Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Apr 28.
Published in final edited form as: AIDS Behav. 2012 May;16(4):858–871. doi: 10.1007/s10461-011-0119-4

Social Network and Other Correlates of HIV Testing: Findings from Male Sex Workers and Other MSM in Shanghai, China

Z Jennifer Huang 1, Na He 2,3, Eric J Nehl 4, Tony Zheng 5, Brian D Smith 6, Jin Zhang 7, Sarah McNabb 8, Frank Y Wong 9
PMCID: PMC8080277  NIHMSID: NIHMS1694452  PMID: 22223298

Abstract

Although the Chinese government provides free-of-charge voluntary HIV counseling and testing, HIV testing rates among men who have sex with men (MSM) are reported to be extremely low. This study examines the association of structural and psychosocial factors and social network characteristics with HIV testing behaviors among “money boys” and general MSM in Shanghai. Overall, 28.5% of “money boys” and 50.5% of general MSM had never tested for HIV despite high rates of reported HIV risk behaviors. Factors associated with not testing for HIV included: not knowing of a testing site, limited HIV knowledge, low perceived HIV risk, concern about HIV testing confidentiality, being a closeted gay, not using the Internet, and having a small social network or network with few members who had tested for HIV. Future efforts to promote HIV testing should focus on outreach to general MSM, confidentiality protection, decreasing the stigma of homosexuality, and encouraging peer education and support through the Internet and social networks.

Keywords: HIV testing, MSM, Money boys, Social network

Introduction

Consistent growth in the numbers of HIV and other STI (sexually transmitted infections) cases in heavily-populated China has resulted in the country playing an increasingly important role in determining the trajectory of these epidemics regionally and globally. Thus, the development of effective interventions in China is essential. China’s HIV epidemic is increasingly transmitted through sex and will grow from the current 740,000 cases [1, 2] to an estimated 1.2 million cases by 2015 [3], while the annual HIV infection rate has been rising steadily since the 1990s: increases of 30% between 1995 and 2000, 58% during 2001, 122% during 2003, and 45% during 2007 [4, 5]. As in many other Asian countries, men who have sex with men (MSM) represent an increasing proportion of new infections in China: 0.2% in 2001, 7.3% in 2005, 12.2% in 2007, and 32.5% in 2009 [5–8]. A large-scale national survey of 18,000 MSM conducted in 2008–2009 across 61 cities throughout China indicated an overall prevalence of 4.9% [9], with four cities reporting an HIV prevalence exceeding 10% [10]. Other regional and sentinel surveys reveal high HIV prevalence rates among MSM in many large metropolitan areas: 5.2% in Beijing in 2006 [11], 7–8% in Chongqing [12, 13], and 10.6% in Chengdu [14], while rising HIV prevalence rates among MSM were also reported in Nanjing [15], Shenyang [16], Shenzhen [17, 18], Jinan [19], and Jiangsu Province [20]. The increased number of reported cases among MSM may result from a combination of factors, including a higher disclosure rate that can be attributed to increasing cultural acceptance of homosexuality and the expanding size of sexual networks [4, 21] and increased presence of risk behaviors [2, 16, 22–26] with persistently high rates of unprotected anal intercourse (UAI) among MSM [10, 14, 27] in China. It is undeniable that China, like other Asian countries, is facing an expanding HIV epidemic among its 10 to 20 million MSM [6].

Recent articles [4, 10, 28] based on research projects in China have summarized the demographic and sexual risk profiles of MSM, emphasizing the unique cultural and demographic characteristics of MSM and the high prevalence of risky behaviors such as unprotected anal intercourse and multiple sex partners among the population. It is still unknown whether unrecognized infections among MSM are a driving force behind the high rate of HIV transmission [29]. Since the emergence of the HIV epidemic, client-initiated HIV testing and counseling [also known as voluntary counseling and testing (VCT)] has allowed millions of people globally to ascertain their HIV status and has been shown to be an efficacious form of primary prevention [30, 31]. Free VCT has also been integrated into China’s “Four Free and One Care” national policy to fight HIV/AIDS since 2003 [32]. As with many countries with a low HIV/AIDS prevalence rate among the general population, however, VCT rates for HIV in China are extremely low. Two studies—a 2006 study in Beijing [33] and a 2008 study in Jinan [29]—found that more than 80% of MSM had not tested for HIV in the past 12 months. Certain demographic patterns were also reported: MSM who were older (at least 35 years old), married, bisexual, or unsure about their sexual identity were more likely to forego HIV testing [29]. Other factors associated with not testing include HIV/AIDS knowledge, accessibility of testing sites, and perceived vulnerability. Choi et al. [33] found that the primary psychosocial barriers to testing were low perceived risk of contracting HIV/AIDS, fear of needles, and lack of knowledge about HIV treatment availability, while the primary structural barriers were lack of knowledge of a testing site and possibility of breach of confidentiality.

What is unclear is how the characteristics of social networks (e.g. network size and homogeneity) might be associated with the psychological and structural factors and ultimately influence HIV testing practices among Chinese MSM from diverse socio-demographic circles. In addition, most studies of HIV testing among Chinese MSM have treated these men as a homogeneous group, whereas our formative research documented several different types of MSM whose diverse demographic and socioeconomic profiles suggest that they may exhibit different patterns of HIV testing [6, 34, 35]. This study aims to describe and examine the association of social network characteristics with structural and psychosocial factors for HIV testing among general MSM and money boys (i.e., MSM who provide commercial sex) in China. We derive the social network characteristics from a respondent driven sampling (RDS)-based study conducted in Shanghai, China. The study is innovative as it is one of the first to explore these social network characteristics of money boys and MSM in China in the context of promoting HIV testing.

Methods

Data Sources

Respondent-driven sampling [13, 18, 36–38] (RDS) was used to recruit study subjects in Shanghai, China between July 2008 and January 2009. To be eligible, participants must have been at least 18 years old, been able to provide written and verbal consent in Mandarin, self-identified as male, and engaged in sex (oral, anal, or both) with another man in the last 12 months. Eight seeds were chosen based on occupation: four money boys and four general MSM (non-money boys). The money boy seeds were further categorized as recruiting clients through either venue- or Internet-based methods, and as self-identifying as either gay or non-gay. General MSM seeds were recruited based on migrant status (migrant vs. local resident) and age range (18–29 years old vs. 30+ years old) (Fig. 1). These sampling strata were adopted after extensive discussions with local community leaders in order to recruit a diverse and representative sample of MSM in Shanghai. A coupon tracking system was used to track the recruitment process, affirm a preexisting relationship to the recruiter, prevent recruitment overlaps, and assess non-response bias.

Fig. 1.

Fig. 1

Open in a new tab

RDS network chart and HIV testing status by seed characteristics, Shanghai, China 2009

Eligible respondents were verbally informed of the purpose and procedures of the study, sensitive nature of the questions, confidentiality parameters, compensation (US $30 for completing the quantitative study, free HIV and STI testing, free STI treatment, and referrals for other needed services), risks and benefits, and freedom to cease participation at any time. Study participants expressed a verbal understanding of these issues and signed Mandarin-language consent forms.

