Gay Community Integration as Both a Source of Risk and Resilience for HIV Prevention in Beirut

International Journal of Behavioral Medicine https://doi.org/10.1007/s12529-019-09829-9 FULL LENGTH MANUSCRIPT Gay Community Integration as Both a Source of Risk and Resilience for HIV Prevention in Beirut Carol Abi Ghanem 1 & Cynthia El Khoury 2 & Matt G. Mutchler 3,4 & Bonnie Ghosh-Dastidar 5 & Susan Kegeles 6 & Elie Balan 7 & Jacques E. Mokhbat 8,9 & Glenn J. Wagner 10 # International Society of Behavioral Medicine 2019 Abstract Background Aspects of sexual identity development, including integration into gay community, have been found to be associated with sexual risk behavior among men who have sex with men (MSM), but not in the Middle East. Method Drawing on the minority stress model and integrated theory of health behavior, we examined the relationships between measures of sexual identity development and HIV protective behaviors (condomless anal sex and HIV testing) and the mediating roles of HIV knowledge, peer judgmentalism, and discrimination, in a sample of 226 young MSM in Beirut, Lebanon. A cross-sectional, mediation analysis was conducted using a bootstrapping approach and logistic regression models. Results Fifteen percent of the sample reported recent condomless anal sex with partners whose HIV status was positive or unknown, and 82.3% had ever been tested for HIV. Multivariate logistic regression analysis showed that greater integration into the gay community was significantly associated with having recent condomless anal sex with positive/unknown HIV status partners and having any history of HIV testing, after controlling for covariates. Knowledge of HIV risk was associated with gay integration as well as both condomless anal sex with positive or unknown HIV status partners and any history of HIV testing, but it only served as a (partial) mediator of HIV testing. Peer judgmental communication about sex and sexuality-related discrimination were related in bivariate analysis to gay integration and condomless anal sex, but they too did not prove to be significant mediators of the relationship between these two constructs. Conclusion These findings highlight the potential sexual health benefits and vulnerabilities associated with increased integration into the gay community for young MSM in Beirut and the need to better understand how to increase HIV knowledge while limiting its potential to increase sexual risk behavior. Keywords Discrimination . HIV knowledge . Social support . Condomless anal sex . HIV testing . Gay community integration 1 The Lebanese Center for Policy Studies, Beirut, Lebanon 2 Beirut, Lebanon 3 California State University Dominguez Hills, Carson, CA, USA 4 AIDS Project Los Angeles Health, Los Angeles, CA, USA 5 RAND Corporation, Santa Monica, CA, USA Susan Kegeles Susan.Kegeles@ucsf.edu 6 Center for AIDS Prevention Studies, University of California at San Francisco, San Francisco, CA, USA Elie Balan eballan@afemena.org 7 Arab Foundation for Freedom and Equality, Beirut, Lebanon * Carol Abi Ghanem carol.abi.ghanem@gmail.com Matt G. Mutchler mmutchler@csudh.edu Bonnie Ghosh-Dastidar bonnieg@rand.org 8 Jacques E. Mokhbat jacques.mokhbat@gmail.com Lebanese AIDS Society, Beirut, Lebanon 9 Glenn J. Wagner gwagner@rand.org Department of Medicine, Lebanese American University School of Medicine, Beirut, Lebanon 10 RAND Corporation, Santa Monica, CA, USA Int.J. Behav. Med. Introduction Processes of sexual identity development can play a key role in motivating sexual health behavior among men who have sex with men (MSM). Aspects of sexual identity development, including self-acceptance, coming out to and receiving support from friends and family, and becoming integrated into the gay community where social support and resources may be more readily available, can serve to motivate self-care and healthy behavior, including consistent condom use and HIV testing [1]. This is in line with the integrated theory of health behavior, which posits that health behavior change is enhanced by improving knowledge, self-regulation skills and social support [2]. Conversely, there are minority stress processes associated with greater expression of a sexual minority identity that can impede the above described mechanisms for health promotion, such as increased exposure to stigma (both internal and external) and sexuality-related discrimination [3]. Consistent with the minority stress model [4, 5], the pervasive stigma and discrimination that sexual minorities experience can contribute to poor psychological well-being, internalized shame and self-loathing [6, 7], as well as engagement in high risk behaviors (e.g., condomless anal sex with partners who are HIVpositive or of unknown HIV status) that result in a vulnerability to HIV and other sexually