Arch Sex Behav DOI 10.1007/s10508-014-0450-8 ORIGINAL PAPER Homosexuality-Related Stigma and Sexual Risk Behaviors Among Men Who Have Sex With Men in Hanoi, Vietnam Huy Ha • Jan M H Risser • Michael W Ross Nhung T Huynh • Huong T M Nguyen • Received: December 2013 / Revised: 16 November 2014 / Accepted: 24 November 2014 Ó Springer Science+Business Media New York 2015 Abstract This article examined the associations between three forms of homosexuality-related stigma (enacted, perceived, and internalized homosexual stigmas) with risky sexual behaviors, and to describe the mechanisms of these associations, among men who have sex with men (MSM) in Hanoi, Vietnam We used respondent-driven sampling (RDS) to recruit 451 MSM into a cross-sectional study conducted from August 2010 to January 2011 Data were adjusted for recruitment patterns due to the RDS approach; logistic regression and path analyses were performed Participants were young and single; most had attended at least some college Nine out of ten participants engaged in sexual behaviors at moderate to high risk levels Compared to those who had no enacted homosexual stigma, men having low and high levels of enacted homosexual stigma, respectively, were 2.23 times (95 % CI 1.35–3.69) and 2.20 times (95 % CI 1.04–4.76) morelikely to engage in high levels of sexual risk behaviors In addition, there was an indirect effect of perceived homosexual stigma and internalized homosexual stigma on sexual risk behaviors through depression and drug and alcohol use Our study provides valuable information to our understanding of homosexual stigma in Vietnam, highlighting H Ha (&) The Institute of Population, Health and Development, 18 Alley 132 Hoa Bang Street, Cau Giay District, 10000 Hanoi, Vietnam e-mail: huy.ha@alumni.uth.edu J M H Risser Á M W Ross School of Public Health, The University of Texas, Houston, TX, USA N T Huynh School of Medicine and Pharmacy, Vietnam National UniversityHanoi, Hanoi, Vietnam H T M Nguyen Center for Community Health Research and Development, Hanoi, Vietnam the need for provision of coping skills against stigma to the gay community and addressing drinking and drug use among MSM, to improve the current HIV prevention interventions in Vietnam Keywords Homosexuality-related stigma Á Sexual risk behavior Á MSM Introduction In Vietnam, homosexuality is often viewed as an immoral and unacceptable lifestyle that conflicts with traditional values that highlight lineage continuity and men’s responsibilities to their families and the society Due to stigma, men may conceal their sexual orientation, ignore the riskiness of some of their sexual behavior, fail to seek HIV/sexually transmitted infection (STI) testing and treatment when needed, and avoid involvement in HIV prevention activities (Doll & Beeker, 1996; Gill, 2002; UNAIDS, 2007) Drawing on stigma theories (Deacon, 2006) and Goffman’s (1963) work, Earnshaw and Chaudoir (2009) have developed a practical and comprehensive HIV-related stigma framework that emphasizes individual-level conceptualizations of stigma This framework depicts three stigma mechanisms—enacted, anticipated, and internalized—to describe how ‘‘social devaluation’’ impacts those who are infected with HIV and those who display HIV-related risk behaviors such as MSM, commercial sex workers, and injecting drug users Earnshaw and Chaudoir indicated that using a framework of multiple-stigma measures is better than a single-stigma mechanism framework in assessing how stigma impacts individuals, and that the multiple-stigma framework is more useful in predicting important psychological, behavioral, and health outcomes 123 Arch Sex Behav The linkage between stigma and increased sexual risk behaviors among MSM is inconsistent in the literature Some studies failed to find an association (Brimlow, Cook, & Seaton, 2003; Courtenay-Quick, Wolitski, Parsons, & Gomez, 2006); while other studies have demonstrated that stigma and high-risk sexual behaviors are related (Alvy, McKirnan, Mansergh, & Koblin, 2011; Koblin, Husnik, Colfax, & Huang, 2006; Preston, Cain, Schulze, & Starks, 2004; USAID and Horizons Program/ Population Council, 2008) A recent large-scale study conducted in 38 European countries indicated that higher levels of internalized stigma were associated with unprotected anal intercourse amongMSM (Ross, Berg, Schmidt, & Hospers,2013) Although there is some evidence of associations between stigma and sexual risk behavior, the mechanism ofsuch associations remains unclear and warrants further research Preston, D’Augelli, Kassab, and Starks (2007) found that low levels of internalized homophobia and certain mental health conditions, such as low self-esteem, mediated the stigma and sexual risk behavior relationship Stigma is often cited as an explanation for the increased mental health outcomes such as depression in MSM (Herek, Gillis, & Ogan, 2009), and may help describe why gay men have more depression compared to heterosexual men (Berlan et al., 2010; King, Semlyen, Tai, Killaspy, & Nazareth, 2008; Marshal, Friedman, Stall, King, & Morse, 2008; Mimiaga, Noonan, Donnell,Safren,&Mayer,2009).Depressionwithotherpsychosocial problems such as substance/alcohol abuse and violence may synergistically interact with one another, resulting in elevated HIV risk behavior among MSM (Hirshfield, Remien, Humberstone, & Walavalkar, 2004; Safren, Raisner, Herrick, & Mimiaga, 2010; Stall, Paul, Greenwood, Pollack, & Catania, 2001) Some studies indicatethatdepressionandsexualrisktakingarerelated(Bancroft et al., 2003; Koblin et al., 2006), while others not (Stall, Mills, Williamson, Hart, & Catania, 2003) In Vietnam, there was an interaction between substance and alcohol abuse, and depression among MSM, especially among male sex workers (MSW), that was linked to sexual risk taking (Colby, 2010) Our article aimed to (1) estimate the prevalence of homosexuality-related stigma, (2) examine the relationship between homosexual stigma and sexual risk behavior among MSM in Hanoi, Vietnam, and (3) to explore the mechanisms of this relationship.Ourhypothesiswas that ahighlevel ofstigmawould havedirect effectsonriskysexual behaviorandindirecteffects on sexual risk taking through depression To our current knowledge, there have been no published studies assessing the effect of stigma against MSM on sexual risk taking behaviors in Vietnam Method Participants A cross-sectional design was used to collect data from 451 MSM in Hanoi from August 2010 to January 2011 Preliminary work 123 in preparation for data collection included identifying networks of MSM and conducting interviews with 10 MSM to pilot and refine the survey instrument Procedure Recruitment was done using the respondent-driven sampling method (RDS), which has been widely used in recent years in different parts of the world to draw probability samples of hidden populations (Heckathorn, 2002) This method combines a modified form of snowball sampling, such as chain referral sampling, with a mathematical system for weighting the data to compensate for the fact that the sample was not drawn randomly With the appropriate weighting, RDS produces unbiased prevalence estimates (Salganik, 2006) To be eligible for this study, individuals had to be biologically male; with age over 18 years; be residents of Hanoi for at least months before the survey implementation; report anal sexual contact (oral and/or anal sexual contact) with other men at least twice during the previous 90 days; have a valid referral coupon (related to the RDS process); and be willing and able to provide informed consent Initially, ten men who were identified during thepreparationprocesstohavelargesocialnetworksofgayfriends were selected as‘‘seeds.’’Each‘‘seed’’received three unique, non-replicable recruitment coupons to give his peers who were expected to fit the eligibility criteria Next, individuals with valid coupons presented themselves to be interviewed Before each interview, a short screening check was performed to ensure eligibility and avoid duplication of subjects, and informed consent form was signed The process was repeated until the desired sample size was reached There were between five and seven waves of RDS recruitment for each ‘‘seed’’ and no RDS waves died out by their own Before the interview, the study purpose—to improve the MSM’s health—was explained and the participant was asked to voluntarily sign the consent form for his participation in the study Face-to face interviews using a structured questionnaire took place in a private room to ensure the confidentiality and protection of subjects’ responses Trained interviewers conducted these interviews at the HIV/STD volunteer counseling and testing clinic placed in Hanoi Dermatology Hospital Upon completion of the interview, each man was given three uniquely numbered coupons to recruit other eligible acquaintances to participate in the study An incentive equivalent of $10 US dollars (USD) was provided to the participant for agreeing to the study and US $2.50 was given for each successful peer recruitment The study was reviewed and approved by the Johns Hopkins University’s Center for Communication Programs and the University of Texas Health Sciences Center at Houston, and the internal review board of the Vietnam Institute for Social Development Studies, which has registered with the NIH (IRB00006556) The study was implemented by the Center for Community Health Research and Development in Hanoi No Arch Sex Behav identifying information, such as names or addresses of respondents, was collected in the study Measures Dependent Variables Sexual risk behaviors were categorized by the degree of vulnerability to HIV infection as (a) no risk, (b) low risk, (c) moderate risk, and (d) high risk The classification was developed with reference to Ostrow, DiFranceisco, and Wagstaff (1998) and Preston et al (2004) and based on (1) number of sexual partners within the 30 days prior to the