Since our past research [25, 34, 35, 39] indicated that participants from rural areas would not be familiar with computer technology, this study used a traditional paper- and-pencil survey/interview methodology. Participants were asked to complete an hour-long questionnaire with assistance from trained interviewers whose first language was Mandarin and who were familiar with the Shanghai money boy and MSM sub-cultures and slang. The Mandarin-language questionnaire underwent standard translation and back-translation (English–Chinese–English) using a cross-cultural adaptation process.

Participants who were willing to be tested for HIV received pre- and post-test counseling at a clinic of the Shanghai Municipal Center for Disease Control and Prevention (CDC). A sequential HIV screening strategy was used to prevent false positives. All participants who received a positive screen received secondary testing with two different test kits, followed by a western blot analysis if diagnosed positive in either test. Individuals confirmed as HIV-positive were reported to the Shanghai Municipal CDC’s Department of HIV/AIDS/STDs for further referrals. Migrants who were identified as HIV-positive had their status reported to their hometown CDC, though they could initially receive HIV treatment in Shanghai if eligible according to national guidelines.

Measures

HIV Testing Status

HIV testing status was based on participants’ responses to the question, “Have you ever been tested for HIV in your life?” Those who answered “yes” or were tested as part of the current study were defined as “tested” and asked to answer questions concerning the reasons why they received testing. Those who answered “no” were asked to identify the reasons for not testing where multiple responses were allowed. Voluntary HIV testing was offered after the subject completed the survey, so as to not affect the participant’s answer to this testing question.

Reasons for not being tested for HIV previously were grouped into two broad categories: (1) structural barriers, which, as defined in established studies on HIV testing [40, 41], are those such as the health care system and public health infrastructure that account for limited access to health care services and lack of HIV knowledge; and (2) psychosocial barriers, which are individual-level barriers to receiving HIV testing that are usually influenced by culture and one’s social environment (Fig. 2). Here we attribute limited HIV knowledge to the lack of structured intervention and prevention efforts in the public health system.

Fig. 2.

Fig. 2

Open in a new tab

A conceptual framework to guide analysis of HIV testing barrier

Psychosocial Correlates of HIV Testing

Based on past research [40], our measures for psychosocial factors include individual-level factors such as perceived risk for HIV, depression, fear of needles, fear of learning test results, and stigma which was measured by “closeted” versus openly gay/bisexual status. Perceived risk for HIV was based on responses to the question, “Do you think you could be HIV-positive?” An answer of “yes” was considered high perceived risk and an answer of “no” or “don’t know” was considered low perceived risk. Depression among study participants was analyzed using the short-form Center for Epidemiological Studies Scale (CES-D) [42], a 12-item questionnaire concerning depressive symptoms experienced in the past week. The CES-D has been shown to accurately screen for DSM-IV-diagnosed depression. Internal consistency was 0.83 for money boys and 0.85 for general MSM. A Sexual Attitudes Scale (SAS) score assessed participants’ attitudes about sexual permissiveness, sexual responsibility, sexual communion, sexual instrumentality, and sexual conventionality [43]. The SAS showed an internal consistency of 0.75 for money boys and 0.81 for general MSM. A Loss of Face Scale (LFS) score assessed social perceptions of “face”, addressing items related to social status, ethical behavior, social propriety, and self-discipline [44]. For example, participants were asked to respond to the statement, “I am more affected when someone criticizes me in public than when I am criticized in private.” A seven-point Likert scale (1 = strongly disagree, 7 = strongly agree) was used with all items scored in the direction of loss of face. The score had an internal consistency of 0.71 for money boys and 0.78 for general MSM.

Structural Correlates of HIV Testing

Measures included knowledge of an HIV testing site and relative knowledge and understanding of HIV/AIDS. Knowledge of a testing site was assessed based on responses to the question, “Where can you receive HIV testing?” Possible responses included (mostly public) hospitals, private clinics, the CDC, or “I don’t know.” Individuals were grouped based on whether or not they could correctly identify a testing site. To assess HIV knowledge and understanding, an “HIV Knowledge Score” was calculated based on responses to eight true-or-false statements concerning HIV infection and prevention. Points were scaled into three units of analysis: (1) no correct answers, (2) one to four correct answers, and (3) five or more correct answers. Other factors include concerns about confidentiality (“I am afraid that friends or family might discover my test results”, “I am afraid that the government might discover my test results”, etc.) and concerns about time and cost (“testing site hours are not convenient”, “I cannot afford testing”, etc.).

Sexual Risk Behaviors

Sexual risk behaviors include unprotected anal sex and bisexual behaviors (“have you ever had sex with a woman?”).

Sexual Identity and Social Support

The sexual orientation of each participant was assessed based on responses to the question, “What is your sexual orientation?” with possible designations of (1) openly gay or bisexual, (2) “closeted” gay or bisexual, and (3) heterosexual or other. The Lesbian, Gay, and Bisexual Identity Scale (LGBIS) [45] was used to further analyze a variety of factors related to gay identity, including intrapersonal dimensions of personal superiority, fear of judgment, homo-negativity, and confusion about identity, as well as extra-personal dimensions including disclosure of one’s gay identity and participation in gay community activities. Items were reverse-scored with a possible range of 1 to 7 (1 = strongly disagree, 7 = strongly agree). Internal consistency was 0.60 for money boys and 0.64 for general MSM. To assess social support variables, the Social Provisions Scale (SPS) [46] was applied based on the six provisions of guidance, reliable alliance, reassurance of worth, attachment, social integration, and opportunity for nurturance. Responses were scored and derived for a global support score indicating relative levels of social support. The SPS showed an internal consistency of 0.82 for money boys and 0.85 for general MSM.

RDS Social Network Characteristics

RDS social network characteristics include homophily index that quantifies the tendency among study participants to recruit others with characteristics similar to their own [47, 48]. Random mixing would produce homophily values that approximate zero, whereas a score of 1 would indicate exclusive in-group recruitment and a score of −1 would indicate exclusive out-group recruitment. We also looked at seed network size and clusters of HIV-tested participants in each network.

Socio-Demographic Covariates

Socio-demographic covariates included age, migrant status, education level (illiterate, or completed elementary school, middle school, high school, or college and above), marital status (never married, married, divorced/widowed, or cohabiting with a significant other), and monthly income (less than RMB1000, RMB1000–2999.9, RMB3000–4999.9, RMB5000 or more; US$1 = ~RMB6.83 at the time of the study). Migrant status was based on self-reported legal residency status or hukou, which classifies participants as being from Shanghai (non-migrants) or elsewhere (migrants).

Study participants were categorized as either money boys or general MSM for separate analysis. Individuals who selected “sex worker” as their occupation were identified as money boys, while participants who selected any other response were classified as general MSM.

Statistical Analysis

The distributions of the variables of interest were tabulated separately for money boys and general MSM. RDS-related information (network size, etc.) was re-formatted and entered into RDSTAT 6.0 software [36] in order to assess the social network characteristics of money boys and general MSM, and to provide weighted prevalence rates of HIV and STIs for each group. Univariate (Chi-square tests and their P values) and multivariate (adjusted odds ratios and their 95% confidence intervals) analyses in SAS software [49] were then used to assess associations between HIV-testing status and socio-economic and demographic variables, as well as structural and psychosocial variables. The reasons cited for having received HIV testing (and for not receiving testing) were ranked for money boys and general MSM. Finally, characteristics of participants’ network characteristics were denoted by a network chart created using NETDRAW software [50] and the average network sizes and homophily scores resulted from RDS analysis using RDSTAT software [36, 51].

