Factors Associated with Loss-to-Follow-Up during Behavioral Interventions and HIV Testing Cohort among Men Who Have Sex with Men in Nanjing, China
Factors Associated with Loss-to-Follow-Up during Behavioral Interventions and HIV Testing Cohort among Men Who Have Sex with Men in Nanjing, China
Weiming Tang 0 1 2 5
Xiping Huan 1 5
Ye Zhang 0 5
Tanmay Mahapatra 3 5
Jianjun Li 1 5
Xiaoyan Liu 1 5
Sanchita Mahapatra 3 5
Hongjing Yan 1 5
Gengfeng Fu 1 5
Jinkou Zhao 4 5
Chenghua Gu * 1 5
Roger Detels 3 5
0 Guangdong Provincial Center for Skin Disease and STI Control , Guangzhou, Guangdong, 510095, China,
1 Jiangsu Provincial Central for Disease Control and Prevention , Nanjing, Jiangsu, 210009, China,
2 University of North Carolina Project-China , Number 2 Lujing Road, Guangzhou, 510095, China,
3 Department of Epidemiology, School National Institute of Cholera and Enteric Diseases , Kolkata, West Bengal , India,
4 Strategy, Investment and Impact Division, The Global Fund to Fight AIDS , Tuberculosis and Malaria, Geneva , Switzerland
5 Editor: Hong Shang, China Medical University , China
Background: Behavioral interventions (BIs) remained the cornerstone of HIV prevention in resource-limited settings. One of the major concerns for such efforts is the loss-to-follow-up (LTFU) that threatens almost every HIV control program involving high-risk population groups. Methods: To evaluate the factors associated with LTFU during BIs and HIV testing among men who have sex with men (MSM), 410 HIV sero-negatives MSM were recruited using respondent driven sampling (RDS) in Nanjing, China during 2008, they were further followed for 18 months. At baseline and each follow-up visits, each participant was counseled about various HIV risk-reductions BIs at a designated sexually transmitted infection (STI) clinic. Results: Among 410 participants recruited at baseline, altogether 221 (53.9%) were LTFU at the 18-month follow-up visit. Overall, 46 participants were found to be positive for syphilis infection at baseline while 13 participants were HIV seroconverted during the follow-up period. Increasing age was less (Adjusted Odds Ratio(aOR) of 0.90, 95% confidence Interval (CI) 0.86-0.94) and official residency of provinces other than Nanjing (AOR of 2.49, 95%CI 1.32-4.71), lower level of education (AOR of 2.01, 95%CI 1.10-3.66) and small social network size (AOR of 1.75, 95%CI 1.09-2.80) were more likely to be associated with higher odds of
Conclusion: To improve retention in the programs for HIV control, counseling and
testing among MSM in Nanjing, focused intensified intervention targeting those who
were more likely to be LTFU, especially the young, less educated, unofficial
residents of Nanjing who had smaller social network size, might be helpful.
Given that HIV is still incurable, non-availability of a safe and effective vaccine,
limited access to anti-retroviral treatment (ART) and declining international
fund, till date, preventive approaches through behavioral interventions (BIs) and
voluntary counseling and testing (VCT) remain the cornerstone of HIV control
. Researchers have argued that a considerable reduction in HIV transmission
could be successfully achieved through a combined effect of wide spread and
sustained behavioral changes by appropriate counseling of a good number of
potentially high risk individuals [4, 5]. Analysis of data from three countries
around the world having different cultures and diverse HIV epidemics: Uganda,
Senegal and Thailand, clearly indicated that behavioral changes were successful in
some containment of HIV epidemic . Likewise, VCT has been also effective
in motivating people to change their risky behaviors and previous meta-analyses
demonstrated that VCT recipients were less likely to engage in unsafe sex than
VCT non-recipients [11, 12]. While the UNAIDS report 2012  revealed an
increasing trend towards uptake of HIV testing globally, it was estimated that only
half of all people living with HIV (PLHIV) knew their HIV sero-status and
probably HIV prevention programs were not adequately reaching the population
groups at highest risk. Globally, population at highest risk like female sex workers
(FSW), men who have sex with men (MSM) and injecting drug users (IDUs)
continue to be disproportionately affected by HIV [1, 13, 14]. It was estimated that
odds of having HIV infection among MSM on average was 13 times more
compared to general population in any capital city of the world . Although
prior research indicated that counseling and BIs were effective in reducing high
risk sexual behavior among MSM , worldwide the median coverage of risk
reduction programs among MSM was estimated to be 55%, only few countries
reported 75% consistent condom use and knowledge about their own HIV
serostatus was reported to be exceptionally low among MSM .