transmitted infections [8]. Greater involvement and integration into the gay community could increase exposure to minority stressors (e.g., stigma and discrimination) as well as opportunities for sexual encounters, thereby increasing the likelihood of HIV-related sexual risk behaviors. A study in San Francisco, where gay community norms in the 1990s favored risky behavior, found that greater integration into gay community (e.g., being “out of the closet”) was associated with increased rates of condomless anal sex in a sample of young gay and bisexual men who were untested for HIV [9]. Studies have in fact revealed mixed effects of sexual identity development on HIV risk behavior [10, 11], and the potential for positive and negative effects of greater involvement in the gay community may in part explain such findings. It is important to gain a better understanding of how the interplay of identity formation, community integration, and minority stress processes related to sexual health behavior. There are multiple pathways by which greater integration into the gay community can relate to sexual health behavior, both in positive and negative ways. Greater integration may be associated with sexual risk behavior through increased opportunities to meet sex partners with whom to engage in condomless sex, as well as more exposure to discrimination and stigma [12]. But greater integration into the gay community can also be associated with healthy sexual behavior through increased access to information about HIV risk and services to promote prevention. Moreover, greater integration often results in more social relations with others in the community, which can result in increased peer support for sexual health and social support in general, as well as heightened exposure to minority stressors [13]. The minority stress model posits that factors such as social support and other resources (e.g., knowledge, access to services) can serve as a buffer from stressors, while negative social aspects (i.e., discrimination and stigmatization) can exacerbate stress [4]. This is supported by research showing that social support attenuates and mediates the association of greater integration into the gay community and increased sexual risk behavior [14], and heightened experiences of sexuality-related discrimination and peer judgmental communication about sex are associated with increased sexual risk taking [6, 15] and may strengthen the association between gay community integration and HIV risk behavior. It would be useful to examine how these constructs may be related through studies of MSM in different cultural contexts. The Middle East North Africa (MENA) region has the second fastest growing HIV epidemic in the world [16], with most cases occurring among marginalized communities including MSM. Although the HIV prevalence rate among MSM in Lebanon was estimated to be 3.5% in 2010 [17], a study in 2015 found a rate of 15% [18]. These high HIV rates among MSM in Beirut are likely related to their reporting rates of condomless anal sex in the range of 60–70% (including 20– 30% with HIV-positive or unknown status partners). Unlike much of the Middle East, the gay community in Beirut has experienced significant development over the past 15 years, despite the continued presence of strong societal stigmatization and discrimination. Beirut presents a rich context for examining how aspects of sexual identity development, including integration into the gay community, and minority stress processes related to HIV protective behaviors. Beirut has a prominent civil society presence that advocates for rights and services the needs of sexual minorities, and there is a growing number of relatively safe spaces (bars, cafes, community centers) for the community. Moreover, the use and efficiency of social networking phone and Internet platforms are increasingly used in Lebanon to access gay-related spaces, events, information, and interpersonal relationships, thereby increasing one’s integration and exposure to the gay community. These environmental elements can not only provide a supportive context for promoting sexual identity development and result in enhanced HIV protective behaviors but also provide increase opportunities for engaging in risky sexual behavior and exposure to facets of the community that may trigger minority stress processes such as internalized stigma and discrimination. Despite the growth and vibrance of the gay community in Beirut, Lebanon continues to criminalize homosexuality, and Beirut remains a place where security forces can stop any person, including men they suspect of being gay, to search them for evidence of their homosexuality. Int.J. Behav. Med. MSM who do not have legal resident status (e.g., refugees) and have materials (e.g., apps on their mobile phones, condoms