survey, (2) whether one engaged in anal sexual intercourse, and (3) the consistency of condom use during sexual intercourse in the past 30 days Men were classified as ‘‘no risk’’ if, within the last 30 days, (1) they did not have a sexual partner or (2) they were in a monogamous relationship, but did not engage in sexual intercourse Men were placed in the‘‘low risk’’group ifthey were in a monogamous relationship, and engaged in sexual intercourse but always used condoms during sexual intercourse This group also included men who had multiple sexual partners but always used condoms during sexual intercourse The‘‘moderate risk’’ was assigned to men who were in a monogamous relationship and did not always use a condom during sexual intercourse Men were put in the‘‘high risk’’group if they had multiple sexual partners and did not always use a condom during sexual intercourse For logistic regression analysis, because the distribution of sexual risk behavior was skewed toward the‘‘high risk’’ group, the other three groups—no risk, low risk, and moderate risk—were combined, and the final comparison was between ‘‘high risk’’and all other risk categories combined Independent and Control Variables Homosexuality-related stigma was measured in three forms: enacted homosexual stigma (enacted stigma), perceived homosexual stigma (perceived stigma), and internalized homosexual stigma (internalized stigma) These forms were adapted from previous studiesconductedinChina(Liu,Feng,& Rhodes,2009; Neilands, Steward, & Choi, 2008) and piloted with Vietnamese MSM in Hanoi (Ha, Ross, Risser, & Nguyen, 2013) The adapted scales showed good construct validity with a fairly high internal reliability overall, as well as with each type of stigma Enacted stigma measures the actual experience of prejudice or discrimination directed toward a man due to his homosexual activities Eight questions were used to measure enacted stigma, including one screening question of whether his family, friends, and others knew about his homosexual behavior, and seven subsequent questions regarding prior experiences with homosexual discrimination (Ha et al., 2013) with a 4-point Likert scale responses (1 = never; = once or twice; = a few times; = many times) Perceived stigma measures a man’s perception of how he is being stigmatized against in the ‘‘eyes’’ of the ‘‘community’’because of his same-sex practices This form was measured by asking eleven questions designed to rate the degree of respondents’ disagreement/agreement (1 = strongly agree; = agree; = disagree; = strongly disagree) concerning negative attitudes in the community toward stigmatized persons (Ha et al., 2013) Internalized stigma measures a man’s negative self-perceptions and beliefs about his same-sex behavior due to feelings of shame and fears of the society’s negative perceptions of homosexuals It was measured using seven questions in which men were asked to rate their level of disagreement/agreement (1 = strongly agree; = agree; = disagree; = strongly disagree) in regard to their self-beliefs and feelings such as shame or disgust, resulting in devaluation and internal conflict (Ha et al., 2013) Each form of stigma was represented by a summative score; a higher score indicates a higher level of stigma An individual with a score of 75th percentile or more was considered having a high stigma score, except for enacted stigma where three levels of stigma were defined: no, low, and high—as the enacted stigma composite score was skewed toward zero Cronbach’s alphas of the enacted stigma, perceived stigma, and internalized stigma scales are 0.82, 0.82, and 0.79, respectively Depression status was measured using the Center for Epidemiologic Studies Depression Scale with 20 items (CESD-20) (Radloff, 1977) Participants were rated on the frequency (from ‘‘most or all of the time’’ to ‘‘rarely or none of the time’’) of depression signs/symptoms they had experienced in the week prior to the survey Depression scores were calculated and a threshold of 16 for indicating the likely presence of significant depression (Radloff, 1977) Cronbach’s alpha of the scale exhibited excellent reliability (a = 0.90) Alcohol use was measured by the 10-item Alcohol Use Disorder Identification Test scale (AUDIT-10) developed by the World Health Organization (2001) The four categories of low, medium, high, and severe levels of alcohol problems were generated on the basis of the total AUDIT-10 score: less than 8, 8–15, 16–19, and 20 or above, respectively (WHO, 2001) The reliability coefficient in this sample was 0.82 Drug use was a binary variable indicating if men reported using any of the following illicit substances: marijuana, ecstasy, methamphetamine, ketamine, cocaine, and heroin within 30 days prior to the survey Knowledge about HIV/AIDS was examined through 25 true/ false questions about the cause of HIV/AIDS, HIV transmission, and prevention of HIV infection A summative total score was calculated.HighversuslowknowledgeofHIV/AIDSwasdefined using the 50th percentile cut-off of the total score The reliability coefficient in this sample was 0.83 A few sociodemographic characteristics, including the men’s age and education, were also included as control variables because there is evidence that they were significantly linked to high-risk sexual activity (Preston et al., 2004) 123 Arch Sex Behav Statistical Analysis We used respondent-driven sampling analysis tool (RDSAT) program (www.respondentdrivensampling.org) to adjust the analysis for the relative sizes of the participants’ network, to estimate the weighted prevalence and 95 % confidence intervals for the socio-economic characteristics and behavioral characteristics of participants Median, mean, and standard deviation (SD) were estimated for continuous variables The individual weights associated with the outcome variable (sexual risk behavior) were generated using RDSAT and then were exported to the STATA software for logistic regression analyses Odds ratios (OR), 95 % confidence intervals (95 % CI), and p values were used in univariate and multi-variable logistic regressions Path analysis using the AMOS 18.0 program was performed to determine the relationships among variables (homosexual stigma, depression, drug use, alcohol drinking, and HIV knowledge) and sexual risk behavior Path analysis is often used to determine the comparative strength of direct and indirect associations between variables, and employs multiple equations to estimate the effects of hypothesized linkages of a set of variables (Lleras, 2005) In this analysis, continuous variables such as total index scores of three stigma forms, depression, alcohol use, and HIV knowledge were used We started with and tested the full model; a reduced model was retained by deleting all paths with p value [0.20, however the paths of three forms of homosexual stigma with sexual risk-taking behaviors were kept regardless of their levels of p value The goodness-of-fit of the reduced model was assessed by a non-significant v2 value, the comparative fit index (CFI), the root mean square error of approximation (RMSEA), and Tucker Lewis Index (TLI) (Hu & Bentler, 1999) Subsequently, a path diagram was constructed to show their linkages Results A total of 451 men participated in the study The majority of the participants were young (87.0 % of participants at age B22 years; mean age: 22.5 years, SD = 5.3), single (96.8 %; [95 % CI 94.7–98.5]) and had attended at least some college (75.5 %; [67.6–82.5 %]) More than three-quarters (76.3 %; [69.5–81.9 %]) of the sample were born outside Hanoi but currently lived in Hanoi Most of the men (82.8 %; [76.8– 88.1]) had low levels of HIV/AIDS knowledge Nearly a half of the sample reported having more than two sexual partners within 30 days before the survey The majority (67.6 %; [59.0–75.0]) had a depression score higher than 16, which has been considered ‘‘significant’’ or ‘‘mild’’ depressive symptomatology (Radloff, 1977) Approximately, two out of five men (40.7 %; [31.1–51.1] in the study had an AUDIT-10 score equal to or greater than 16, indicating high or severe levels of alcohol use (WHO, 2001) (Table 1) 123 Most men (79.4 %; [74.2–83.6]) reported that they did not experience homosexual stigma and discrimination Among those who did (21 %), many (17.5 %; [13.7–22.5]) exhibited high rates of stigma/discrimination experience For perceived stigma and internalized stigma, the majority of men reported low levels of perceived stigma (72.7 %; [66.7–78.1]) and internalized stigma (70.1 %; [64.3–76.1]) (Table 1) Nine out of ten men in the sample admitted that they engaged in sexual behaviors at a moderate or high level of risk; more than a half reported sexual behaviors at the high level of risk (results not shown) We found that only enacted stigma was associated with the high level of risky sexual behaviors in both the univariate and multivariate models (Table 2) Compared to those who reported no enacted stigma, men reporting low and high levels of enacted stigma were 2.23 [1.35–3.69] times and 2.20 [1.04–4.76] times more likely, respectively, to have a higher level of engagement inriskysexualbehaviors.Therewasastrongrelationshipbetween high AUDIT-10 scores and risky sexual behavior: men who had AUDIT-10 scores of 16–19 and[19 in the past 30 days were 2.38 [1.24–4.56] times and 2.47 [1.21–5.05] times, respectively, more likely to engage in high levels of sexual risk intercourse than those who had an AUDIT-10 score of \8 Also, men who reported using marijuana, ecstasy, methamphetamine, ketamine, cocaine, or heroin within the past 30 days were 1.98 [1.25–3.13] times more likely to engage in high-risk sexual behaviors than others Figure illustrates results of the path analysis All three forms of homosexuality-related stigma were significantly correlated Similarly to the findings shown in Table 2, Fig indicates that among the three stigma forms, enacted stigma was directly and significantly related to