Results

Overall, 39.4% (404 of 1,026 coupons) of coupons were returned, 39.2% among MBs and 39.7% among general MSM. Socio-demographic characteristics of participants are shown in Table 1. The average age of our sample was about 30 years old, though money boys were much younger than general MSM (24 years old vs. 35 years old, P < 0.001). Money boys were more likely to be migrants (97.5% vs. 62.3%, P < 0.001), self-identify as non-gay (13.5% vs. 3.5%, P < 0.01), have less education (14.6% had a college degree compared to 31.5% of general MSM, P < 0.001), higher incomes (52.3% vs. 34.3% had income >RMB3000 per month, P = 0.001), and were less likely to be married (5.5% vs. 24.3%, P < 0.001). The only variable that was not statistically different between the two groups was the percentage of participants who had ever had sex with a woman: 75.3% of money boys versus 74.6% of general MSM.

Table 1.

Socio-demographic characteristics of money boys and general MSM in Shanghai, China, 2009

Overall (n = 404) Money boys (n = 200) General MSM (n = 204) Significance
M (SD) M (SD) M (SD)
Current age 29.6 (10.4) 24.27 (4.98) 34.91 (11.6) F = 143.2, P < 0.001
Age at first MSM experience 19.9 (6.1) 18.87 (4.71) 20.85 (7.06) F = 10.87, P < 0.001
n (%) n (%) n (%)
Hukou (resident status)
 Shanghai (non-migrant) 82 (20.3) 5 (2.5) 77 (37.7) x2 (1) = 77.55, P < 0.001
 Other (migrant) 322 (79.7) 195 (97.5) 127 (62.3)
Sexual orientation
 Openly gay/bisexual 49 (12.1) 26 (13.2) 23 (11.1) x2 (2) = 14.18, P < 0.01
 “Closeted” gay/bisexual 321 (79.5) 147 (73.5) 174 (85.3)
 Other 34 (8.4) 27 (13.5) 7 (3.4)
Ethnicity
 Han 386 (96.0) 187 (93.5) 199 (98.5) x2 (1) = 6.61, P = 0.01
 Other ethnicity 16 (4.0) 13 (6.5) 3 (1.5)
Occupation
 Student 2 (0.8) 1 (0.6) 1 (1.2) x2 (5) = 204.50, P < 0.001
 White collar 23 (9.1) 1 (0.6) 22 (27.2)
 Blue collar 23 (9.1) 1 (0.6) 22 (27.2)
 Government official 2 (0.8) 0 (0.0) 2 (2.5)
 Self-employed 42 (16.5) 11 (6.3) 31 (39.7)
 Sex worker 162 (63.8) 162 (92.0) 0 (0.0)
Education
 Illiterate or primary school 22 (5.5) 8 (4.0) 14 (6.9) x2 (3) = 19.58, P < 0.001
 Middle school 126 (31.3) 72 (36.2) 54 (26.6)
 High School or equivalent 161 (40.0) 90 (45.2) 71 (35.0)
 College or above 93 (23.1) 29 (14.6) 64 (31.5)
Marital status
 Never married 301 (75.1) 181 (91.0) 120 (59.4) x2 (2) = 53.31, P < .001
 Married/cohabiting 60 (15.0) 11 (5.5) 49 (24.3)
 Divorced/widowed 40 (10.0) 7 (3.5) 33 (16.3)
Monthly income
 < 1,000 Yuan 25 (6.2) 6 (3.0) 19 (9.3) x2 (3) = 17.18, P = 0.001
 1,000 Yuan–2,999 Yuan 204 (50.6) 89 (44.7) 115 (56.4)
 3,000 Yuan–4,999 Yuan 115 (28.5) 71 (35.7) 44 (21.6)
 ≥5,000 Yuan 59 (14.6) 33 (16.6) 26 (12.7)
Had sex with woman in lifetime
 Yes 299 (74.9) 149 (75.3) 150 (74.6) x2 (1) = 0.021, P = 0.885
 No 100 (25.1) 49 (24.7) 51 (25.4)
Open in a new tab

Note n varies based on missing responses

HIV Testing and RDS Network Characteristics

A total of 200 money boys and 204 general MSM were recruited in one to seven waves of RDS recruitment with eight seeds (Fig. 1). The network sizes of each seed varied from 13 recruits to 128 recruits. The high homophily score of 0.967 for money boys and 0.955 for general MSM indicates that the two groups, money boys and general MSM, were recruited exclusively in this study. Regarding affiliation among the participants, a figure of 0.4 is commonly used as the cutoff point to indicate strong affiliation. While most participants without college degrees were likely to recruit across education categories, those with college degrees were more likely to stay within their own education level, with a homophily score of 0.4. Money boys and general MSM of all income levels (homophily = −0.22–0.13) and marital status (homophily between 0.06 and 0.3) tended to mingle with individuals from different economic backgrounds and marital status. Those who indicated that they had tested for HIV were more likely to have been recruited by others who also tested for HIV (homophily = 0.2) than by those who indicated that they had not tested for HIV (homophily = 0.02).

Having tested for HIV was significantly associated with being a money boy (71.5% vs. 28.5%, P < 0.001). Average network size (adjusted for weights of money boys and general MSM) was 6.2 for those who tested for HIV and 4.8 for those who had never tested. Further analysis on clustering of HIV testing among recruits of each of the eight initial seeds (Fig. 1) showed that Internet-based money boy seeds had networks characterized by the highest proportion of HIV-tested participants (81 to 83%), while the networks of young migrants and older local resident MSM had the lowest testing rates (38 to 40%).

HIV Testing and Sexual Identity and Social Support

Sexual identity played an important role in HIV testing status. Overall, 80% (n = 321) of MSM reported that they were “closeted” gay, meaning that they did not disclose their gay or bisexual identity to friends and family (Table 2), compared to 12% (n = 49) open gay or bisexual, and 8% (n = 34) who were heterosexual but had an MSM experience or were not sure whether they were heterosexual or homosexual. Those who were “closeted” gay or bisexual were mostly likely to not having HIV testing (43.9%) compared to opened-gay/bisexual/other (43.9% vs. 22.9%, P = 0.0005). This trend repeated among money boys (32.7% vs. 17.7%, P = 0.03) and general MSM (53.5% vs. 33.13%, P = 0.042).

Table 2.