in their pockets) can be arrested and thrown into jail. With data from a sample of young MSM in Beirut, and drawing on the minority stress model and integrated theory of health behavior, we examined how aspects of sexual identity development (comfort with one’s sexuality, integration into the gay community, and being out to one’s family) were related to condom use and HIV testing, and how these relationships may be mediated by knowledge of HIV risk, sexuality-related discrimination, and both positive and negative forms of social support. Method interview, as well as for each recruit ($15) (up to 3) who enrolled in the study. Measures The survey was administered in English or Arabic, depending on the preference of the participant, with computer-assisted interview software. The survey was developed in English and translated into Arabic using standard translation and back translation methods. Participants were given the option of completing the survey on their own or having the interviewer administer the survey, but the study interviewers reported that it was very rare (less than 10%) for a participant to choose to self-administer the survey. Survey measures were developed by the study team, unless otherwise noted by cited sources. Study Design and Participant Recruitment This study is an open trial of a community-based HIV prevention and sexual health promotion intervention that uses a longitudinal cohort to examine the intervention effects on the larger young MSM community. The baseline data from this cohort provide the data for this analysis. Recruitment of the cohort took place between July 2016 and March 2017 using long-chain peer referral methods primarily, though other methods such as recruitment flyers, postings on social media, and word of mouth were added near the end of recruitment in order to enable the study to reach its target sample. Eligibility criteria consisted of being biologically male and male-identified, age 18 to 29 years, fluent in English or Arabic, residing in greater Beirut, and having had oral or anal sex with a man in the past 12 months. Only participants, who reported being untested for HIV or being HIV-negative at their last test, were included in the analysis for this paper. For the long-chain peer referral methods, which are commonly used to recruit hidden, stigmatized populations, recruitment began with a small number of eligible persons designated as “seeds” who were identified through recommendations from community organizations working with MSM and our community advisory board study and were purposively selected to be well-connected and to represent the diversity in the community. All participants, including seeds and those recruited through flyers, postings, and word of mouth, received three recruitment coupons to recruit members of their social network, resulting in multiple waves of participants. Participants were instructed to give a coupon to eligible MSM peers who were interested in participating and to inform the recruit to call the study coordinator for coupon verification, eligibility screening, verbal consent procedures, and scheduling of an interview. The survey interview was administered at the project office, by either an MSM or female interviewer, depending on the preference of the participant. Participants were compensated $40 for completing the Outcomes HIV Protective Behaviors To assess recent condom use during anal sex with partners whose HIV status was believed to be positive or unknown, respondents were asked to indicate their number of male sex partners in the past three months. For receptive and insertive anal intercourse, respondents were asked: how many times they engaged in the act over the past three months, how many of those acts involved the use of a condom, and the HIV status of the partners with whom condomless acts were engaged with. Respondents indicated how many of these partners “told you he/she was HIV negative and you had no reason to doubt it,” “you knew this man/ woman was HIV positive,” and “you were not completely sure of this man/woman’s HIV status.” A dichotomous variable was created to indicate whether or not any condomless anal sex took place with a male partner who was known to be HIV-infected or whose HIV status was unknown to the respondent, in the past 3 months. This variable indicates engagement in sexual behavior that poses a risk for HIV transmission; however, it should be noted that we did not assess perception of viral load status among HIV-positive partners. Individuals in Lebanon are required to pay for their viral load status laboratory tests, which are quite expensive, so most individuals are unaware of their current viral loads. Therefore, it is plausible that some reports of condomless sex with HIV-positive partners did not involve HIV risk if the partner had an