a high level of sexual risk behaviors In addition, enacted stigma and internalized stigma were significantly associated with depression, which inturnwassignificantly associated with drug use and alcohol use; both drug and alcohol use was related to increased sexual risk behaviors In addition, internalized stigma and HIV knowledge were significantly associated; however, HIV knowledge was not significantly linked with other factors such as depression, alcohol use, drug use, or sexual risk behaviors The model provided an excellent fit to thesample data, as indicated by the v2 goodness-offit statistics (v2 square = 4.3; df = 13; p value = 0.99 and goodness-of-fit indices (CFI = 1.00; RMSEA = 0.00; TLI = 1.00) Discussion The purpose of this article was to examine the relationship between homosexuality-related stigma and sexual risk behaviors We tested the hypothesis that homosexuality-related stigma had both direct and indirect effects on sexual risk taking through depression Our findings partially supported the hypothesis: enactedstigmawasdirectlyandsignificantlyassociatedwithsexualrisk behavior This result was consistent with previous studies (Koblin et al., 2006; Thiede, Valleroy, MacKellar, & Celentano, 2003; Arch Sex Behav Table Sociodemographic characteristics of men who have sex with men in Hanoi, Vietnam in 2011 (n = 451) Background characteristics Population-based estimatesa % (95 % CIc) No/low/moderate levels of sexual risk Crude estimatesb % (n) High levels of sexual risk Total sample Age (Median = 21; M: 22.5; SD = 5.3) 18–20 years 57.4 (45.8–71.1) 42.6 (28.9–54.3) 30.6 (23.8–40.2) 20.8 (94) 10–22 years 52.3 (44.4–61.4) 47.7 (38.6–55.6) 56.4 (46.3–62.4) 52.4 (236) C23 years 51.2 (39.8–68.3) 48.8 (31.7–60.3) 23.0 (9.0–18.8) 26.8 (121) Education level Less than college 45.5 (34.1–57.0) 45.5 (34.1–57.0) 24.5 (17.5–32.4) 32.4 (146) 55.1 (47.9–63.3) 44.9 (36.7–52.1) 75.5 (67.6–82.5) 67.6 (305) Single, never get married 52.7 (46.3–60.3) 47.3 (39.7–53.8) 96.8 (94.7–98.5) 93.8 (423) Married/divorced/separated 49.1 (25.7–76.5) 50.9 (23.5–74.3) 3.2 (1.5–5.3) 6.2 (28) Born in Hanoi 58.0 (47.7–70.3) 42.0 (29.7–52.3) 23.7 (18.5–30.5) 31.7 (143) Born outside Hanoi 51.4 (43.6–59.3) 48.6 (40.7–56.4) 76.3 (69.5–81.6) 68.3 (308) Student 96.7 (81.2–62.0) 46.9 (38.0–55.4) 66.5 (57.5–74.5) 57.2 (258) Non-student 52.0 (43.0–62.2) 48.0 (37.8–57.0) 33.5 (25.5–42.6) 42.8 (193) Some college and more Marital status Place of birth Occupation HIV/AIDS Knowledge score (Median = 10; M = 10; SD = 2.7) Low knowledge 56.6 (47.4–64.0) 43.4 (36.0–52.6) 82.8 (76.8–88.1) 81.6 (368) High knowledge 49.1 (28.3–66.8) 50.9 (33.2–71.7) 17.2 (11.9–23.2) 18.4 (83) B1 partner 80.0 (72.0–87.6) 20.0 (12.4–28.1) 54.0 (47.4–60.4) 37.7 (169) 2–4 [4 partners 23.2 (16.1–31.7) 13.8 (5.9–23.4) 76.8 (68.3–83.9) 86.2 (76.6–94.1) 33.9 (28.3–39.7) 12.1 (8.9–15.6) 42.9 (192) 19.4 (87) Number of male partners Depression score (CESD-20) (Median = 22; M = 23; SD = 10) B16 60.5 (44.4–75.3) 39.5 (24.7–55.6) 32.4 (25.0–41.0) 31.9 (144) [16 67.7 (58.3–76.9) 32.3 (23.1–41.7) 67.6 (59.0–75.0) 68.1 (307) AUDIT-10 score (Alcohol use) (Median = 14; M = 14.2; SD = 6.4) \8 61.9 (49.1–78.4) 38.1 (22.0–50.9) 12.4 (8.0–18.2) 16.2 (73) 8–15 60.2 (50.7–69.9) 39.8 (30.3–49.0) 46.9 (39.7–53.5) 41.5 (187) 16–19 42.5 (30.1–55.5) 57.5 (44.5–69.9) 23.1 (18.5–28.3) 23.1 (104) [19 40.0 (23.9–55.3) 60.0 (44.7–76.1) 17.6 (12.6–22.7) 19.3 (87) Yes 60.7 (54.1–69.0) 39.3 (31.0–45.9) 33.9 (27.5–40.7) 30.6 (138) No 35.1 (23.3–48.6) 64.9 (51.4–76.7) 66.1 (59.4–72.5 69.4 (313) 68.3 (308) Drug use Enacted homosexual stigma (Median = 0; M = 3.8; SD = 6.1) No stigma 60.3 (53.3–68.1) 39.7 (31.9–46.7) 79.4 (74.2–83.6) Low stigma 27.5 (18.0–39.3 72.5 (60.7–82.0) 3.2 (1.5–5.0) 5.8 (26) 76.4 (49.3–89.5) 17.5 (13.7–22.5) 25.9 (117) High stigma 23.6 (10.5–50.8) Perceived community stigma (Median = 25; M = 25.1; SD = 5.4) Low stigma 56.3 (50.3–65.2) 43.7 (34.8–49.7) 72.7 (66.7–78.1) 76.1 (343) High stigma 43.8 (32.1–56.1) 56.2 (44.0–67.9) 27.3 (21.9–33.3) 24.0 (108) Internalized homosexual stigma (Median = 17; M = 17.5; SD = 4) Low stigma 55.6 (46.6–64.5) 44.4 (35.5–53.4) 70.1 (64.3–76.1) 71.2 (321) High stigma 48.1 (39.1–57.8) 51.9 (42.2–60.9) 29.9 (23.9–35.7) 28.8 (130) a RDSAT was used b RDSAT was NOT used c 95 % confidence interval 123 Arch Sex Behav Table The relationship between different forms of homosexual stigma and degree of risky sexual behaviors among men who have sex with men in Hanoi, Vietnam (n = 451) Variables Univariate OR (95 % CI) Multivariatea AOR (95 % CI) Enacted homosexual stigma No stigma 1.00 1.00 Low stigma 2.10 (1.31–3.35)** 2.23 (1.35–3.69)* High stigma 2.39 (1.14–5.00)* 2.20 (1.04–4.76)* Perceived community homosexual stigma Low stigma 1.00 1.00 High stigma 1.03 (0.69–1.53) 0.96 (0.62–1.47) Low stigma 1.00 1.00 High stigma 1.01 (0.69–1.48) 0.98 (0.65–1.50) Self-homosexual stigma Depression score (CESD-20) B16 1.00 1.00 [16 1.35 (0.91–2.02) 0.95 (0.61–1.49) 1.00 1.00 AUDIT-10 