Correlates for percentage of money boys and general MSM who never had HIV testing in Shanghai, China, 2009

Never had HIV testing (before this study) Overall (n = 404) Money boys (n = 200) General MSM (n = 204)
n %/M (SD) n %/M (SD) n %/M (SD)
Demographic and socioeconomic traits
 Type of MSM
  Money boy 200 28.5%
  General MSM 204 50.5%
P < 0.001
 Hukou (residency status)
  Shanghai (non-migrant) 322 35.4% 195 28.2% 127 46.5%
  Other (migrant) 82 56.1% 5 40% 77 57.1%
P = 0.0016 P = 0.6246 P = 0.1389
 Age group
  <23 years 99 44.4% 76 36.8% 23 69.6%
  23–35 years 204 30.4% 117 23.9% 87 39.1%
  >35 years 101 53.5% 7 14.3% 94 56.4%
P = 0.0006 P = 0.1060 P = 0.0101
 Marital status
  Never married 301 36.5% 181 27.6% 120 50.0%
  Other (currently or previously married) 100 48.0% 18 33.3% 82 51.2%
P = 0.0422 P = 0.6075 P = 0.8648
 Education
  Middle school or less 148 38.5% 80 30.0% 68 48.5%
  High school equivalent 161 39.1% 90 28.9% 71 52.1%
  College or above 93 43.0% 29 24.1% 64 51.6%
P = 0.7662 P = 0.8341 P = 0.9031
 Monthly income
  <1,000 Yuan 25 44.0% 6 33.3% 19 47.4%
  1,000 Yuan–2,999 Yuan 204 41.2% 89 31.5% 115 48.7%
  3,000 Yuan–4,999 Yuan 115 40.0% 71 28.2% 44 59.1%
  ≥5,000 Yuan 59 32.2% 33 21.2% 26 46.2%
P = 0.6219 P = 0.7329 P = 0.6325
Sexual risks
 Had sex with woman in lifetime
  Yes 303 41.9% 140 30.7% 163 51.5%
  No 101 32.7% 60 23.3% 41 60.9%
P = 0.100 P = 0.2893 P = 0.5522
 Ever had unprotected anal sex?
  Yes 96 46.9% 50 34.0% 46 60.9%
  No 308 37.3% 150 26.7% 158 80.7%
P = 0.0953 P = 0.3198 P = 0.1096
Structural correlates
 Knowledge of testing site
  Know of a site 355 34.7% 182 24.7% 173 45.1%
  Did not know of a site 49 75.5% 18 66.7% 31 80.7%
P < 0.001 P = 0.002 P = 0.003
 Mean HIV knowledge score (had HIV testing: no HIV testing) 4.2(1.5):3.6 (1.6) 4.1(0.1):3.9(0.2) 4.2(0.2):3.5(0.2)
P = 0.008 P = 0.2869 P = 0.0016
Psychosocial
 Perceived risk for HIV
  Somewhat likely/very likely 92 37.0% 54 31.5% 38 44.7%
  Not very likely 312 40.4% 146 27.4% 166 51.8%
P = 0.5546 P = 0.57 P = 0.4317
 Depression (CESD ≥ 16)
  Yes 169 34.9% 108 25.9% 61 50.8%
  No 235 43.0% 92 31.5% 143 50.4%
P = 0.1019 P = 0.3823 P = 0.9510
Sexual identity and social support
 Sexual identity
  Openly gay/bisexual/other 83 22.9% 53 17.7% 30 33.3%
  “Closeted” gay/bisexual 321 43.9% 147 32.7% 174 53.5%
P = 0.0005 P = 0.0302 P = 0.0418
 Lesbian, gay, and bisexual identity scale score (had HIV testing: no testing) 3.9 (0.7): 4.1 (0.7) 4.0 (0.1):4.1 (0.1) 3.9 (0.1):4.0 (0.1)
P = 0.0615 P = 0.1100 P = 0.1493
 Sexual attitudes score (had HIV testing: no testing) 73.5 (17.9): 71.5 (17.3) 76.3 (1.4):73.7 (1.9) 69.7 (1.8): 70.3 (1.8)
P = 0.2594 P = 0.3120 P = 0.7999
 Social provision scores (had HIV testing: no testing) 68.7 (6.9):68.3 (7.0) 67.8 (0.6):68.2 (0.8) 69.9 (0.7):68.4 (0.7)
P = 0.5793 P = 0.7224 P = 0.1251
Open in a new tab

The bold highlighted results were statistically significant with Chi-square P value <0/05

Scores for the sexuality-related LGBIS and SAS did not differ for those who had tested for HIV versus those who had not. Similarly, the SPS score did not differ based on testing status (Table 2).

Structural and Psychosocial Determinants for HIV Testing

Not knowing of a testing site was one of two significant factors associated with never having tested among all study participants (75.5% vs. 34.7%, P < 0.001), whether money boys (66.7% vs. 24.7%, P < 0.0002) or general MSM (80.7% vs. 45.1%, P = 0.003). Lack of HIV knowledge was the other significant factor for all participants and both sub-groups. Lower mean scores for HIV knowledge were significantly associated with HIV testing status for all study subject (4.2 vs. 3.6, P = 0.008) (Table 2), and general MSM (4.2:3.5, P = 0.0016), but not for money boys whose overall mean score was higher (4.1 vs. 3.9, P = 0.28).

Neither of the psychosocial variables analyzed in this study, including perceived risk for HIV and depression, was significantly associated with HIV testing. However, an effect modification of being a money boy or general MSM may exist: those money boys who perceived themselves at high-risk were more likely NOT to have tested for HIV (31.5% vs. 27.4%) while general MSM who perceived themselves at high-risk were more likely to seek HIV testing (44.7% vs. 51.8% had not tested for HIV), even though these differences were not significant due to small sample sizes in each group. On the other hand, a higher proportion of money boys than general MSM disclosed more severe depressive symptoms (CESD ≥ 16 for 54% of money boys vs. 30% of general MSM, P = 0.001).

HIV Testing and Sexual Risks

The two indicators of risky sexual behaviors were not associated with HIV testing status. However, similar offset findings among money boys versus general MSM were intriguing. In particular, those money boys who had engaged in unprotected anal sex were more likely than those who had not to forego HIV testing (34% vs. 26.7%, P = 0.3), whereas among general MSM, those who had engaged in unprotected anal sex were less likely than those who had not to forego HIV testing (60% vs. 80.7%, P = 0.1). In general, three-quarters of the MSM had sex with women, and one-quarter admitted that they had engaged in unprotected anal sex.

HIV Testing and Socio-Demographic Characteristics

Finally, significant demographic and socio-economic characteristics that correlated with HIV testing status were age and hukou (residency status) (Table 2). Those who were 23 to 35 years old (69.6%) were more likely to have tested for HIV compared to younger (55.6%) and older (46.5%) (P = 0.00016) participants. Migrants were more likely to have tested, but since hukou was highly correlated with money boy or general MSM status, and since money boys were more likely to be tested for HIV, hukou could be a confounding variable. Those who ever married were more likely to lack HIV testing compared to those who never married (48.0% vs. 36.5%, P = 0.042). The factors not significantly associated with HIV testing include education level and income level.

Multivariable Analysis of Factors Associated with HIV Testing

We conducted separate multi-variable analysis to look at multiple correlates of HIV testing among money boys and general MSM (Table 3). Not all the factors were included in the final model due to high correlations among the factors. Factors entered the final models reflected the different domains of the conceptual model and results of bi-variable analysis: “engaged in unprotected anal sex”, “knowledge of testing site (before this study)”, “HIV knowledge score”, “education”, and “monthly income”. For both money boys and general MSM, knowing of a testing site became a significant factor after adjusting for other factors. Comparing those who lacked knowledge of a testing site to those who knew of a site, the risk of not having tested for HIV was 6.5 times greater (95% CI: 2.2–18.9) among money boys and 5.2 times greater (95% CI: 1.9–14.2) among general MSM. Knowledge of HIV was positively correlated with having tested for HIV among general MSM (AOR = 0.7, 95% CI: 0.6–0.9) but not among money boys. Other factors did not contribute to explaining HIV testing status.

Table 3.

