Previous studies have reported a possibly increasing HIV prevalence among men who have sex with men (MSM) in China. However there have been limited systematic analyses of existing surveillance data to learn the trend of HIV prevalence and factors driving the trend. The aims of this study were to examine the trend of HIV prevalence among MSM in Guangzhou and to explore the role of unprotected anal intercourse (UAI) in the trend.
Snow-ball sampling was applied in the subject recruitment for the annual serological and behavioral surveys among MSM from 2008 to 2013. Data collected in the behavioral survey include demographic information, HIV related sexual behavior with men and women, access to HIV prevention services, and symptoms of sexually transmitted infections. Chi-square test was used to analyze the trend of HIV prevalence. Multivariate logistic regression was conducted to test the factors associated with HIV infection.
HIV prevalence increased significantly from 5.0% in 2008 to 11.4% in 2013 while syphilis prevalence decreased from 17.4% to 3.3% in the same period. UAI rates were high and stable in every single year, ranging from 54.5% to 62.0%. Those who were having UAI (OR = 1.80, 95% confidence interval (CI): 1.26–2.58), being migrants, having more than 10 partners, and infected with syphilis had higher risk for HIV infection.
HIV epidemic is expanding in Guangzhou. The persistently high UAI may have played a major role in the increasing trend of HIV prevalence. Targeted prevention program should be conducted among MSM who are migrants, low educational level, syphilis infected, or having multiple partners to encourage HIV test and change UAI behavior. The general high UAI calls for tailored intervention program to promote healthy culture and form a safe sex social norm in the MSM community.
Citation: Zhong F, Liang B, Xu H, Cheng W, Fan L, Han Z, et al. (2014) Increasing HIV and Decreasing Syphilis Prevalence in a Context of Persistently High Unprotected Anal Intercourse, Six Consecutive Annual Surveys among Men Who Have Sex with Men in Guangzhou, China, 2008 to 2013. PLoS ONE 9(7): e103136. https://doi.org/10.1371/journal.pone.0103136
Editor: Marti Vall, Catalan Health Institute, Spain
Received: April 7, 2014; Accepted: June 13, 2014; Published: July 25, 2014
Copyright: © 2014 Zhong et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The authors confirm that, for approved reasons, some access restrictions apply to the data underlying the findings. Data used in this paper are from Chinese national HIV sentinel surveillance program, and Guangzhou Center for Disease Control and Prevention owns the data. Data are available upon request. Such a request should be sent to Dr. Huifang Xu (email@example.com), the director of HIV sentinel surveillance program at Guangzhou Center for Disease Control and Prevention.
Funding: This research has been supported by the China Medical Board (12-111), Medical Scientific Research Foundation of Guangdong Province (A2013539), Medical Scientific Program of Guangzhou (20131A011114), and Science and Technology Program of Guangzhou (2012Y2-00021). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
During the past decade, the HIV epidemiology has changed in China, from predominantly driven by injection drug use and unsafe plasma collection to by unprotected sex . Among reported HIV/AIDS cases in China, the proportion of men who have sex with men (MSM) increased from 7.3% in 2005, to 11.0% in 2007 and 17.4% in 2011 . Unprotected sex between men has become one of the major HIV transmission modes in China, accounting for 29.4% of estimated annual new HIV infection in China in 2011 .
Guangzhou, the capital of Guangdong province and one of the largest cities located in southern China, is remarkable for its rapid economic growth during the past three decades, thanks to the open-door policy since 1978 . The open-door policy brought the opportunities not only for economic development, but for multiple culture environments. Guangzhou gradually became socially tolerant city, which attracts MSM from all over the country and other surrounding Asian cities. During the past decade, lots of gay cruising areas and partner hunting ways emerged, such traditional ones as gay bars, sauna, tea house, parks, public toilets and clubs, and modern ones as gay website, internet chat-room, instant chatting groups, and mobile networking. In addition, highways and high speedy trains make partner hunting very easy for MSM between Guangzhou and its neighboring cities. Easy mobility, together with advanced modern networking tools, incubates spreading of sexually transmitted diseases including HIV when sex is unprotected.
Unprotected anal intercourse (UAI) might have played a key role in the spread of HIV among MSM in Guangzhou. The relationship between UAI and HIV infection and sexually transmitted infections (STIs) is biologically plausible and approved by numerous studies in the United States and Australia –. Reports in different Chinese cities indicated that UAI has increased or maintained at a high level, despite a high level of knowledge about HIV/AIDS, since early 2000s when most of the studies were conducted .
