Objective: To evaluate the effect of the "Interner Plus-based AIDS Comprehensive Prevention Service System" among MSM in Guangzhou, during 2010-2017, using a dynamic compartmental model. Methods: A dynamic compartmental model was developed to describe the HIV situation among MSM in Guangzhou. This model was parameterized on data from published literature or surveillance programs from the Guangzhou CDC. The Matlab 7.0 software was used for coding and analysis on collected data. HIV prevalence was analyzed among MSM under the status quo data and estimated the impact by the "Internet Plus" AIDS prevention services project. Results: HIV prevalence would have increased to 22.75% in 2017, and the total number of new HIV infections would have been 11 038, from 2010 to 2017, using the data status quo. Under the Guangzhou "Internet Plus" AIDS prevention services project, the prevalence of HIV among MSM from 2010 to 2017 was estimated to be 8.44%, 9.68%, 10.65%, 11.34%, 11.73%, 11.83%, 11.71% and 11.43% in Guangzhou, which were similar to the surveillance data. The total number of new infections in the past 8 years under the "Internet Plus" scenario was estimated to be 4 009. The "Internet Plus" program would have prevented 7 029 (63.68%) new infections during 2010-2017 as compared to the number, status quo. Conclusions: The fitting result of dynamic compartmental model seemed more reasonable, which was applicable to predict HIV epidemic among MSM in Guangzhou, suggesting that the increase of HIV prevalence had been curbed since the "Internet Plus" project which was launched in 2010, and the "Interner Plus-based AIDS Comprehensive Prevention Service System" had achieved the purpose as planned, epidemiologically.
目的： 应用传染病动力学模型评估"互联网+艾滋病综合预防服务体系"对2010－2017年广州市互联网型MSM艾滋病流行趋势的影响。 方法： 构建广州市互联网型MSM的传染病动力学模型（模型），收集整理广州市哨点监测的互联网型MSM相关数据，进行文献检索，调整模型运行的输入参数及校正参数。运用Matlab 7.0软件将模型参数及微分方程进行编程和计算，对2010年互联网艾滋病预防服务开展前后，分析MSM HIV流行趋势，评估广州市互联网艾滋病预防服务对MSM艾滋病的防治效果。 结果： 模型估计，按照2010年之前的基线情况下，模型估计2017年MSM HIV感染率将上升至22.75%，2010－2017年累计新发HIV感染者11 038例。按照开展互联网艾滋病预防服务的情况下，模型估计2010－2017年MSM的HIV感染率分别为8.44%、9.68%、10.65%、11.34%、11.73%、11.83%、11.71%和11.43%，与哨点监测数据相近，模型拟合度高；2010－2017年累计新发HIV感染者4 009例，减少新发HIV感染者7 029例（63.68%）。 结论： 构建的模型拟合结果较为合理，适用于预测广州市互联网型MSM HIV流行趋势；广州市自2010年建立"互联网+艾滋病综合预防服务体系"以来，到2015年基本遏制了互联网型MSM新发HIV感染的上升势头，其防治效果较好。.
Keywords: Dynamic model; HIV/AIDS; Internet Plus; Men who have sex with men.