Adjusted odds ratio for NOT having tested for HIV among money boys and general MSM, Shanghai, China, 2009

AOR 95% CI Money boys General MSM
AOR 95% CI AOR 95% CI
Engaged in unprotected anal sex
 Yes 1.5 (0.9–2.5) 1.4 (0.7–3.0) 1.9 (0.9–4.0)
 No Ref Ref Ref Ref Ref Ref
Knowledge of testing site
 Know of a site Ref Ref Ref Ref Ref Ref
 Did not know of a site 5.5 (2.7–11.3) 6.5 (2.2–18.9) 5.2 (1.9–14.2)
Perceived risk for HIV
 Somewhat likely/very likely 0.9 (0.6–1.6) 1.4 (0.6–2.8) 0.7 (0.3–1.6)
 Other (not likely, don’t know) Ref Ref Ref Ref Ref Ref
Knowledge score 0.8 (0.7–0.9) 0.9 (0.7–1.2) 0.7 (0.6–0.9)
Education
 Middle school or below 0.7 (0.4–1.2) 1.0 (0.4–2.8) 1.1 (0.5–2.4)
 High school 0.6 (0.3–1.0) 1.1 (0.4–3.1) 0.7 (0.3–1.7)
 College or above Ref Ref Ref Ref Ref Ref
Monthly income
 <1,000 Yuan 1.5 (0.5–4.4) 1.7 (0.2–12.8) 0.8 (0.2–3.2)
 1,000 Yuan < 2,999 Yuan 1.7 (0.8–3.2) 2.0 (0.7–5.5) 0.9 (0.3–2.5)
 3,000 Yuan–4,999 Yuan 1.6 (0.8–3.2) 1.6 (0.6–4.7) 1.9 (0.7–5.4)
 ≥5,000 Yuan Ref Ref Ref Ref Ref Ref
Open in a new tab

The bold highlighted results were statistically significant with 95% confidence interval not cross 1 and P value <0/05. All variable listed were in the model

Ranking of Reasons for Testing and Not Testing for HIV

Finally, other than the scales for each domain associated with HIV testing, we also asked and ranked the self-reported reasons for having been tested or never having been tested (Table 4). Among money boys who had not tested for HIV, the three most common reasons cited for never having tested for HIV were “I did not know where to be tested” (38.6%), “I did not know I should be tested” (36.84%), and “I am not worried about HIV” (33.33%). These responses were also the three most common reasons that general MSM gave for never having been tested for HIV, though they were ranked in a different order. The majority of general MSM identified “I am not worried about HIV” (51.46%) as the main reason, while “I did not know a test site” (27.18%) and “I did not know I should be tested” (25.24%) each accounted for about a quarter of the sample.

Table 4.

Rank of reasons for having and not having tested for HIV among money boys and general MSM, Shanghai, China, 2009

Money boys (n, %) Rank General MSM (n, %) Rank
Reasons for not testing (n = 160)
 Structural barriers
  Did not have time for testing 5 8.8 15 14.6 5
  Testing site hours not convenient 3 5.3 8 7.8
  Could not afford testing 3 5.3 4 3.9
  Did not know where to go 22 38.6 1 28 27.2 2
  Did not know he should get tested 21 36.8 2 26 25.2 3
 Confidentiality issues
  Afraid that friends/family might discover results 11 19.3 5 12 11.7
  Afraid that government might see results 2 3.5 6 5.8
  Did not want to be seen at testing site 9 15.8 14 13.6
 Psychosocial risks-Low perceived risks
  Was not worried about HIV 19 33.3 3 53 51.5 1
  Did not have risky sex 13 22.8 4 16 15.5 4
  Had sex with only one partner 4 7.0 14 13.6
 Other psychosocial barriers
  Afraid that results might be positive 11 19.3 5 12 11.7
  Does not like needles 8 14.0 10 9.7
Reasons for testing (n = 244)
 Perceived risks
  Was having sex with a new partner 45 31.5 4 34 33.7 3
  Had unprotected sex 66 46.2 2 37 36.6 2
  Was asked by partner/boyfriend 23 16.1 5 25 24.8 4
  Had HIV-positive partner/boyfriend 3 2.1 0 0.0
  Afraid that he contracted HIV 23 16.1 5 10 9.9
 Outreach/regulation
  Was asked by health department 6 4.2 0 0.0
  Was part of a research study 46 32.2 3 15 14.9 5
  Doctor recommended 10 7.0 3 3.0
  Required for employment/military service 9 6.3 0 0.0
  Required for medical/surgical procedure 1 0.7 0 0.0
  Donated blood/plasma 9 6.3 8 7.9
  Other 69 48.3 1 49 49.0 1
Open in a new tab

Among money boys who had tested for HIV, unspecified reasons or “other” (48.25%) and “had unprotected sex” (46.15%) were the most common reasons identified for doing so, whereas “being part of a research study” was the third most common reason for testing (32.2%). Among general MSM who had tested for HIV, the most common reasons for doing so were “had unprotected sex” (36.63%) and “having sex with a new partner” (33.66%).

Discussion

The present study is one of the first to explore the association between social network characteristics and HIV testing behaviors and to systematically examine psychosocial and structural barriers to HIV testing among two types of MSM in China. Key findings from this study will provide critical insights for helping to chart a new course in battling the rising HIV infection rate among MSM in China.

Overcome Structural Barriers I: Outreach Programs Targeting General MSM

Both money boys and general MSM in Shanghai were at high-risk for HIV infection, even though they perceived themselves to be low-risk. Among the 404 MSM who completed the survey, 23.7% (n = 96) reported having unprotected anal sex (25% of money boys, 22.5% of general MSM), and most (80%) also had sex with women. Despite the prevalence of high-risk behaviors in both groups, we found that money boys were much more likely than general MSM to have tested for HIV (71% vs. 28%) when their sexual risk behaviors were similar. The testing rate among money boys was higher than it was in our preliminary study conducted three years ago in Shanghai [34]. This finding is consistent with results from studies in other parts of China [52], likely reflecting either the effectiveness of extensive governmental and academic outreach efforts targeting sex workers in recent years, or money boys’ ability to benefit from prevention programs that target either sex workers or MSM. Indeed, compared to MSM in Shanghai who were not sex workers, the money boys in our study had better knowledge of HIV, held a higher perceived risk for HIV, and were more likely to know of a testing site despite the barriers they faced as rural-to-urban migrants (97% of money boys were migrants) with lower levels of education.

Among money boys, HIV knowledge was not significantly correlated with testing, probably due to their high average knowledge score, while HIV knowledge was significantly correlated with HIV testing among general MSM, even after adjusting for sexual risks, perceived risk, and demographics. These results highlight the importance of scaling up structured interventions among general MSM such as basic HIV education.

Knowing of an HIV testing site and having better knowledge of HIV were also strong predictors for having tested for HIV among both money boys and general MSM, thus echoing some of Choi’s 2006 [33] results, in which 28% of MSM in Beijing reported having been tested for HIV and the primary barriers to testing were low perceived risk of HIV infection and not knowing of a testing site. Among general MSM, the most common reasons cited for not having tested for HIV were “I am not worried about HIV infection”, followed by “I do not know of a testing site” and “I do not know I should get tested.” Among money boys, the most common reasons for not having tested for HIV were “I do not know where to go for testing”, “I do not know I should get tested”, and “I am not worried about HIV.” Overall, low perceived risk (“I did not know I should get tested, and “I am not worried about HIV”) and not knowing of a testing site were top reasons cited for not being tested for HIV.