Previous surveys have reported a possibly increasing average HIV prevalence among MSM in China, from 0.9% in 2003, 1.3% in 2004, 3.0% in 2006 and 5.0% in 2008 . Increased HIV prevalence has been reported in big Chinese cities. In Beijing, HIV prevalence of MSM increased from 0.4% in 2004 to 5.8% in 2006, and 8.0% in 2009 , . In Harbin, the prevalence of HIV increased from 1.0% in 2006 to 7.5% in 2010 . The HIV prevalence was reported to be increasing in some medium-sized cities like Hangzhou (from 1.8% in 2006 to 8.3% in 2009) , Jinan (from 0.05% in 2007 to 3.1% in 2008) , Suzhou (from 7.1% in 2008 to 8.2% in 2012) , . This increase is in the context of a regional increase in Asia such as from 17.3% in 2003 to 28.3% in 2005 in Bangkok , from 2.5% in 2002 to 8.0% in 2007 in Jakarta , from 9.4% in 2006 to 20% in 2010 in Hanoi .
There might be an increase in HIV prevalence in Guangzhou. However, there have been limited systematic analyses of existing surveillance data to examine the trend of HIV prevalence and factors driving the trend . Taking advantage of being part of 61 cities' survey among MSM in China organized by Chinese Center for Disease Control and Prevention (China CDC) , Guangzhou started the annual biological and behavioral surveys among MSM since 2008 using an identical survey protocol and the same questionnaire, implemented by the same group of interviewers at the same survey site.
In the present study, we examined the trend of HIV prevalence and behaviors using annual biological and behavioral surveys from 2008 to 2013. We also explored the role of UAI in the trend of HIV prevalence.
Materials and Methods
Participants and recruitment
Subjects included in the surveys were men who had anal or oral sex with men in the last 12 months, at least 18 years old and had lived in Guangzhou for at least 6 months prior to the survey year.
Snow-ball sampling was applied in the subject recruitment. An incentive, including a gift (worth about 2 US dollars) and 20 Chinese Yuan cash (approximately 3 US dollars), was given for participation in the questionnaire survey and the provision of 5 ml blood for serological testing. Pre- and post-test counseling services were provided to each of subjects participating in the study. Referral services were provided to HIV or syphilis positive cases.
The sample size was estimated based on the following formula :
The MSM population size (N) in Guangzhou was estimated to be about 35,000 (Personal communication, Huifang Xu, unpublished data based on capture-recapture exercise in 2008). HIV prevalence among MSM (p) was 5.2% in 2008 , confidence limit (d) of 5%, a design effect (DEFF) of 2, and an alpha (α) of 0.05. Given these conditions, the required smallest sample size would be 152. The sample size was increased to 400 to allow for greater representativeness and possible refusals during the recruitment.
The study protocol was developed by Guangzhou CDC and approved by its Ethics Committee. All subjects provided written informed consents.
After written informed consent obtained from each of the eligible subjects, 5 ml intravenous blood was drawn prior to the questionnaire interview. Face to face questionnaire interviews were conducted by the same interviewers using an identical structured questionnaire in a private room at the voluntary counseling and testing (VCT) clinic of Guangzhou CDC.
Data collected in the questionnaire include demographic information, knowledge and attitudes towards HIV or AIDS, access to HIV prevention services, and sexual behavior with men and women in the past 6 months, and symptoms of STIs in the past 12 months. The questionnaire interviews were administered anonymously but such contact information as mobile phone numbers, QQ (Instant messaging software) account or email address were collected to inform the laboratory test results and referral services when necessary.
HIV screening was conducted by using two enzyme-linked immunoassays (ELISAs; Diagnostic Kit for Antibody to Human Immunodeficiency Virus, BioMérieux, Boxtel, The Netherlands, Beijing BGI-GBI Biotech, Beijing, China). If the result of one ELISA was positive, a Western Blot (WB) test was conducted for confirmation (WB;MP Biomedicals Asia Pacific Pte Ltd, Singapore). Syphilis screening was performed by rapid plasma regain (RPR; Shanghai Kehua Bioengineering Co. Ltd, Shanghai, China) to test anticardiolipin. Specimens testing positive by RPR were confirmed by the treponema pallidum particle agglutination test (TPPA; Livzon Group Reagent Factory, Zhuhai, China) to test treponema pallidum antibody. Specimen which is positive in both RPR and TPPA can be diagnosed to be syphilis infection.