undetectable viral load. Also, preexposure prophylaxis (PrEP) is difficult to access and expensive in Lebanon, adding further support to condomless anal sex representing a high-risk behavior when involving a sex partner whose HIV status is positive or unknown. To assess history of HIV testing, respondents were asked whether or not they had ever tested for HIV, and if they had tested, whether they had tested in the past 6 months. Int.J. Behav. Med. Predictors Aspects of Sexual Identity Development Comfort with sexual orientation was assessed with a single item in which participants were asked to rate their comfort from 1 “very comfortable” to 5 “very uncomfortable”; for analysis, we used a binary variable to represent whether or not the respondent was very comfortable with their orientation. Gay community integration was measured with five items that assessed the proportion of social time spent with MSM (from 0 “none at all” to 4 “all of the time”), degree of being open about one’s sexuality in one’s personal life and at work or school (from 0 “not at all” to 4 “completely,” in separate items), frequency of spending time at predominantly gay venues such as bars and coffee houses (from 0 “never” to 4 “several times a week”), and frequency of using gay social networking phone apps or websites (from 0 “never” to 4 “several times a day”); the mean item score was calculated and used in analyses. Cronbach’s alpha for this scale was .59. Respondents were also asked with a single item to rate the degree to which they feel there is a sense of community among MSM in Beirut (from 0 “not at all” to 4 “A lot”). Family awareness of respondent’s sexual identity was measured as a proxy for MSM disclosure, using a single item in which participants were asked whether or not “some members of your family know your sexual identity and that you have sex with men.” often (never, rarely, often, always) they have expressed or felt judgmental attitudes with/by their peers when communicating about sexual behavior (e.g., “I have called a friend “stupid” or “dumb” for having sex without a condom (even as a joke)”; “If I had sex without a condom, a friend would judge me for it”). The mean item score was calculated, with higher scores representing greater judgmentalism; Cronbach’s alpha was .76. Peer Communication Regarding HIV Prevention Using an 8item scale developed by Kegeles et al. [22], respondents were asked to report how many times in the last 60 days that they and their MSM friends had talked about or encouraged each other to engage in HIV protective behaviors (e.g., how to negotiate condom use, shared experiences of having used condoms, HIV testing) or given each other condoms to use or safe sex literature. The mean item score was calculated; Cronbach’s alpha was .88. Knowledge of Factors Influencing HIV Risk Using a measure developed by Bingham et al. [20], respondents were asked to indicate whether 18 statements related to how one can contract or transmit HIV risk were true or false. Examples of items include, “Having a sexually transmitted infection does not affect a person’s risk of getting HIV” and “If an HIV positive person has an undetectable viral load, they cannot transmit HIV through sex.” A sum of correct responses was tabulated. Mediators Social Support Three items from the Social Relationship Scale [19] were used to measure general social support; respondents were asked about access to someone for emotional (to talk to about personal problems), caregiver (when sick or unable to care for self), and tangible (e.g., money, transport) support. Response options range from 1 “definitely not” to 5 “definitely yes,” with a mean item calculated and higher scores representing greater support; Cronbach’s alpha for this scale was .74. Peer support for sexual health was measured with two items developed by [20] that asks respondents to rate their level of agreement with these statements: “I have friends who I can talk to if I find out I have a sexually transmitted infection” and “I have friends who I can talk to if I have unprotected sex”; response options ranged from 1 “strongly agree” to 5 “strongly disagree,” scores were reversed and mean item score calculated, and higher scores represent greater peer support. Cronbach’s alpha for this 2-item scale was .84. Judgmental Peer Communication About Sexuality Using an 11-item measure developed by McDavitt and Mutchler [15, 21], respondents were asked about their experiences and communications that they have with their friends and others regarding sexual health behaviors. They were asked to rate how Sexuality-Based Discrimination Discrimination was measured with the subscale of the Multiple Discriminations Scale [23] that asks the respondents to indicate whether or not they experienced any of five types of discriminatory events (e.g., insulted or made fun of; denied or lost a job; physically assaulted) in the past year as a result of others thinking the respondent was gay or bisexual. The sum of types of discrimination experienced was used in analyses; Cronbach’s alpha was .51. Covariates These included age, education level (for analysis, a binary variable was created representing whether or not any university level education had been received), current work status (employed or not), and relationship status. Data Analysis Initial analyses using bivariate statistics (chi-squared tests, two-tailed, independent t tests, Pearson correlation coefficients) were conducted to examine associations between the covariates, predictors, proposed mediators, and outcomes of interest. Next, we evaluated whether the mediators statistically mediated the significant associations between predictors and Int.J. Behav. Med. outcomes. To identify our mediation models, we required there be significant associations between (1) the independent variables (IV; measures of sexual identity development) and the dependent variable (DV; HIV protective behaviors), (2) the IV and the proposed mediator, and (3) the proposed mediators and the DV. If these three requirements were met, we proceeded with mediation analysis. In this part of the analysis, we first estimated a logistic regression model with the IV and the DV, excluding the potential mediator. Second, we estimated the same model including the potential mediator and assessed whether the mediator changed the magnitude of the association between the IV and the DV. These regression models included four important socio-demographic characteristics (age, any university education, relationship status, employment status) as covariates. The direct effect is the association between the IV and the DV, and the indirect effect represents the portion of the relationship between the IV and the DV that is mediated by the mediator. We conducted significance testing of direct and indirect effects associated with a mediation hypothesis, using a bootstrap approach with 1000 samples [23]. All of the logistic regression models described above included a cluster adjustment to account for dependence among persons recruited by the same individual (via long-chain referral recruitment). We used the Taylor series (linearization) method for computing cluster-adjusted variances [24]. The cluster-adjusted regression analyses were conducted using SAS survey analysis procedures [25]. Results Sample Characteristics A sample of 226 YMSM enrolled in the study; eight respondents reported being HIV-positive and were thus removed from the analysis, resulting in an analytic sample of 218 men. Table 1 lists the characteristics of the 218 YMSM, including socio-demographics, HIV protective behaviors, aspects of sexual identity development and potential mediators of the relationship between aspects of sexual identity development and HIV protective behaviors. The mean age of the sample was 23.8 years (SD = 3.0), nearly half (46.8%) were currently attending university, and a quarter (24.8%) were in a committed relationship; 74.3% were born in Lebanon. HIV Protective Behaviors Condom Use in Recent Anal Sex Of the 218 men, 176 (80.7%) reported having anal intercourse with men in the past 3 months, including 108 who had receptive anal sex (RAS) and 141 who had insertive anal sex (IAS). Just under half (n = 86; 48.9%) of the 176 men who had anal sex reported any condomless anal sex within the past 3 months, of whom 30 (17.0%) had such sex with an HIV-positive (n = 4) or unknown status (n = 26) partner. When considering the whole sample, with those not having recent anal sex represented as not having condomless anal sex, 39.4% (n = 86) reported any recent condomless anal sex, including 13.8% (n = 30) who had such sex with an HIV-positive or unknown status partner. HIV Testing The vast majority (81.7%) of the sample reported being tested for HIV in their lifetime, but only half (50.9%) had been tested within 6 months prior to the survey. Aspects of Sexual Identity Development and Their Relationship to HIV Protective Behaviors Participants in the sample were generally comfortable with their sexual identity with 70.2% being very comfortable with their sexual identity, 58.1% being mostly or completely out in their social/personal life, 38.4% being mostly/ completely out at their workplace, and 73.0% having some family members who are aware of their sexual identity. As a group, respondents were somewhat integrated into the gay community (M = 2.26, SD = .84) with 40.6% having frequented a gay friendly venue at least once a week and 48.6% accessing gay online social networking applications at least several times a week. In bivariate analysis, the only measure of sexual identity development that was significantly