Clearly, future studies must focus on addressing the existing challenges to disseminating knowledge about HIV risks and the health services available to all MSM in Shanghai. This will likely be especially challenging when targeting the general MSM population, since these men constitute a larger and more heterogeneous population than money age in terms of age and socio-economic and marital status. Some general conclusions about the effect of age and socio-economic mix on HIV: the social networks of the younger migrant and older local resident MSM seeds had low HIV testing rates (40 and 38%) compared to the networks of older migrant and younger local resident MSM seeds (50 and 77%).

Overcome Structural Barriers II: Improving Confidence in HIV Testing Confidentiality

Secondly, our data showed that concern about the confidentiality of HIV testing is a barrier for MSM to undergo routine testing. Three cited reasons related to such concerns were “I am afraid that friends or family might discover my test results”, “I am afraid that the government might see my test results”, and “I did not want to be seen at a testing site.” Those who cited at least one of these concerns accounted for 11% of money boys and 16% of general MSM. Since most MSM and money boys were married and “closeted” gay, concerns about HIV testing confidentiality could be particularly strong among these populations. Future efforts to promote HIV testing among MSM in China will need to address policies regarding testing confidentiality.

Overcome Psychosocial Barriers: Weakening Social Taboos and Stigma

Our study also found a strong association between being a “closeted” gay/bisexual versus openly gay (43.9% vs. 26.5%, P = 0.0016), and ever married (48% vs. 36.5%) versus never married, with not testing for HIV. This finding highlights the low HIV testing rate among MSM who consider homosexuality or foregoing marriage to be social taboos and fear being stigmatized. Most MSM in our study identified themselves as “closeted” gay or bisexual (85% of general MSM and 73% of money boys) and had sex with women (79% of general MSM and 70% of money boys), and many were currently or formerly married to women (40% of general MSM and 9% of money boys). These data support previous findings that a significant proportion of MSM in Asia also have sex with women [53] and highlight the importance of de-stigmatizing homosexuality as part of the battle against HIV infection in China.

Even though the Chinese government and society are becoming increasingly open to accepting the rights of homosexuals as evidenced by the government’s recent open statement on state-controlled television [54], many MSM still fail to disclose their sexual orientation to friends and family. In particular, these men were particularly reluctant to disclose their gay identity to their parents. Qualitative interviews from the study reveal that many MSM would “get married and have a child with a woman because a gay relationship has no future and is not stable” or would “go home and marry a woman, then… come back to city and live my life.” These comments reflect the cultural emphasis on producing offspring in order to continue the family name. Among those who either are married to a woman or remain closeted, many may not test for HIV due to concerns about confidentiality (concern that their spouse or a family member will discover their sexual orientation). Due to the dual lives they lead, others may have less exposure to HIV-related knowledge and thus underestimate the risk they face of contracting HIV. Future studies can help to elucidate the role of social taboos as a barrier to HIV testing among MSM.

Internet Use and Larger Social Networks May Help to Promote HIV Testing

Finally, this study highlights the importance of Internet use in promoting HIV testing. Our analysis of social networks showed that, regardless of sexual orientation, a higher concentration of participants recruited through the two Internet-based money boy seeds (80 and 81%) compared to the two venue-based money boy seeds (60 and 65%) had tested for HIV. Online sex-seeking among MSM in China has surged in recent years, especially among younger generations [55]. The Internet thus may provide an efficient means to disseminate information about HIV and HIV testing. Several studies in the US have successfully promoted STI/HIV testing using interactive Internet games and activities that educate adolescents and young adults about sexual health risks [56–58]. Findings from our study provide further evidence of the Internet’s importance in promoting HIV testing.

MSM who were members of larger and more homogenous social networks were more likely to test for HIV. Those who tested for HIV, however, tended to have larger networks (an average network size of 6.2 vs. 4.8) and were more likely to be part of a more homogenous network (a homophily score of 0.24 vs. 0.02). Our study intentionally discouraged MB and MSM seeds to recruit anyone who did not have the same status, while preliminary network analysis showed that the members of MB or MSM social networks were quite diverse within the group in terms of education levels (though those with a college degree were more likely to be exclusive than others), income levels, and marital status. Recruitment ability among seeds was unbalanced: one young general MSM seed recruited only five participants, while an older migrant general MSM recruited 128 participants. This imbalance might be the result of social network size for participants with particular characteristics, or willingness to participate in a study like ours. Future studies are needed to examine the characteristics of social networks in greater depth in order to explore their impact on risk behaviors and information dissemination.

There are several limitations to the study. Since study subjects were recruited from Shanghai, our results may not apply to money boys and general MSM in other areas of China. For the question asking reasons for having tested for HIV, half of respondents reported “other” rather than one of the specific reasons listed. Future studies should provide better line-item choices for this variable. Finally, social network analysis was limited to RDS recruitment questions (size of the network, etc.) and linkage among study participants only.

Conclusions

Our study indicates that HIV testing rates remain very low among general MSM in Shanghai, while testing rates have increased rapidly among Shanghai’s male sex workers. Lack of HIV knowledge and not knowing of an HIV testing site, as well as low perceived risk of HIV, were still the top reasons cited for not testing for HIV. HIV/AIDS education and testing promotion should focus particularly on young migrants and older local resident MSM in Shanghai. Strategies and policies that can decrease the stigma of homosexuality, promote sexual health education using the Internet, and build stronger communities among Chinese MSM may be effective in helping the MSM community to better utilize the free, government-provided HIV testing.

Acknowledgments

Preparation of this article was supported in part by grants from the National Institutes of Health (R01HD046354; PI: Wong) and the Emory Center for AIDS Research (P30 AI050409; Nehl and Wong).

Contributor Information

Z. Jennifer Huang, Department of International Health, Georgetown University School of Nursing and Health Studies, 3700 Reservoir Road NW, St. Mary’s Hall 224, Box 571107, Washington DC, WA 20057, USA.

Na He, Department of Epidemiology, School of Public Health, Fudan University, Shanghai, China; Emory University’s Rollins School of Public Health, Atlanta, GA, USA.

Eric J. Nehl, Emory University’s Rollins School of Public Health, Atlanta, GA, USA

Tony Zheng, Shanghai Piaoxue, Shanghai, China.

Brian D. Smith, Department of International Health, Georgetown University School of Nursing and Health Studies, 3700 Reservoir Road NW, St. Mary’s Hall 224, Box 571107, Washington DC, WA 20057, USA

Jin Zhang, Department of International Health, Georgetown University School of Nursing and Health Studies, 3700 Reservoir Road NW, St. Mary’s Hall 224, Box 571107, Washington DC, WA 20057, USA.

Sarah McNabb, Department of International Health, Georgetown University School of Nursing and Health Studies, 3700 Reservoir Road NW, St. Mary’s Hall 224, Box 571107, Washington DC, WA 20057, USA.