Although the sampling method of survey was a non-probability sample, the sample size was considered large enough to be representative of the population. In this study, UAI was defined as failure to consistently use a condom during the past 6 months when having anal sex. Having no anal sex reported was considered to be safe. UAI was categorized as ‘No’ if there was no anal intercourse or a condom was used every time, and ‘Yes’ if a condom was not used all the time. Unprotected vaginal intercourse (UVI) was categorized in the same method. Coverage of prevention services was defined as receiving any service including condom distribution, lubricant distribution, peer education, STI diagnosis or treatment, HIV counseling or testing, or AIDS/STI educational materials in the past 12 months.
Data were double entered and cleaned using EpiData (version 3.1, Denmark). SPSS (version 17.0, LEAD Technologies Inc.) was used to perform Pearson chi-square test, linear-by-linear association chi-square test, likelihood ratio chi-square test, and multivariate logistic regression analysis. The multivariate forward stepwise logistic regression was conducted to test the factors associated with HIV infection. The independent variable input in the regression model of HIV infection included age, marital status, Hukou (registered permanent residence), education level, ethnicity, monthly income, age at first sex with male, years of being MSM, sexual orientation, venues for meeting partners, UAI, UVI, HIV testing history, preventive intervention services received, STI symptoms in the past 12 months, syphilis infection, and with survey year as dummy variable. All statistical significance test results are reported as p-values, where less than 0.05 was used to define significance.
From 2008 to 2013, 379, 385, 405, 400, 401 and 633 subjects were recruited respectively during April to July each year. The demographic characteristics and key behaviors of the participants are shown in Table 1. From 2008 to 2013, along with the development of economy and culture in China, the increasing trends were found in some demographic data, including increased proportions of being single, having higher educational level and income, and being internet surfers for meeting partners. There was an increase in the rates of HIV testing and coverage of HIV intervention during the past 12 months from 2008 to 2013. In the meantime, the proportion of subjects, who reported STI symptoms during the past 12 months, with multiple partners, or having UVI in the past 6 months, decreased statistically. Syphilis prevalence decreased from 17.4% in 2008 to 3.3% in 2013 during the same period of time. However, across the six years, the most commonly reported risk behavior remained stable, the percentage of UAI still range 54.5% to 62.0% (p>0.05) (Table 1).
HIV prevalence increased significantly from 5.0% in 2008 to 11.4% in 2013 (p<0.001). The increasing trend of HIV prevalence was also found among most of subgroups (Table 2). While in some groups, the rates have been stable over six years. HIV prevalence did not increase significantly in those who cohabited with male partners, having educational level of junior high school or lower, having monthly income lower than 2000 Chinese Yuan, meeting sex partners in bar or bath house, having 4 or more male partners, and syphilis infected, and prevalence among them was higher than average rate in each single year. In another aspect, HIV prevalence did not increase in subgroups those who having income higher than 4000, having less than 2 partners, and having no UAI in the past 6 months, and prevalence in these groups was lower than average rate.
Factors associated with HIV infection were reported in Table 3. Those having UAI (OR = 1.80, 95% CI: 1.26–2.58), without Guangdong Province Hukou (OR = 2.02, 95% confidence interval (CI): 1.33–3.07, compared with those with Guangzhou local Hukou), having more than 10 partners (OR = 2.10, 95% CI: 1.14–3.87, compared with those having only 1 partner), and being syphilis positive (OR = 2.72, 95% CI: 1.74–4.26) were associated with higher risk for HIV infection. Having a college or higher education level (OR = 0.54, 95% CI: 0.34–0.85) and having HIV test in the past 12 months (OR = 0.73, 95% CI: 0.53–1.00) were associated with lower risk for HIV infection. After adjusting for demographic characteristics and key behaviors of the participants, the survey year remained an associated factor for HIV infection, indicating an increasing trend in HIV prevalence over years.
UAI rates were high and stable in every single year, ranging from 54.5% to 62.0%. In Table 4, further analysis indicated the UAI among HIV-positive MSM was higher than that among HIV-negative MSM (71.0% versus 57.2%, p<0.001). UAI was generally higher among HIV positive cases in many subgroups especially those having a Hukou in other cities in Guangdong Province, having a college or higher education level, being syphilis positive, and having multiple male partners (Table 4).
