基于在线社交网络男男性行为人群的HIV超级传播者筛查及干预策略研究

Men who have sex with men (MSM) are disproportionately driving the rising of HIV incidence rate as key targeted population in China HIV response. Online- social-network brings new type of aggregation of HIV transmission risk. Our previous studies showed that the social network was the main channel for seeking sexual partners among MSM and that finding could be used for the design of targeted interventions. We propose to use Deep Structured Semantic Model (DSSM) analysis of big data to capture the high-risk clusters in the online social network. The DSSM will be used to search the terms or queries regarding risky behaviors in order to detect high risk hot points who might be the potential HIV hyper-transmitters. We will identify 10,000 hot points via DSSM and send recruiting information via a pop-up dialogue box or banner in the main page of the mobile App. Two thousand participants will be surveyed and taken a blood specimen for HIV serological test and all HIV positive specimens will be performed high-throughput sequencing to detect HIV subtype and mutation level for drug resistance. Based on the characteristic of HIV transmission risks clustering, we will be able to develop an integrated intervention strategy that harmonize App users’ interest and information needs and generate an innovative way of HIV preventive information dissemination by personalized push. The identified social hot points will be mobilized to further disseminate the key information on condom use, HIV testing, and adherence to antiviral therapy. With this research, we will try to inform the policy makers of the development of Internet-adaptive intervention strategy to address the alarmingly high HIV incidence rate of MSM.

男男性行为人群（MSM）是中国防控HIV新发感染上升的重点和难点人群，在线社交网络为MSM提供了HIV传播的新的风险聚集模式。课题组前期研究发现针对MSM的社交网络开展干预是有效的策略。本研究通过分析北京MSM社交网络大数据中的高危行为词频，构建深度结构语义模型，筛查MSM社交网络中潜在的HIV感染高危亚群，确定1万个社交热点；采用社交网络平台推送信息，招募其中2,000人开展流行病学调查，检测HIV血清学状态，对HIV阳性标本的RNA进行高通量测序和生物信息学分析，判断HIV亚型和药物敏感谱，寻找耐药相关核酸变异，确定HIV超级传播者；针对MSM在线社交网络的风险聚集特点，建立以MSM信息需求为导向，以专家筛选与包装的干预信息内容为核心，对用户兴趣归类，实现个性化HIV干预信息推送，动员社交热点参与的二级传播干预策略，为遏制HIV的传播，建立适应互联网时代的MSM干预策略提供科学依据

男男同性性行为人群（MSM）因其高发的无保护性行为方式和隐蔽的性伴网络是HIV防控的重点和难点，而在线社交网络为HIV传播提供了新的风险聚集模式。本研究通过手机APP在线捕捉MSM中HIV高危亚群，基于中国互联网信息收集政策环境的变化，将最初的研究策略直接进行社交信息抓取分析的策略，调整为采用线上问卷调查进行危险行为评分以及高风险人群“知-信-行”一致性分析；通过横断面研究和重复测量数据纵向研究分析MSM的HIV重复检测行为与社交特征的关系及HIV感染的危险因素；进行HIV血清学检测和阳性标本的实时PCR，提取RNA进行高通量测序和生物信息学分析，检测HIV亚型和耐药核酸相关变异；通过基于网络数据的文本挖掘方法和词频对比分析，了解MSM对于HIV知识需求和主流推送信息现状；探索以HIV知识实际需求为导向，不同知识分类的干预信息内容为核心，个性化干预为原则，动员社交热点参与为重点的干预策略。适应国家对大学生群体中HIV感染病例持续上升的关注，重点优先研究了大学生群体的社交网络特征及其社会社交人群的关联与分布，为遏制HIV传播及建立适应互联网时代的MSM干预策略提供科学依据。