Frank Y. Wong, Emory University’s Rollins School of Public Health, Atlanta, GA, USA

References

  • 1.Wu Z, Wang Y. Introduction: China meets new AIDS challenges. J Acquir Immune Defic Syndr. 2010;53(Suppl 1):S1–3. [DOI] [PubMed] [Google Scholar]
  • 2.He Q, Wang Y, Lin P, Raymond HF, Li Y, Yang F, et al. High prevalence of risk behaviour concurrent with links to other high-risk populations: a potentially explosive HIV epidemic among men who have sex with men in Guangzhou, China. Sex Transm Infect. 2009;85(5):383–90. [DOI] [PubMed] [Google Scholar]
  • 3.Juan S China to have 1.2 million HIV-positive people by 2015 [Internet]. China Daily. 2010. http://www.chinadaily.com.cn/china/2010-12/04/content_11652055.htm. [Google Scholar]
  • 4.Guo Y, Li X, Stanton B. HIV-related behavioral studies of men who have sex with men in China: a systematic review and recommendations for future research. AIDS Behav. 2011;15(3):521–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Wang L HIV prevalence among populations at risk, using sentinel surveillance data from 1995 to 2009 in China. Chung-Hua Liu Hsing Ping Hsueh Tsa Chih. 2011;32(1):20. [PubMed] [Google Scholar]
  • 6.USAID. Health policy initiative. The value of investing in MSM programs in the Asia-Pacific region [Internet]. 2007. http://www.msmasia.org/The_Value_of_Investing_in_MSM_Programs_in_the_Asia-Pacific_Region.pdf.
  • 7.USAID. The value of investing in MSM programs in the Asia-Pacific region [Internet]. 2006. http://www.msmasia.org/The_Value_of_Investing_in_MSM_Programs_in_the_Asia-Pacific_Region.pdf.
  • 8.State Council AIDS Working Committee Office and UN Theme Group on HIV/AIDS in China. A Joint Assessment of HIV/AIDS Prevention, Treatment and Care in China [Internet]. 2004. www.chinaaids.cn/worknet/download/2004/report2004en.pdf. Accessed 30 Dec 2011.
  • 9.China Ministry of Health. China estimation report-en.pdf [Internet]. 2011. http://www.unaids.org.cn/download/2009%20China%20Estimation%20Report-En.pdf. Accessed 3 Aug 2011.
  • 10.Lau JT, Lin C, Hao C, Wu X, Gu J. Public health challenges of the emerging HIV epidemic among men who have sex with men in China. Public Health. 2011;125(5):260–5. [DOI] [PubMed] [Google Scholar]
  • 11.Ma X, Zhang Q, He X, Sun W, Yue H, Chen S, et al. 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. 2007;45(5):581–7. [DOI] [PubMed] [Google Scholar]
  • 12.Han M, Feng LG, Jiang Y, Shen S, Ling H, Ding XB, et al. Surveillance on HIV-1 incidence among men who have sex with men in Chongqing, China, 2006–2008. Zhonghua Liu Xing Bing Xue Za Zhi. 2009;30(9):878–81. [PubMed] [Google Scholar]
  • 13.Ouyang L, Feng LG, Ding XB, Zhao JK, Xu J, Han M, et al. A respondent-driven sampling survey on HIV and risk factors among men who have sex with men in Chongqing. Zhonghua Liu Xing Bing Xue Za Zhi. 2009;30(10):1001–4. [PubMed] [Google Scholar]
  • 14.Feng Y, Wu Z, Detels R, Qin G, Liu L, Wang X, et al. HIV/STD prevalence among men who have sex with men in Chengdu, China and associated risk factors for HIV infection. J Acquir Immune Defic Syndr. 2010;53(Suppl 1):S74–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Yang H, Hao C, Huan X, Yan H, Guan W, Xu X, et al. HIV incidence and associated factors in a cohort of men who have sex with men in Nanjing, China. Sex Transm Dis. 2010;37(4):208–13. [DOI] [PubMed] [Google Scholar]
  • 16.Xu JJ, Zhang M, Brown K, Reilly K, Wang H, Hu Q, et al. Syphilis and HIV seroconversion among a 12-month prospective cohort of men who have sex with men in Shenyang, China. Sex Transm Dis. 2010;37(7):432–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Hong FC, Zhou H, Cai YM, Pan P, Feng TJ, Liu XL, et al. Prevalence of syphilis and HIV infections among men who have sex with men from different settings in Shenzhen, China: implications for HIV/STD surveillance. Sex Transm Infect. 2009;85 (1):42–4. [DOI] [PubMed] [Google Scholar]
  • 18.Liu H, Liu H, Cai Y, Rhodes AG, Hong F. Money boys, HIV risks, and the associations between norms and safer sex: a respondent-driven sampling study in Shenzhen, China. AIDS Behav. 2009;13(4):652–62. [DOI] [PubMed] [Google Scholar]
  • 19.Ruan S, Yang H, Zhu Y, Wang M, Ma Y, Zhao J, et al. Rising HIV prevalence among married and unmarried among men who have sex with men: Jinan, China. AIDS Behav. 2009;13(4):671–6. [DOI] [PubMed] [Google Scholar]
  • 20.Guo H, Wei J-F, Yang H, Huan X, Tsui SK-W, Zhang C. Rapidly increasing prevalence of HIV and syphilis and HIV-1 subtype characterization among men who have sex with men in Jiangsu, China. Am Sex Transm Dis Assoc. 2009;36(2):120–5. [DOI] [PubMed] [Google Scholar]
  • 21.Choi KH, Ning Z, Gregorich SE, Pan QC. The influence of social and sexual networks in the spread of HIV and syphilis among men who have sex with men in Shanghai, China. J Acquir Immune Defic Syndr. 2007;45(1):77–84. [DOI] [PubMed] [Google Scholar]
  • 22.Choi KH, Diehl E, Guo Y, Qu S, Mandel J. High HIV risk but inadequate prevention services for men in China who have sex with men: an ethnographic study. AIDS Behav. 2002;6(3):255–66. [Google Scholar]
  • 23.Choi KH, Gibson DR, Han L, Guo Y. High levels of unprotected sex with men and women among men who have sex with men: a potential bridge of HIV transmission in Beijing, China. AIDS Educ Prev. 2004;16(1):19–30. [DOI] [PubMed] [Google Scholar]
  • 24.Choi KH, Liu H, Guo Y, Han L, Mandel JS, Rutherford GW. Emerging HIV-1 epidemic in China in men who have sex with men. Lancet. 2003;361(9375):2125–6. [DOI] [PubMed] [Google Scholar]
  • 25.He N, Wong FY, Huang ZJ, Thompson EE, Fu C. Substance use and HIV risks among male heterosexual and “money boy” migrants in Shanghai, China. AIDS Care. 2007;19(1):109–15. [DOI] [PubMed] [Google Scholar]
  • 26.Liu S, Wang K, Yao S, Guo X, Liu Y, Wang B. Knowledge and risk behaviors related to HIV/AIDS, and their association with information resource among men who have sex with men in Heilongjiang province, China. BMC Public Health. 2010;10:250. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Cai WD, Zhao J, Zhao JK, Raymond HF, Feng YJ, Liu J, et al. HIV prevalence and related risk factors among male sex workers in Shenzhen, China: results from a time-location sampling survey. Sex Transm Infect. 2010;86(1):15–20. [DOI] [PubMed] [Google Scholar]
