利用网络规模迭加法估计新余市艾滋病相关高危人群规模

发布时间：2018-03-19 14:56

本文选题：艾滋病　切入点：网络规模迭加法　出处：《中国疾病预防控制中心》2014年硕士论文　论文类型：学位论文

【摘要】：研究目的利用网络规模迭加法(NSUM)估计新余市居民的社交网络规模,间接估计新余市艾滋病相关高危人群规模,探讨NSUM在中国用于艾滋病相关高危人群规模估计的可行性。研究方法在2013年10月至2014年2月间通过多阶段随机整群抽样选取调查对象,采用面对面入户调查的方式在新余市辖区进行居民问卷调查,应用NSUM估计社区居民的社交网络规模,通过人口学特征校正、反推估计剔除估计不准确的已知人群、箱式图剔除异常值的方式对社交网络规模的估计结果进行校正,应用校正后的社交网络规模、调查对象所认识的各类艾滋病高危人群的人数,间接估计新余市艾滋病相关高危人群规模。结果本研究共完成问卷1382份,有效问卷1325份,问卷有效率为95.88%；其中分宜县464份,渝水区861份；男性683名(51.55%),女性642名(48.45%),男女比例为1.06：1；平均年龄为35.65±17.04岁；文化程度以初中及以上为主；婚姻状况以在婚居多；多数调查对象在当地居住时间在3年及以上；新余市辖区居民社交网络规模的均值为697。不同县区(P0.05)、户籍类别(P0.05)、性别(P0.05)、年龄组(P0.05)的调查对象社交网络规模均呈显著性差异。应用NSUM估计新余市暗娼人群(FSWs)规模为2564(95%CI：2453-2676)人,嫖客人群为3030(95%CI:2909-3151)人,吸毒者为2017(95%CI：1918-42117)人,注射吸毒者为1576(95%CI：1488-1663)人,MSM为1329(95%CI：1249-1409)人,多性伴人群为4958(95%CI：4802-5113)人；嫖客人群与FSWs的规模比为1.18：1,注射吸毒者为吸毒者的78.14%；估计出FSWs的规模为新余市2013常住人口15-49岁女性人口总数的0.82%,嫖客人群的规模为新余市2013年常住人口中15-49岁男性人口总数的0.88%,MSM的规模为15-49岁男性人口总数的0.39%,多性伴人群占15-49岁人口总数的1.44%。结论利用NSUM可同时对多个高危人群进行规模估计,施行较为经济快捷,但已知人群的选择以及NSUM固有的传播效应、屏障效应以及估计效应均会影响规模估计的结果。应用NSUM的对暗娼和吸毒人群的估计结果较为可信,对人群流动性较大的嫖客和MSM人群的估计结果较传统规模估计方法偏低,而对注射吸毒人群的估计结果较当地公安部门登记数据偏高。本研究在借鉴国内外经验的基础上,根据调查地区的实际情况进行了相应调整,为今后开展NSUM的相关研究提供了一定的借鉴和指导建议。
[Abstract]:Research purpose. Using the method of network scale superposition (NSUM) to estimate the social network scale of residents in Xinyu city and indirectly estimate the scale of AIDS-related high risk population in Xinyu city, and to explore the feasibility of using NSUM to estimate the scale of AIDS-related high-risk population in China. Research method. From October 2013 to February 2014, the survey subjects were selected by multi-stage random cluster sampling, and the residents were surveyed by face-to-face household survey in Xinyu City area. NSUM was used to estimate the social network scale of community residents. Through demographic feature correction, backward estimation is used to eliminate known people whose estimation is not accurate, box diagram is used to eliminate outliers, and the result of social network scale estimation is corrected by using corrected social network scale. The size of AIDS-related high-risk population in Xinyu City was estimated indirectly by the number of AIDS-related high-risk groups. Results. In this study, 1382 questionnaires and 1325 valid questionnaires were completed, and the effective rate was 95.8888, including 464 in Yixian, 861 in Yushui District, 683 in male and 642 in female, the ratio of male to female was 1.066: 1; the average age was 35.65 卤17.04 years old; the education level was mainly in junior middle school and above, and the average age was 35.65 卤17.04 years old, and the average age was 35.65 卤17.04 years old, and the average age was 35.65 卤17.04 years, and the average age was 35.65 卤17.04 years old. Most of the respondents lived in the area for 3 years or more. The average size of social network of residents in Xinyu district was 697.There were significant differences in the scale of social network among different counties, households, sex and age group. NSUM estimates that the size of FSWs in Xinyu City is 256495 CIW: 2453-2676), the client population is 3030U 95 CI: 2909-3151), the drug addict is 2017-95 CI: 1918-42117), the injecting drug user is 157695CIW 1488-1663), the MSM is 1329995 CIW 1249-1409), and the majority of sexual partners is 495895 CIW 4802-5113). The ratio of clients to FSWs is 1.18: 1, and the ratio of injecting drug users to drug users is 78.14. The size of FSWs is estimated to be 0.82% of the total female population aged 15-49 of 2013 resident population in Xinyu City, and the size of the client population is 15-49 years old men of Xinyu City in 2013. The MSM of the total sex population is 0.39% of the male population aged 15-49, and the sexual partner group accounts for 1.44% of the total population aged 15-49 years. Conclusion. Using NSUM can estimate the scale of many high risk population at the same time, which is more economical and faster, but the choice of known population and the inherent transmission effect of NSUM. Both the barrier effect and the estimation effect will affect the results of the scale estimation. The results of the NSUM estimation for the prostitutes and drug users are more reliable, and the estimates for the people with high mobility and MSM population are lower than those of the traditional methods. However, the estimated results of injecting drug users are higher than that of local public security departments. Based on the experience of domestic and foreign countries, this study adjusts accordingly according to the actual situation of the investigation areas. It provides some reference and guidance for the research of NSUM in the future.
【学位授予单位】：中国疾病预防控制中心
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：R512.91

【参考文献】