Results from 6 consecutive annual surveys indicated an increasing HIV prevalence among MSM in Guangzhou. The increase is a continuation of increasing trend since 2006 , coincides with the increase in the neighboring cities like Shenzhen , in similar Chinese metropolitan cities , , – and in the Asian region , , , . Studies in similar metropolises such as Beijing and Shenyang also reported the high HIV incidence of 8.1 and 5.4 per 100 person-year, respectively , . The increase of epidemic in China is different from the trend in San Francisco where HIV epidemic stabilized among MSM during the past decade in a context of decreasing HIV incidence and increased survival of HIV positive individuals due to ART .
The increase in HIV prevalence was also found in the most subgroups, by most demographic variables and related behaviors. This is consistent with our previous studies and the results from some studies in China , –. Stable prevalence over years was only found in some groups with high risk factors (having educational level of junior high school or lower, having 4 or more male partners, and syphilis infected) or protective factors (having less than 2 partners and no UAI), remaining high or low prevalence, respectively. In our study, after adjusting for years, UAI, migrant, multiple male partners, and syphilis infection were independent related risk factors of HIV infection, while higher educational level and taking an HIV test were protective factors.
Contrary to the trend of HIV prevalence, decreasing syphilis prevalence was found among MSM in Guangzhou from 2008 to 2013. Syphilis is a treatable disease which is passed on through sexual intercourse, therefore syphilis infection can be viewed as a marker for high-risk sexual practices, such as unprotected sex and/or multiple partners, and can also be controlled through screening and treatment. Previous study showed receiving HIV test is a protective factor of syphilis infection among MSM in Guangzhou , because when they take an HIV test, they can also receive free syphilis screening and referral services of treatment besides behavior intervention at the same time. Although there is limited data of syphilis screening and treatment among MSM in Guangzhou, our study showed the rate of having HIV test in the past 12 months has increased significantly during six years, arriving 53.4% in 2013, thus the rise of HIV test rate may contribute to control syphilis infection among these MSM. From 2008 to 2013, risk factors found associated with HIV infection decreased while protective factors increased across years. However, UAI, being consistently high and stable from 2008 to 2013, did not change from the level observed in our previous surveys and were not different from those observed in other studies in China , , , –. The consistently high level of UAI may explain the steady increase in HIV prevalence since UAI is the only biologically plausible but uncontrolled risk factor related to HIV during the past six years. HIV positive cases have higher UAI than those not infected yet. The finding, of 100% UAI among those aged below 20 years, is of a particular concern. This 100% UAI in the younger age group, together with high level of UAI in all subgroups, indicates an unhealthy culture environment and incorrect social norm in the MSM community. Social norms refer to expectations of acceptable behavior or attitudes within a community or peer group prescribed by the respective members . The social norms of MSM community are to protect himself and his partners anytime having sex. Research among Chinese MSM found that stronger endorsement of positive social norms around condom use strongly predicted lower prevalence of HIV infection . Therefore intervention program which focus on positive social norms and condom using may protect MSM from being infected with HIV.
This study has some limitations. Despite the snow-ball sampling applied, the majority of the sampled MSM were from venues or internet. This may bias the samples towards the relatively active MSM subgroups. Similar to other behavioral surveys, behavioral information was relied on self-reporting, recall bias and social desirability bias may exist. Information bias may also exist given the nature of sensitive sex-related questions. To address these possible biases, the same group of trained interviewers conducted all six surveys at the same site using an identical questionnaire. Consistency was ensured in the methodology and implementation of questionnaire interviews, biases may be towards one direction across years. The risk behaviors, such as UAI, reported by interviewees may be conservative due to social desirability bias. Despite these limitations, the present study is one of the few studies in China to examine the trend of the prevalence of HIV and behaviors among MSM for 6 consecutive years. In further research, cohort study will be conducted to measure incidences of HIV and syphilis among MSM and to modify cause inference.
In conclusion, HIV epidemic is expanding in Guangzhou. The persistently high UAI may have played a major role in the increasing trend of HIV prevalence. Targeted prevention program should be conducted among MSM who are migrants, low educational level, syphilis infected, or having multiple partners to encourage HIV test and change UAI behavior. The general high UAI calls for tailored intervention program to promote healthy culture and form a safe sex social norm in the MSM community.