  • 28.Wong FY, Huang ZJ, Wang W, He N, Marzzurco J, Frangos S, et al. STIs and HIV among men having sex with men in China: a ticking time bomb? AIDS Educ Prev. 2009;21(5):430–46. [DOI] [PubMed] [Google Scholar]
  • 29.Wei C, Ruan S, Zhao J, Yang H, Zhu Y, Raymond HF. Which Chinese men who have sex with men miss out on HIV testing? Sex Transm Infect. 2011;87(3):225–8. [DOI] [PubMed] [Google Scholar]
  • 30.Marks G, Crepaz N, Senterfitt JW, Janssen RS. 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. J Acquir Immune Defic Syndr. 2005;39 (4):446–53. [DOI] [PubMed] [Google Scholar]
  • 31.Marseille E, Hofmann PB, Kahn JG. HIV prevention before HAART in sub-Saharan Africa. Lancet. 2002;359(9320):1851–6. [DOI] [PubMed] [Google Scholar]
  • 32.Shen J, Yu DB. Governmental policies on HIV infection in China. Cell Res. 2005;15(11–12):903–7. [DOI] [PubMed] [Google Scholar]
  • 33.Choi KH, Lui H, Guo Y, Han L, Mandel JS. Lack of HIV testing and awareness of HIV infection among men who have sex with men, Beijing, China. AIDS Educ Prev. 2006;18(1):33–43. [DOI] [PubMed] [Google Scholar]
  • 34.Wong FY, Huang ZJ, He N, Smith BD, Ding Y, Fu C, et al. HIV risks among gay- and non-gay-identified migrant money boys in Shanghai, China. AIDS Care. 2008;20(2):170–80. [DOI] [PubMed] [Google Scholar]
  • 35.Wong FY, Huang ZJ, He N, Young D, O’Conor CA, Ding Y, et al. Migration and illicit drug use among two types of male migrants in Shanghai, China. J Psychoactive Drugs. 2010;42(1):1–9. [DOI] [PubMed] [Google Scholar]
  • 36.RDS Inc. RDS analysis tool, V.5.6, user manual Ithaca: RDS Inc.; 2006. [Google Scholar]
  • 37.Ramirez-Valles J, Heckathorn DD, Vazquez R, Diaz RM, Campbell RT. From networks to populations: the development and application of respondent-driven sampling among IDUs and Latino gay men. AIDS Behav. 2005;9(4):387–402. [DOI] [PubMed] [Google Scholar]
  • 38.Malekinejad M, Johnston LG, Kendall C, Kerr LR, Rifkin MR, Rutherford GW. Using respondent-driven sampling methodology for HIV biological and behavioral surveillance in international settings: a systematic review. AIDS Behav. 2008;12(4 Suppl): S105–30. [DOI] [PubMed] [Google Scholar]
  • 39.He N, Wong FY, Huang ZJ, Ding Y, Fu C, Smith BD, et al. HIV risks among two types of male migrants in Shanghai, China: money boys vs. general male migrants. AIDS. 2007;21(Suppl 8): S73–9. [DOI] [PubMed] [Google Scholar]
  • 40.Huang ZJ, Wong FY, De Leon JM, Park RJ. Self-reported HIV testing behaviors among a sample of southeast Asians in an urban setting in the United States. AIDS Educ Prev. 2008;20(1):65–77. [DOI] [PubMed] [Google Scholar]
  • 41.Do TD, Hudes ES, Proctor K, Han CS, Choi KH. HIV testing trends and correlates among young Asian and Pacific Islander men who have sex with men in two U.S. cities. AIDS Educ Prev. 2006;18(1):44–55. [DOI] [PubMed] [Google Scholar]
  • 42.Radloff LS. The CES-D scale: a self report depression scale for research in the general population. Appl Psychol Meas. 1977;1: 385–401. [Google Scholar]
  • 43.Hendrick S, Hendrick C. Multidimensionality of sexual attitudes. J Sex Res. 1987;23:502–26. [Google Scholar]
  • 44.Zane N, Yeh M. The use of culturally-based variables in assessment: studies on loss of face. In: Asian American mental health: assessment theories and methods. New York, NY: Kluwer Academic/Plenum Publishers; 2002. p. 123–38. [Google Scholar]
  • 45.Mohr J, Fassinger R. Measuring dimensions of lesbian and gay male experience. Meas Eval Couns Dev. 2000;33:66–90. [Google Scholar]
  • 46.Russell D, Cutrona CE, Rose J, Yurko K. Social and emotional loneliness: an examination of Weiss’s typology of loneliness. J Pers Soc Psychol. 1984;46(6):1313–21. [DOI] [PubMed] [Google Scholar]
  • 47.Iguchi MY, Ober AJ, Berry SH, Fain T, Heckathorn DD, Gorbach PM, et al. Simultaneous recruitment of drug users and men who have sex with men in the United States and Russia using respondent-driven sampling: sampling methods and implications. J Urban Health. 2009;86(Suppl 1):5–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Heckathorn D Deriving valid population estimates from chain-referral samples of hidden populations. Soc Probl. 2002;49:11–34. [Google Scholar]
  • 49.SAS Institute Inc. SAS/STAT 9.2 user’s guide Cary: SAS Institute Inc.; 2010. [Google Scholar]
  • 50.Borgatti S A brief guide to using NETDRAW [Internet]. 2009. http://www.analytictech.com/Netdraw/netdraw.htm.
  • 51.Heckathorn D Respondent driven sampling [Internet]. 2011. http://www.respondentdrivensampling.org/. Accessed 14 Dec 2011. [DOI] [PMC free article] [PubMed]
  • 52.Liu H, Liu H, Cai Y, Rhodes AG, Hong F. Money boys, HIV risks, and the associations between norms and safer sex: a respondent-driven sampling study in Shenzhen, China. AIDS Behav. 2009;13:652–62. [DOI] [PubMed] [Google Scholar]
  • 53.Sheridan S, Phimphachanh C, Chanlivong N, et al. HIV prevalence and risk behaviour among men who have sex with men in Vientane Capital, Lao People’s Democratic Republic. AIDS. 2009;23(3):409–14. [DOI] [PubMed] [Google Scholar]
  • 54.呂麗萍反同志 央視批「要反省」 | 港陸傳真 | 娛樂追星 | 聯合新聞網 [Internet]. World News. 2011. http://udn.com/NEWS/ENTERTAINMENT/ENT8/6444368.shtml. Accessed 3 Aug 2011.
  • 55.Zou H, Wu Z, Yu J, Li M, Ablimit M, Li F, et al. Sexual risk behaviors and HIV infection among men who have sex with men who use the internet in Beijing and Urumqi, China. J Acquir Immune Defic Syndr. 2010;53(Suppl 1):S81–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Rosser BR, Oakes JM, Konstan J, Hooper S, Horvath KJ, Danilenko GP, et al. Reducing HIV risk behavior of men who have sex with men through persuasive computing: results of the Men’s INTernet Study-II. AIDS. 2010;24(13):2099–107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Coleman E, Horvath KJ, Miner M, Ross MW, Oakes M, Rosser BR. Compulsive sexual behavior and risk for unsafe sex among internet using men who have sex with men. Arch Sex Behav. 2010;39(5):1045–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Mansergh G, Koblin BA, McKirnan DJ, Hudson SM, Flores SA, Wiegand RE, et al. An intervention to reduce HIV risk behavior of substance-using men who have sex with men: a two-group randomized trial with a nonrandomized third group. PLoS Med. 2010;7(8):e1000329. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES