabled and used at least one online VCT intervention function.By comparing change of HIV-related high-risk sexual behavior between baseline and the lastest follow-up,we found that the incidence of CAI has decreased from 56.2% at baseline to 39.9% at the lastest follow-up,which yields into a 16.3% reduction in CAI(p<0.0001).There was no statistically significant reduction observed among the none-users.The use of online VCT intervention-type functions(HR=0.2,95%CI:0.1-1.1,p=0.065)and the use of maintenance-type functions(HR=0.8,95%ci:0.7-1.0,p=0.074)are margina associated with the reduction of syphilis incidence.Conclusion: 1.The digital cohort strategy is efficient in recruiting large samples-sized,multiple-sourced and representative high-risk MSM population samples.It has crucial prospects in future application.Based on this representative sample,we obtained the HIV prevalence of the local high-risk MSM.The HIV infections were significantly correlated with infrequent HIV testing behavior and low self-percieved HIV infection risk.2.Throug the VCT based digital cohort,we observed a high rate of HIV incidence and retention among high risk MSM populations and its subpopulations.Moreover,the strategy has the advantage in cost-efficiency of the maintenance of the migrants MSM,which needs to be urgently applied in the local VCT.3.The online VCT intervention function has high frequency of use among high-risk MSM population and it expands the time and ability of traditional VCT to provide service.The use of online intervention has significantly reduced the CAI and syphilis incidence among the MSM users.更多还原