China has also observed a similar pattern in HIV epidemic among MSM
population in recent years. The estimated HIV prevalence among Chinese MSM
was found to be quite high [14, 1821]. At the end of 2011, the estimated number
of PLHIV was 780,000, 48,000 got newly infected and 17.4% of PLHIV were
infected through homosexual route in China [18, 20]. Approximately 18 million
men were estimated to be involved in homosexual activities in 2011 and HIV
transmission between homosexuals rose from 0.3% before 2005 to more than
13.7% in 2011 . Even though the Chinese government recently recognized the
emerging burden of HIV in MSM communities and emphasized the need to
expand the coverage of BIs and HIV testing among them, surveillance and
targeted prevention strategies among these men were still under-developed and
non-specific . An analysis of 2011 surveillance data revealed that
approximately 77% MSM were covered by prevention programs, 74% reported
using a condom during the last sexual act and only half of them were tested for
HIV and were aware of their sero-status . Major public health challenges of
the emerging HIV epidemic among MSM were delayed national response to the
HIV epidemic among MSM until 2007, lack of experienced public health workers
being engaged in delivering effective intervention to this hard-to-reach
population, unexplored bisexual behaviors as many had traditional marriage to
hide their homosexuality while continuing to have a secret same sex relationship,
poor societal acceptance of same sex behavior leading to rapid congregation of
MSM communities in different venues of metropolitan areas like sex clubs, bars,
saunas, bathhouses, public toilets and parks which promoted homosexual
prostitution, erotic activities and practice of unprotected sex and lastly under
utilization of available HIV prevention services [19, 23, 24]. Furthermore, most of
the previous studies involving MSM in China had focused on assessing factors
associated with HIV risk and evidence-based intervention studies targeting this
group were less documented [24, 25]. To the best of our knowledge, till date only
three published papers have documented the role of preventive intervention
strategies among MSM in China. Out of these three BIs, two peer-based
communication programs regarding safer sex behaviors were effective in
promoting condom use and HIV testing among MSM in Chengdu  and
Anhui province of China  while the efficacy of another internet-based HIV.
BIs among MSM in Hong Kong was inconclusive . Although BIs and
counseling were effective in altering high risk sexual behaviors, acceptance of these
prevention services among MSM was low. To improve the acceptance of such
programs among MSM, comprehensive understanding of the predictors of being
loss-to-follow-up (LTFU) among MSM seemed crucial.
The objective of this study was to identify socio-demographic and behavioral
characteristics of MSM who were LTFU during BIs and HIV testing among HIV
sero-negative MSM attending a designated sexually transmitted infection (STI)
clinic of the Jiangsu Provincial Center for Disease Control in Nanjing, China, so
that adequate counseling targeting these at-risk groups could help in the
improvement of retention in HIV control programs.
Study Design and Sampling Methods
Details of the sampling methods and recruitment of the participants were
described elsewhere [29, 30]. Briefly, respondent-driven sampling (RDS) was used
to recruit the eligible and consenting participants. Male persons, aged 18 years or
more and reported to have had oral and/or anal sex with men during the past 12
months, who were HIV sero-negative, did not participate in a similar survey
within the past 3 months, willing to complete the 18-month follow-up study and
provided informed consent were eligible for the study. Overall, 10 seeds having
diverse characteristics in terms of income, age, occupation, and cruising areas
were recruited. Interviews were conducted and blood samples were collected at the
STI clinic of the Jiangsu Provincial Center for Disease Control in Nanjing.
Follow-up visits and intervention
Written informed consent was obtained from all eligible participants before the
interview and blood collection in each visit. A face-to-face interview was
conducted using an interviewer administered structured questionnaire by trained
professionals in a separate and private room. At the end of the interview, specific
risk reduction counseling was provided to each participant on HIV and other STIs
(like increase in condom use/reducing number of sexual partner/avoiding high
risk sexual practices like UVI/UAI/risk of alcohol/drug use during sex etc.) by an
experienced counselor. Blood sample was collected next from each consenting
participant for HIV and syphilis testing. Post-test counseling was provided to each
subject when they returned to the clinic to collect HIV test results. Participants
were asked to come back to the designated clinic for the follow-up assessment
every 6 months. Any participant who was found to be HIV sero-positive at
baseline or during any follow-up visits were excluded from the cohort and were
linked to national free anti-retroviral treatment center.