We would like to thank colleagues in Department of AIDS Control and Prevention, Guangzhou Center for Disease Control and Prevention, for the implementation of surveys. We also feel grateful for MSM volunteers of Lingnan Partner Health Center who help the mobilization and organization efforts during the field surveys and participants for their voluntary participation.
Conceived and designed the experiments: FZ HX WC LF ZH. Performed the experiments: FZ WC LF ZH CL KG HM FQ. Analyzed the data: FZ BL. Contributed to the writing of the manuscript: FZ BL JZ LL.
- 1. Lu F, Wang N, Wu Z, Sun X, Rehnstrom J, et al.. (2006) Estimating the number of people at risk for and living with HIV in China in 2005: methods and results. Sex Transm Infect 82 : iii 87–91.
- 2. Ministry of Health of the People's Republic of China (2012) National report for HIV/AIDS estimation in China, 2011 [In chinese]. Beijing: Chinese Center for Disease Control and Prevention.
- 3. Bui TX YD, Jones WD, Li JZ (2003) China's economic powerhouse: economic reform in Guangdong Province. New York, NY: Palgrave Macmillan. 288p.
- 4. Chris B, Andrea LW, Damian W, Benjamin J, Frangiscos S, et al.. (2011) The Global HIV Epidemics Among Men who Have Sex with Men. Washington DC: The World Bank. 350p.
- 5. Ackers ML, Greenberg AE, Lin CY, Bartholow BN, Goodman AH, et al. (2012) High and persistent HIV seroincidence in men who have sex with men across 47 U.S. cities. PloS one 7: e34972.
- 6. Jin F, Prestage GP, Zablotska I, Rawstorne P, Imrie J, et al. (2009) High incidence of syphilis in HIV-positive homosexual men: data from two community-based cohort studies. Sex Health 6: 281–284.
- 7. Liu J, Qu B, Ezeakile MC, Zhang Y (2012) Factors associated with unprotected anal intercourse among men who have sex with men in Liaoning Province, China. PloS one 7: e50493.
- 8. Wang L, Norris JL, Li DM, Guo W, Ding ZW, et al. (2012) HIV prevalence and influencing factors analysis of sentinel surveillance among men who have sex with men in China, 2003–2011 [In chinese]. Chin Med J (Engl) 125: 1857–1861.
- 9. Ma X, Zhang Q, He X, Sun W, Yue H, et al. (2007) Trends in prevalence of HIV, syphilis, hepatitis C, hepatitis B, and sexual risk behavior among men who have sex with men. Results of 3 consecutive respondent-driven sampling surveys in Beijing, 2004 through 2006. J Acquir Immune Defic Syndr 45: 581–587.
- 10. Fan S, Lu H, Ma X, Sun Y, He X, et al. (2012) Behavioral and serologic survey of men who have sex with men in Beijing, China: implication for HIV intervention. AIDS Patient Care STDS 26: 148–155.
- 11. Wang K, Yan H, Liu Y, Leng Z, Wang B, et al. (2012) Increasing prevalence of HIV and syphilis but decreasing rate of self-reported unprotected anal intercourse among men who had sex with men in Harbin, China: results of five consecutive surveys from 2006 to 2010. Int J Epidemiol 41: 423–432.
- 12. Zhang RS, Shang CX (2011) Investigation on the infection of HIV and Sex Transmitted Diseases in the man who have sex with mam from 2006 to 2009 in Hangzhou [In chinese]. Chinese J of Health Laboratory Technology 21: 732–733.
- 13. Ruan S, Yang H, Zhu Y, Wang M, Ma Y, et al. (2009) Rising HIV prevalence among married and unmarried among men who have sex with men: Jinan, China. AIDS Behav 13: 671–676.
- 14. Bai H, Huan X, Tang W, Chen X, Yan H, et al. (2011) A survey of HIV infection and related high-risk factors among men who have sex with men in Suzhou, Jiangsu, China. J Biomed Res 25: 17–24.
- 15. Hao C, Lau JT, Zhao X, Yang H, Huan X, et al. (2014) Associations Between Perceived Characteristics of the Peer Social Network Involving Significant Others and Risk of HIV Transmission Among Men Who Have Sex with Men in China. AIDS Behav 18: 99–110.
- 16. Centers for Disease Control and Prevention (CDC) (2006) HIV prevalence among populations of men who have sex with men—Thailand, 2003 and 2005. Morb Mortal Wkly Rep 55: 844–848.
- 17. Morineau G, Nugrahini N, Riono P, Nurhayati, Girault P, et al. (2011) Sexual risk taking, STI and HIV prevalence among men who have sex with men in six Indonesian cities. AIDS Behav 15: 1033–1044.
- 18. Garcia MC, Meyer SB, Ward P (2012) Elevated HIV prevalence and risk behaviours among men who have sex with men (MSM) in Vietnam: a systematic review. BMJ Open 2:pii: e001511.
- 19. Lau JT, Lin C, Hao C, Wu X, Gu J (2011) Public health challenges of the emerging HIV epidemic among men who have sex with men in China. Public Health 125: 260–265.
- 20. Scheaffer RL, Mendehall W, Ott RL, Gerow KG (2011) Elementary Survey Sampling, Seventh Edition. Boston, Cengage Learning. 448p.
- 21. Zhong F, Lin P, Xu H, Wang Y, Wang M, et al. (2011) Possible increase in HIV and syphilis prevalence among men who have sex with men in Guangzhou, China: results from a respondent-driven sampling survey. AIDS Behav 15: 1058–1066.
- 22. Zhao J, Cai WD, Gan YX, Zhang Y, Yang ZR, et al. (2012) A comparison of HIV infection and related risk factors between money boys and noncommercial men who have sex with men in Shenzhen, China. Sex Transm Dis 39: 942–948.
- 23. Zhou C, Raymond HF, Ding X, Lu R, Xu J, et al. (2012) Anal Sex Role, Circumcision Status, and HIV Infection Among Men Who Have Sex with Men in Chongqing, China. Arch Sex Behav 42: 1275–1283.
- 24. Wang Z, Lau JT, Hao C, Yang H, Huan X, et al. (2013) Syphilis-related perceptions not associated with risk behaviors among men who have sex with men having regular male sex partner(s) in Nanjing, China. AIDS Care 25: 1010–1017.
- 25. She M, Zhang HB, Wang J, Xu J, Duan YW, et al. (2012) Investigation of HIV and syphilis infection status and risk sexual behavior among men who have sex with men in four cities of China [In chinese]. Zhonghua Yu Fang Yi Xue Za Zhi 46: 324–328.
- 26. Centers for Disease Control and Prevention (CDC) (2013) HIV and Syphilis Infection Among Men Who Have Sex with Men - Bangkok, Thailand, 2005–2011. Morb Mortal Wkly Rep 62: 518–520.
- 27. Mimiaga MJ, Biello KB, Sivasubramanian M, Mayer KH, Anand VR, et al. (2013) Psychosocial risk factors for HIV sexual risk among Indian men who have sex with men. AIDS Care 25: 1109–1113.
- 28. Li D, Li S, Liu Y, Gao Y, Yu M, et al. (2012) HIV incidence among men who have sex with men in Beijing: a prospective cohort study. BMJ Open 2: e1810–e1829.
- 29. Xu JJ, Zhang M, Brown K, Reilly K, Wang H, et al. (2010) Syphilis and HIV seroconversion among a 12-month prospective cohort of men who have sex with men in Shenyang, China. Sex Transm Dis 37: 432–439.
- 30. Raymond HF, Chen YH, Ick T, Scheer S, Bernstein K, et al. (2013) A new trend in the HIV epidemic among men who have sex with men, San Francisco, 2004–2011. J Acquir Immune Defic Syndr 62: 584–589.
- 31. Zhang L, Chow EP, Jing J, Zhuang X, Li X, et al. (2013) HIV prevalence in China: integration of surveillance data and a systematic review. Lancet Infect Dis 13: 955–963.
- 32. Meng X, Zou H, Beck J, Xu Y, Zhang X, et al. (2013) Trends in HIV prevalence among men who have sex with men in China 2003-09: a systematic review and meta-analysis. Sex Health 10: 211–219.
- 33. Guy RJ, Wand H, Wilson DP, Prestage G, Jin F, et al. (2011) Using population attributable risk to choose HIV prevention strategies in men who have sex with men. BMC Public Health 11: 247.
- 34. Peterson JL, Rothenberg R, Kraft JM, Beeker C, Trotter R (2009) Perceived condom norms and HIV risks among social and sexual networks of young African American men who have sex with men. Health Educ Res 24: 119–127.