score \8 8–15 1.41 (0.82–2.44) 1.42 (0.81–2.53) 16–19 2.56 (1.38–4.74)* 2.38 (1.24–4.56)* [19 3.01 (1.57–5.77)** 2.47 (1.21–5.05)* Knowledge of HIV/AIDS Low knowledge 1.00 1.00 High knowledge 1.08 (0.74–1.58) 1.07 (0.71–1.60) Drug use within 30 days No 1.00 1.00 Yes 2.45 (1.59–3.77)** 1.98 (1.25–3.13)* RDSAT weights were included in all analyses OR odds ratio, CI confidence Interval, AOR adjusted odds ratio * p\0.05, ** p\0.001 a Adjusted for education, age, and birth place Fig Final model of the relationship of homosexualityrelated stigma—enacted stigma, perceived stigma, and internalized stigma—to sexual risk behavior among MSM in Hanoi, Vietnam (n = 451) *p\0.05, **p\0.001 123 USAID and Horizons Program/Population Council, 2008) We also found indirect effects of enacted stigma and internalized stigma on sexual risk behaviors through depression, alcohol use, and drug use Depression itself was also indirectly related to sexual risk behavior through drug use and alcohol use Men with an elevated depression score were more likely to report drug and alcohol use, which increased one’s likelihood of engaging in sexual risktaking behaviors at a high level These findings suggest that interventions addressing stigma should also reduce alcohol, drug use, and depression because of the associations between them Like Stall et al (2003), we did not find direct effects of depression on sexual risk behaviors, although others have found evidence of its effect on sexual risk taking (Alvy et al., 2011; Chesney et al., 2003; Koblin et al., 2006; Rogers, Curry, Oddy, & Pratt, 2003; The EXPLORE Study Team, 2004) In our study, the association between depression and sexual risk taking was statistically significant in the univariate analysis, but not significant when we controlled for alcohol use, drug use, education, and age This study also indicated that the relationship between depression and sexual risk behavior is complex and often mediated by one or several psychological factors (Alvy et al., 2011) For instance, a few studies have also indicated similar indirect effects of depression on increased sexual risk taking after adjusting for factors such as social support, education level, sexual abuse (Strathdee,Hogg,Martindale,Cornelisse,&Schechter, 1998),and sexual risk cognition (Beck, McNally, & Petrak, 2003) As previously mentioned, depression had indirect effects on sexual risk taking via increased alcohol and drug use Vietnamese gay men may experience psychological disorders such as stress, depressive symptoms, loneliness, high anxiety levels, and even suicidal ideation due to fears or actual experience with discrimination by their families and communities Consequently, as coping mechanisms, many of them may feel compelled to look for comfort in behaviors that may put them at risk, such as alcohol Arch Sex Behav drinking, drug use, and/or unprotected sex Previous research has documented links between stress, mental health, and stigma (Meyer, 2003) and highlighted sexual risk taking as a way to release stress (Folkman, Chesney, Pollack, & Phillips, 1992; McKusick, Hoff, Stall, & Coates, 1991) Some limitations of the study should be acknowledged First, our data were from Hanoi, which may not be representative of other, smaller cities in Vietnam However, there is evidence that there are more similarities than differences in causes, forms, and consequences of HIV/AIDS stigma between cultures (Songwathana & Manderson, 2001) Thus, it is reasonable to assume that stigma may not vary much by region within Vietnam Second, it is possible that some men chose not to participate in our study due to fears of stigma and discrimination, and that young MSM may be more visible in Vietnam than older MSM, despite much effort to recruit men from various segments of the MSM population in Hanoi As a result, a large proportion of the sample were students in their early to mid-20s, which may not represent the greater MSM population Finally, we did not collect data on other psychosocial factors such as self-efficacy, avoidant coping, and cognitive escape, all of which have been documented to have considerable effects on stigma and high levels of sexual risk taking (Alvy et al., 2011) Despite these limitations, our study is the first time to provide valuable evidence concerning the relationship between homosexuality-related stigma and sexual risk-taking behaviors and their mediating factors among MSM in Vietnam Our findings highlight the important role of stigma reduction activities in HIV prevention efforts in Vietnam While it is difficult to change the views of society toward homosexuality; it is important to empower the gay community with coping skills against stigma, so that we can prevent the negative impacts of homophobic cultures, and address drug and alcohol use, to ultimately improve the effectiveness of HIV/AIDS prevention program in Vietnam Acknowledgments This research was supported by Research to Prevention (R2P) Small Grants Program funded by United States Agency for International Development, awarded to Dr Le Minh Giang at the Center for Community Health Research and Development (CCRD) The John Hopkins Bloomberg School of Public Health/Center for Communication Program is the prime contractor for R2P Authors would like to thank Dr Giang for his supports and CCRD’s staffs, especially Mr Nguyen Tri Trung and Ms Le Mai Phuong for data collection Additional support was provided by Fogarty International Center through grant D43TW007669 The content of this study is solely the responsibility of the authors and does not necessarily represent the official views of the Fogarty International Center or the National Institutes of Health References Alvy,L.M.,McKirnan,D.J.,Mansergh,G.,&Koblin,B.(2011).Depression is associated with sexual risk among men who have sex with men, but is mediatedbycognitiveescapeandself-efficacy.AIDSandBehavior,15, 1171–1179 Bancroft, J., Janssen, E., Strong, D., Carnes, L., Vukadinovic, Z., & Long, J S (2003) Sexual risk-taking in gay men: The relevance of sexual arousability, mood, and sensation seeking Archives of Sexual Behavior, 32, 555–572 Beck, A., McNally, I., & Petrak, J (2003) Psychosocial predictors of HIV/ STI risk behaviors in a sample of homosexual men Sexually Transmitted Infections, 79, 142–146 Berlan, E D., Corliss, H L., Field, A E., Goodman, E., & Austin, S B (2010) Sexual orientation and bullying among adolescents in the growing up today study Journal of Adolescent Health, 46, 366–371 Brimlow, D L., Cook, J S., & Seaton, R (2003) Stigma and HIV/AIDS: A review of theliterature.Washington,DC: USDHHS,Health Resources and Services Administration, HIV/AIDS Bureau Chesney, M A., Koblin, B A., Barresi, P J., Husnik, M J., Celum, C L., Colfax, G., … EXPLORE Study Team (2003) An individually tailored intervention for HIV prevention: Baseline data from the EXPLORE study American Journal of Public Health, 93, 933–938 Colby, D J (2010) Results of research on MSW in Ho Chi Minh city Vietnam: Ho Chi Minh City Courtenay-Quick, C., Wolitski, R J., Parsons, J T., & Gomez, C A (2006) Is HIV/AIDS stigma diving the gay community? Perception of HIV-positive men who have sex with men AIDS Education and Prevention, 18, 56–57 Deacon, H (2006) Towards a sustainable theory of health-related stigma: Lessons from the HIV/AIDS literature Journal of Community & Applied Social Psychology, 16, 418–425 Doll, L S., & Beeker, C (1996) Male bisexual behavior and HIV risk in the United States: Synthesis of research with implications for behavioral interventions AIDS Education and Prevention, 8, 205–225 Earnshaw, V A., & Chaudoir, S R (2009) From conceptualizing to measuring HIV stigma: A review of HIV stigma mechanism measures AIDS and Behavior, 13, 1160–1177 Folkman, S., Chesney, M., Pollack, L., & Phillips, C (1992) Stress, coping, and high-risk sexual behavior Health Psychology, 11, 218–222 Gill, B (2002) China’s HIV/AIDS crisis: Implications for human rights, the Rule of Law and U.S.–China relations Testimony before the Congressional-Executive Commission on China, Roundtable on HIV/AIDS Goffman, E (1963) Stigma: Notes on the management of spoiled identity New York: Simon & Schuster Ha, H., Ross, M W., Risser, J M H., & Nguyen, H T M (2013) Measurement of stigma in men who have sex with men in Hanoi, Vietnam: Assessment of a Homosexuality-related Stigma Scale Journal of Sexually Transmitted Diseases http://dx.doi.org/10.1155/ 2013/174506 Heckathorn, D (2002) Respondent-driven sampling II: Deriving valid population estimates from chain-referral samples of hidden populations Social Problems, 49, 11–34 Herek, G M., Gillis, J R., & Ogan, J C (2009) Internalized stigma among sexual minority adults: insights from a social psychological perspective Journal of Counseling Psychology, 56, 32–43 Hirshfield, S., Remien, R H., Humberstone, M., & Walavalkar, I (2004) Substance use and high-risk sex among men who have sex with men: A national online study in the USA AIDS Care, 16, 1036–1047 Hu, L., & Bentler, P M (1999) Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives Structural Equation Modeling, 6, 1–55 King, M., Semlyen, J., Tai, S S., Killaspy, H., & Nazareth, I (2008) A systematic review of mental disorder, suicide, and deliberate self harm in lesbian, gay, and bisexual people BMC Psychiatry, 8, 70 doi:10.1186/1471-244X-8-70 Koblin, B A., Husnik, M J., Colfax, G., & Huang, Y (2006) Risk factors for HIV infection among men who have sex with men AIDS, 20, 731–739 Liu, H., Feng, T., & Rhodes, G A (2009) Assessment of the Chinese version of HIV and homosexuality related stigma scales Sexually Transmitted Infections, 85, 65–69 123 Arch Sex Behav Lleras, C (2005) Path analysis Encyclopedia of social measurement (pp 25–30) New York: Academic Press Marshal, M P., Friedman, M S., Stall, R., King, K M., & Morse, J Q (2008) Sexual orientation and adolescent substance use: A metaanalysis and methodological review Addiction, 103, 546–556 McKusick, L., Hoff, C C., Stall, R., & Coates, T J (1991) Tailoring AIDS prevention: Differences in behavioral strategies among heterosexual and gay bar patrons in San Francisco AIDS Education and Prevention, 3, 1–9 Meyer, I H (2003) Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: Conceptual issues and research evidence Psychological Bulletin, 129, 674–697 Mimiaga, M J., Noonan, E., Donnell, D., Safren, S A., & Mayer, K H (2009) Childhood sexual abuse is highly associated with HIV risk taking behavior and infection among MSM in the EXPLORE Study Journal of Acquired Immune Deficiency Syndromes, 51, 340–348 Neilands, B T., Steward, W T., & Choi, H K (2008) Assessment of stigma towards homosexuality in China: A study of men who have sex with men Archives of Sexual Behavior, 37, 838–844 Ostrow, D G., DiFranceisco, W., & Wagstaff, D (1998) The Coping and Change Study of Men at Risk of AIDS: Sexual Behavior and Behavior Change Questionnaire In C M Davis, W L Yarber, R Bauserman, G Sheer, & L Davis (Eds.), Handbook of sexualityrelated measures (pp 547–553) Thousand Oaks, CA: Sage Preston, D B., Cain, R E., Schulze, F W., & Starks, M T (2004) The influence of stigma on the sexual risk behavior of rural men who have sex with men AIDS Education and Prevention, 16, 291–303 Preston, D B., D’Augelli, A R., Kassab, C D., & Starks, M T (2007) The relationship of stigma to the sexual risk behavior of rural men who have sex with men AIDS Education and Prevention, 19, 218–230 Radloff, L S (1977) The CES-D scale: A self report depression scale for research in the general population Applied Psychological Measurement, 1, 385–401 Rogers, G., Curry, M., Oddy, J., & Pratt, N (2003) Depressive disorders and unprotected casual anal sex among Australian homosexually active men in primary care HIV Medicine, 4, 271–275 Ross, M W., Berg, R C., Schmidt, A J., & Hospers, H J (2013) Internalized homonegativity predicts HIV-associated risk behavior in European men who have sex with men in a 38-country acrosssectional study: Some public health implications of homophobia BMJ Open, 3, e001928 123 Safren, S A., Raisner, S L., Herrick, A., & Mimiaga, M J (2010) Mental health and HIV risk in men who have sex with men Journal of Acquired Immune Deficiency Syndromes, 55, S74–S77 Salganik, M J (2006) Variance estimation, design effects, and sample size calculations for respondent-driven sampling Journal of Urban Health, 83(6 Suppl.), i98–112 Songwathana, P., & Manderson, L (2001) Stigma and rejection: Living with AIDS in villages in southern Thailand Medical Anthropology, 20, 1–23 Stall, R., Mills, T C., Williamson, J., Hart, T., & Catania, J A (2003) Association of co-occurring psychosocial health problems and increased vulnerability to HIV/AIDS among urban men who have sex with men American Journal of Public Health, 93, 939–942 Stall, R., Paul, J., Greenwood, G., Pollack, L M., & Catania, J A (2001) Alcohol use, drug use, and alcohol related problems among men who have sex with men: The Urban Men’s Health Study Addiction, 96, 1589–1601 Strathdee, S A., Hogg, R S., Martindale, S L., Cornelisse, P G., & Schechter, M T (1998) Determinants of sexual risk-taking among young HIV-negative gay and bisexual men Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology, 19, 61–66 The EXPLORE Study Team (2004) Effects of a behavioral intervention to reduce acquisition of HIV infection among men who have sex with men: The EXPLORE randomized controlled study Lancet, 364, 41– 50 Thiede, H., Valleroy, L A., MacKellar, D A., & Celentano, D D (2003) Regional patterns and correlates of substance use among young men who have sex with men in US urban areas American Journal of Public Health, 93, 1915–1921 UNAIDS (2007) Men who have sex with men, the missing piece in national response to AIDS in Asia and the Pacific Available at: http://www wisdomofwhores.com/wp-content/uploads/2010/12/MSM-themissing-piece-Aug-2007.pdf Accessed on August 30, 2013 USAID and Horizons Program/Population Councill (2008) Assessment of risk factors for HIV infection among men who have sex with men in the metropolitan area of Campinas City, Brazil, using respondent-driven sampling Population Council World Health Organization (2001) The Alcohol Use Disorders Identification Test: Guidelines for use in primary care Available at: http:// whqlibdoc.who.int/hq/2001/who_msd_msb_01.6a.pdf Accessed on August 30, 2013  ... examine the relationship between homosexuality-related stigma and sexual risk behaviors We tested the hypothesis that homosexuality-related stigma had both direct and indirect effects on sexual risk. .. concerning the relationship between homosexuality-related stigma and sexual risk- taking behaviors and their mediating factors among MSM in Vietnam Our findings highlight the important role of stigma. .. aimed to (1) estimate the prevalence of homosexuality-related stigma, (2) examine the relationship between homosexual stigma and sexual risk behavior among MSM in Hanoi, Vietnam, and (3) to explore