After the participants were enrolled, the phone number, email address and QQ
number of them were collected. In addition, appointments regarding the next
visit/follow up were scheduled, while reminding cards which includes the date of
the next appointment were given them. Before two weeks of the scheduled follow
up time, a reminding message was send to every eligible HIV-negative participant
through texted message, email and QQ. If we did not get any responses from the
participant one week before the appointed time, an additional call was made to
If the participant missed the scheduled appointment, two more reminding
messages were delivered to him. Also, two calls were made to him one/two weeks
after the appointment. If needed, new appointments were. We continued to call
them 12 months after the initial interview, to ask whether they are willing to
attend the follow up. If they are not willing to attend the follow up scurvies again,
the detailed reasons for the recursion were asked and recorded.
Information was collected on socio-demographic factors (age/marital status/
official residency/education/income/whether student or not), behavioral
information (sexual orientation/history of unprotected vaginal/anal intercourse (UVI/
UAI) in the past six months). In our study, the size of the social network was
determined on the basis of the reported number of MSM known to the
participants (familiar with face/name/nickname/had his contact information and
could get in touch with him within the next month). Information was also
collected on self-reported history (yes/no) on any of the following symptoms:
burning sensation during urination/genital discharge/genital ulcer and if a
participant reported any one of the aforementioned symptoms, then he was
considered to be suffering from one or more STIs.
The participants were categorized into two groups: 1) consistent participation:
participants who attended the baseline survey and had returned for the follow-up
visits during 6th, 12th and 18th months to the designated STI clinic; 2)
nonconsistent participation/LTFU: participants who attended the baseline survey but
did not return for one or more follow up visits to the designated STI clinic during
the 18-month follow up period. The participants who were tested syphilis positive
but remained HIV negative were retained in the study.
Cruising areas/venues, HIV knowledge, and coverage of HIV preventive services
have been reported previously [29, 30]. Briefly, cruising areas/venues were
categorized as conventional venues (gay-bars/parks/massage-parlors/spas/saunas/
internet) and non-conventional venues (specifically, meeting partners on
campuses/introduced by friends, etc.).
HIV and syphilis testing
Five ml of venous blood sample was collected from each participant for HIV and
syphilis testing. Initially samples were screened for HIV using a rapid test (Acon
Biotech Co. Ltd). Samples positive at screening were confirmed by Western blot
(HIVBLOT 2.2, Genelabs Diagnostics, Singapore). Samples positive for Western
blot were considered as HIV sero-positive. Samples were also screened for syphilis
using the Rapid Plasma Reagin test (RPR: Beijing Wantai Biological Pharmacy
Enterprise Co. Ltd) and confirmed by the Treponema Pallidum Particle
Agglutination assay (TPPA: Livzon Group Reagent Factory). Syphilis positivity
was defined as current when both TPPA and RPR were positive. Participants
were asked to return to the designated lab to collect the test results.
The study protocol was approved by the Institutional Review Board of the Jiangsu
Provincial Center for Disease Control and Prevention (JSCDC). A signed
informed consent was obtained from each participant prior to the interview and
blood collection. Each participant had the discretion to freely decline or withdraw
from this study at any given point of time.
Data was double-entered using EpiData 3.0  and multiple logic checks were
used to ensure the data quality. SAS version 9.1  was used for all statistical
analyses. Descriptive analyses were conducted to determine the distribution of
socio-demographic characteristics, sexual behaviors, coverage of HIV/STI related
prevention services and compared between two groups of participant (those who
were LTFU & those who were retained in the study). In addition, to assess the
strength and direction of associations between LTFU and potential correlates,
both bivariate and multivariate regression analyses were performed [expressed in
Odds Ratio (OR) & 95% confidence interval (CI)]. The main outcome variable
was LTFU to BIs and predictor variables were age, marital status, official
residency, education, income, occupation, sexual orientation, UAI, UVI,
HIVrelated knowledge, coverage of HIV/STI preventive services, STI-related
symptoms, cruising venues, social network size, and syphilis positivity. Variables
with a p-value less than 0.3 in bivariate analyses were included in the multivariate
Demographics, behaviors, STI related symptoms and syphilis
Overall, 430 eligible MSM of Nanjing were screened at the baseline. Of these
subjects, 20 were HIV sero-positives at the baseline and were excluded from the
study, 410 HIV sero-negatives were invited to come to the designated clinic for
the follow-up visits at 6-month, 12-month and 18-month. Of the 410 participants,
189 (46.1%) came for all the three and 221 (53.9%) missed one or more follow-up
visits. 13 (6.9%) participants were sero-converted for HIV infection during the
follow-up period (Fig. 1).
Overall, the mean age was 28.38.6 years, 61.2% of the participants were of age
2029 years, 76.1% were never married, 59.5% were official residents of Nanjing,
68.3% attended college or above, 28.8% had a monthly income less than 1,000
Chinese Yuan (1 US Dollar56.12 Chinese Yuan), 29.5% were students, 57.6%
contacted their casual partner through the internet and 87.1% received HIV/STI
related prevention services during the past 1 year, 60.0% and 16.8% of them had
UAI and UVI, respectively, during the past 6 months preceding the baseline
survey. In addition, about 36.8% of the eligible participants reported that they
have social network size less than 10. Overall, 24.6% reported one or more STI
related symptoms and 11.2% were tested positive for syphilis infection.
Sociobehavioral characteristics of the participants who attended all the follow-up visits
were significantly different from those who were LTFU, as evident from the
nonoverlapping 95% CIs (Table 1).
Fig. 1. Flowchart of the Follow Up among MSM in Nanjing, China, 20082010.
Correlates of LTFU
Both crude and adjusted models revealed that age of the participant was a
significant predictor of LTFU (crude Odds Ratio (COR) of 0.93, 95% CI 0.90
0.95, adjusted Odds Ratio(AOR) of 0.90, 95% CI 0.860.94). In the unadjusted
model and after controlling all other variables, LTFU was more frequent among
MSM who were official residents of the other provinces than participants from
Nanjing (COR of 3.97, 95% CI 2.266.98, AOR of 2.49, 95% CI 1.324.71). Alike
the bivariate analysis, with reference to the those who completed education up to
college or above, the adjusted analysis also indicated that participants completing
education up to high school or below were more likely to be LTFU (COR of 1.65,
95% CI 1.082.53, AOR of 2.01, 95% CI 1.103.66). Size of the social network was
found to be significantly associated with LTFU in both crude (COR of 1.64, 95%
CI 1.092.47) and adjusted (AOR of 1.75, 95% CI 1.092.80) analyses.
Participants who were single/married, had bisexual sexual orientation and found
their partner at pub/disco/tearoom/club showed positive associations with LTFU
in the crude analyses but after adjusting for potential confounders, these
associations were no longer significant in the final model (Table 2).
Table 1. Demographic characteristics, behaviors, STI related symptoms and proportion of syphilis infection among HIV sero-negative MSM at baseline in
Nanjing, China, 2008(N5410).
Official resident (Hukou)
Coverage of HVI/STI
Unprotected anal sex
Unprotected vaginal sex
Official resident (Hukou)
Unprotected anal sex
Unprotected vaginal sex
Reporting any STI related symptoms
Coverage of HIV/STI prevention services
HIV related knowledge
Size of social network
OR5Odds ratio 95% CI595% Confidence interval.
To the best of our knowledge this is the first study analyzing the role of
sociodemographic and behavioral correlates of LTFU to BIs among MSM in China. In
this cohort of 410 HIV sero-negative MSM attending a STI clinic for BIs and HIV
testing in Nanjing, China during 2008, 221(53.9%) of the participants were LTFU
during the 18-month period. This burden of LTFU was much higher than that
observed in a 12-month follow-up study among MSM in Beijing (13.8%) during
2007 , in a cohort of MSM in Argentina (8.5%) during 20032004  and in
EXPLORE study where 14.2% of MSM enrolled from six US cities during 1999
2001 did not attend the 48-month follow-up visits . It was much lower than
that reported in the Smart Sex Quest study in US during 20022003, where
approximately 85% of MSM did not have the complete follow-up data . Even
though MSM form a high risk population for HIV infection, it seems that
perceived homophobia, ignorance about HIV, low self-perceived HIV risk and
insensitivity of health care providers continues to limit access and uptake of these
essential preventive services among this vulnerable population [1, 23, 36, 37].
In our study the chances of LTFU fell with age and participants who were
official residents of other provinces in China had higher odds of drop out. These
findings might be partially explained by Chinas massive and most extensive
internal migration of younger population. Young, rural-to-urban male migrants
have been recognized as the tipping point for AIDS epidemic in China . A
subgroup of these migrants is known as money boy who engage in same sex
behavior mostly in MSM communities for earning money. They are mostly
hidden, hard to reach population and vulnerable to HIV due to their high risk
sexual behaviors . A cross-sectional study among money boys in Shanghai,
China demonstrated that compared to general male migrants, money boys had
lower rate of condom use, more likely to have multiple partners, less
knowledgeable about HIV/AIDS and only half of them were aware about free HIV
counseling and testing . Thus, a peer-driven counseling and behavioral
intervention programs targeting these highly mobile and active MSM and their
clients might be helpful in promoting HIV testing and safer sex among this at-risk
Consistent with the previous studies [33, 35], the participants in our study who
completed the 18-month follow-up visit had higher level of education than who
were LTFU. This study also revealed a positive association between lower level of
education and LTFU which corroborated with a prior study among MSM in
Beijing . This might be due to the fact that MSM with less education are
relatively less aware of the available HIV prevention services and probability of
their participation being more affected by social stigma and discrimination might
be another contributing factor.
The current study also indicated that small social network size was positively
associated with LTFU. Prior research demonstrated a strong and consistent
association between network characteristics and MSM sexual behavior [42, 43].
Constant vigilance by peer group, numerous social interconnections at individual
and community levels, and peer driven counseling about the importance of HIV
prevention interventions in a large social network were likely to influence the
adherence to behavioral changes, counseling and HIV testing among MSM
compared to relatively smaller social network [42, 44].
The present study had several limitations. Due to the potential lack of
generalizability, extrapolation of results beyond the study sample was not
recommended and any such effort should be made with caution. Although care
was taken to avoid over-representation of participants linked with large social
networks, chances of such bias might not be completely ruled out. Behavioral
information was all self-reported and degree to which social desirability and
accuracy of memory influenced these responses, remained questionable. We could
not include too many variables in the final multivariate regression model, thus
chances of residual confounding might still be there.
To conclude, MSM of younger age, having lower level of education, residing
unofficially in Nanjing and being linked with smaller social network were all
associated with higher likelihood of being LTFU. Despite of study limitations and
scarcity of information on correlates of LTFU among MSM, we believe that the
findings of our study stressed the need to better understand the specific
characteristics of MSM who were vulnerable to LTFU so that a culturally
competent, peer-driven, sustainable, multilevel collaborative approach might be
designed for preventing or minimizing LTFU from such preventive efforts and
HIV testing among these at risk MSM.
This work was supported by National Natural Science Foundation of China
(Grant No. 81373125), Jiangsu Provincial Technologies Research Program
(BE2009685), Jiangsu Provinces Outstanding Medical Academic Leader Program
(RC2011086, 2011087), and NIH program (1D43TW009532-01).
Conceived and designed the experiments: XH JL XL HY GF CG RD. Performed
the experiments: XH JL XL HY GF CG. Analyzed the data: WT TM SM GF JZ.
Contributed reagents/materials/analysis tools: YZ TM SM. Wrote the paper: WT
TM SM CG RD.
1. (2011) Prevention and treatment of HIV and other sexually transmitted infections among men who have
sex with men and transgender people, Recommendations for a public health approach, 2011, World
Health Organization (WHO).
2. Granich RM , Gilks CF , Dye C , De Cock KM , Williams BG ( 2009 ) Universal voluntary HIV testing with immediate antiretroviral therapy as a strategy for elimination of HIV transmission: a mathematical model . The Lancet 373 : 48 - 57 .
3. 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 . The Lancet infectious diseases 13 : 955 - 963 .
4. Coates TJ , Richter L , Caceres C ( 2008 ) Behavioural strategies to reduce HIV transmission: how to make them work better . The Lancet 372 : 669 - 684 .
5 . ( 2013 ) Effective Behavioral interventions , HIV/AIDS, Prevention research, Programs, Centers for Disease Control and Prevention (CDC).
6 . ( 2001 ) HIV prevention needs and successes: a tale of three countries, An update on HIV prevention success in Senegal, Thailand and Uganda , UNAIDS.
7. Auerbach JD , Coates TJ ( 2000 ) HIV prevention research: accomplishments and challenges for the third decade of AIDS . American Journal of Public Health 90.
8. Stoneburner RL , Low-Beer D ( 2004 ) Population-level HIV declines and behavioral risk avoidance in Uganda . Science 304 : 714 - 718 .
9. Rojanapithayakorn W , Hanenberg R ( 1996 ) The 100% condom program in Thailand . Aids 10 : 1 - 8 .
10. Me da N , Ndoye I , Souleymane M , Ndiayee S , Niang C , et al. ( 1999 ) Low and stable HIV infection rates in Senegal: natural course of the epidemic or evidence for success of prevention? Aids 13 : 1397 - 1405 .
11. Denison JA , O9Reilly KR , Schmid GP , Kennedy CE , Sweat MD ( 2008 ) HIV voluntary counseling and testing and behavioral risk reduction in developing countries: a meta-analysis , 1990 - 2005 . AIDS and Behavior 12 : 363 - 373 .
12. Marks G , Crepaz N , Senterfitt JW , Janssen RS ( 2005 ) Meta-analysis of high-risk sexual behavior in persons aware and unaware they are infected with HIV in the United States: implications for HIV prevention programs . JAIDS Journal of Acquired Immune Deficiency Syndromes 39 : 446 - 453 .
13. ( 2012 ) UNAIDS World AIDS Day Report , 2012 .
14. Beyrer C , Baral SD , van Griensven F , Goodreau SM , Chariyalertsak S , et al. ( 2012 ) Global epidemiology of HIV infection in men who have sex with men . The Lancet 380 : 367 - 377 .
15. Dilley JW , Woods WJ , Loeb L , Nelson K , Sheon N , et al. ( 2007 ) Brief cognitive counseling with HIV testing to reduce sexual risk among men who have sex with men: results from a randomized controlled trial using paraprofessional counselors . JAIDS Journal of Acquired Immune Deficiency Syndromes 44 : 569 - 577 .
16. Koblin BA ( 2004 ) Effects of a behavioural intervention to reduce acquisition of HIV infection among men who have sex with men: the EXPLORE randomised controlled study . The Lancet 364 : 41 - 50 .
17. Herbst JH , Sherba RT , Crepaz N , DeLuca JB , Zohrabyan L , et al. ( 2005 ) A meta-analytic review of HIV behavioral interventions for reducing sexual risk behavior of men who have sex with men . JAIDS Journal of Acquired Immune Deficiency Syndromes 39 : 228 - 241 .
18. ( 2012 ) China AIDS Response Progress Report, Ministry of Health of the People's Republic of China , UNAIDS.
19. Choi K-H , Diehl E , Yaqi G , Qu S , Mandel J ( 2002 ) High HIV risk but inadequate prevention services for men in China who have sex with men: an ethnographic study . AIDS and Behavior 6 : 255 - 266 .
20. Gao L , Zhang L , Jin Q ( 2009 ) Meta-analysis: prevalence of HIV infection and syphilis among MSM in China . Sexually transmitted infections 85 : 354 - 358 .
21. Wu Z , Xu J , Liu E , Mao Y , Xiao Y , et al. ( 2013 ) HIV and syphilis prevalence among men who have sex with men: a cross-sectional survey of 61 cities in China . Clinical infectious diseases 57 : 298 - 309 .
22. Shang H , Xu J , Han X , Li JS , Arledge KC , et al. ( 2012 ) HIV prevention: Bring safe sex to China . Nature 485 : 576 - 577 .
23. Liu H , Yang H , Li X , Wang N , Liu H , et al. ( 2006 ) Men who have sex with men and human immunodeficiency virus/sexually transmitted disease control in China . Sexually transmitted diseases 33 : 68 - 76 .
24. Lau J , 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 .
25. Wong FY , Huang ZJ , Wang W , He N , Marzzurco J , et al. ( 2009 ) STIs and HIV among men having sex with men in China: a ticking time bomb ? AIDS Education & Prevention 21 : 430 - 446 .
26. Yun Gao M , Wang S ( 2007 ) Participatory communication and HIV/AIDS prevention in a Chinese marginalized (MSM) population . AIDS care 19 : 799 - 810 .
27. Zhang H , Wu Z , Zheng Y , Wang J , Zhu J , et al. ( 2010 ) A pilot intervention to increase condom use and HIV testing and counseling among men who have sex with men in Anhui, China . JAIDS Journal of Acquired Immune Deficiency Syndromes 53 : S88 - S92 .
28. Lau J , Lau M , Cheung A , Tsui H ( 2008 ) A randomized controlled study to evaluate the efficacy of an Internet-based intervention in reducing HIV risk behaviors among men who have sex with men in Hong Kong . AIDS care 20 : 820 - 828 .
29. Tang W , Huan X , Mahapatra T , Tang S , Li J , et al. ( 2013 ) Factors Associated with Unprotected Anal Intercourse Among Men Who Have Sex with Men: Results from a Respondent Driven Sampling Survey in Nanjing , China, 2008 . AIDS and Behavior : 1 - 8 .
30. Huan X , Tang W , Babu GR , Li J , Zhang M , et al. ( 2013 ) HIV risk-reduction counseling and testing on behavior change of MSM. PloS one 8: e69740 .
31. Lauritsen J , Bruus M ( 2003 ) EpiData (version 3). A comprehensive tool for validated entry and documentation of data Odense : EpiData Association 2004 .
32. Institute S ( 1996 ) The SAS system for Windows . SAS Institute Cary , NC.
33. Ruan Y , Jia Y , Zhang X , Liang H , Li Q , et al. ( 2009 ) Incidence of HIV-1, syphilis, hepatitis B, and hepatitis C virus infections and predictors associated with retention in a 12-month follow-up study among men who have sex with men in Beijing, China . JAIDS Journal of Acquired Immune Deficiency Syndromes 52 : 604 - 610 .
34. Segura M , Estani SS , Marone R , Bautista CT , Pando MA , et al. ( 2007 ) Buenos Aires cohort of men who have sex with men: prevalence, incidence, risk factors, and molecular genotyping of HIV type 1 . AIDS research and human retroviruses 23 : 1322 - 1329 .
35. Salyers Bull S , Lloyd L , Rietmeijer C , McFarlane M ( 2004 ) Recruitment and retention of an online sample for an HIV prevention intervention targeting men who have sex with men: the Smart Sex Quest Project . AIDS care 16 : 931 - 943 .
36. Song Y , Li X , Zhang L , Fang X , Lin X , et al. ( 2011 ) HIV-testing behavior among young migrant men who have sex with men (MSM) in Beijing, China . AIDS care 23 : 179 - 186 .
37. Choi KH , Lui H , Guo Y , Han L , Mandel JS ( 2006 ) Lack of HIV testing and awareness of HIV infection among men who have sex with men , Beijing, China. AIDS Education & Prevention 18 : 33 - 43 .
38. Anderson A , Qingsi Z , Hua X , Jianfeng B ( 2003 ) China's floating population and the potential for HIV transmission: a social-behavioural perspective . AIDS care 15 : 177 - 185 .
39. Wong FY , Huang ZJ , He N , Smith BD , Ding Y , et al. ( 2008 ) HIV risks among gay-and non-gayidentified migrant money boys in Shanghai, China . AIDS care 20 : 170 - 180 .
40. He N , Wong FY , Huang ZJ , Thompson EE , Fu C ( 2007 ) Substance use and HIV risks among male heterosexual and 'money boy'migrants in Shanghai, China . AIDS care 19 : 109 - 115 .
41. Liu H , Liu H , Cai Y , Rhodes AG , Hong F ( 2009 ) Money boys, HIV risks, and the associations between norms and safer sex: a respondent-driven sampling study in Shenzhen, China . AIDS and Behavior 13 : 652 - 662 .
42. Smith A , Grierson J , Wain D , Pitts M , Pattison P ( 2004 ) Associations between the sexual behaviour of men who have sex with men and the structure and composition of their social networks . Sexually Transmitted Infections 80 : 455 - 458 .
43. Choi K-H , Ning Z , Gregorich SE , Pan Q- c ( 2007 ) The influence of social and sexual networks in the spread of HIV and syphilis among men who have sex with men in Shanghai, China . JAIDS Journal of Acquired Immune Deficiency Syndromes 45 : 77 - 84 .
44. Hightow LB , Leone PA , MacDonald PD , McCoy SI , Sampson LA , et al. ( 2006 ) Men who have sex with men and women: a unique risk group for HIV transmission on North Carolina College campuses . Sexually transmitted diseases 33 : 585 - 593 .