correlated with having any condomless anal sex with partners whose HIV status was positive or unknown was greater integration into the gay community (see Table 2). Greater integration into the gay community was also the only significant correlate of ever having been tested for HIV, among aspects of sexual identity development (see Table 2); having been tested for HIV in the past 6 months was not significantly associated with any measure of sexual identity development, though it was marginally associated with greater integration into the gay community (data not shown). Potential Mediators of the Relationships Between Aspects of Sexual Identity Development and Condomless Anal Sex Among the potential mediators, those significantly correlated with having had any condomless anal sex in the past 3 months with partners whose HIV status was positive or unknown consisted of greater knowledge of HIV risk, greater perceived judgmentalism in communication about sex, greater number of types of gay-related discrimination experienced, and lower general social support (see Table 2). Of these, significant correlates of integration into the gay community consisted of knowledge of HIV risk (r = .23, p < .001), judgmentalism (r Int.J. Behav. Med. Table 1 Sample characteristics of socio-demographics, HIV protective behaviors, sexual development, and the proposed mediators Mean (SD)/n (%) Socio-demographics Mean age (years) Highest level of formal education: Did not complete high school Completed high school 23.8 (3.0) 11 (5.0%) 21 (9.6%) Attended some university University degree Currently attending university Employed Low monthly income (< $1000 USD) In a committed relationship Born in Lebanon Self-identify as gay HIV protective behaviors Any condomless anal sex with male partners whose HIV status is positive or unknown, in past three months Any history of being HIV tested 112 (51.4%) 74 (33.9%) 102 (46.8%) 125 (57.3%) 153 (71.5%) 54 (24.8%) 162 (74.3%) 177 (82.3%) Tested for HIV in the past 6 months Aspects of sexual identity development Discomfort with sexual identity Integration into the gay community Family aware of respondent’s sexual identity and sex with men Potential mediators Knowledge of HIV risk Peer communication about HIV prevention Social support General social support Peer support for sexual health Peer judgmentalism regarding sex Number of types of sexuality-related discrimination experienced in past year 111 (50.9%) 30 (13.8%) 178 (81.7%) 1.47 (0.88) 2.26 (0.84) 154 (73.0%) 12.79 (2.76) 4.47 (6.53) 4.32 (0.97) 1.53 (1.06) 1.97 (0.61) 1.72 (1.65) SD standard deviation = .18, p = .008), and sexuality-related discrimination (r = .28, p < .001), which enabled us to examine whether these three variables mediate the relationship between gay integration and condomless anal sex. As depicted in Fig. 1, regression analysis showed that knowledge of HIV risk partially mediated the relationship between integration into the gay community and condomless anal sex (see Table 3); integration into the gay community remained significantly associated with condomless anal sex [OR (95% CI) = 1.8 (1.02, 3.1)], which is the direct effect, while knowledge of HIV risk was marginally correlated [OR (95% CI) = 1.2 (0.98, 1.4)], but the indirect effect was significant [OR (95% CI) = 1.1 (1.01, 1.3)]. Neither peer judgmental communication about sex nor sexuality-related discrimination mediated the relationship between gay integration and condomless anal sex, as neither were significant correlates of condomless anal sex and the indirect effects were only marginally significant (see Table 3). Analysis of Potential Mediators of the Relationship Between Aspects of Sexual Identity Development and HIV Testing Among the potential mediators, the only correlate of any lifetime history of HIV testing was greater knowledge of HIV risk (see Table 2). As stated above, integration into the gay community was the only aspect of sexual identity development that was significantly correlated with lifetime HIV testing, and it was also positively correlated with knowledge of HIV risk. Regression analysis showed that knowledge of HIV risk partially mediated the relationship between gay integration and HIV testing, as gay integration remained significantly associated with HIV testing [OR (95% CI) = 1.6 (1.01, 2.5)], as was HIV knowledge [OR (95% CI) = 1.2 (1.01, 1.3)], and both the direct [OR (95% CI) = 1.6 (1.01, 2.5)] and indirect effects [OR (95% CI) = 1.1 (1.0, 1.2)] were significant (see Table 3). Int.J. Behav. Med. Table 2 Bivariate correlates of any condomless anal sex with male partners whose HIV status was positive or unknown in the past 3 months and any history of HIV testing Variable Any condomless anal sex with HIV+/unknown status partner(s) Any history of HIV testing No Yes p No Yes p Discomfort with sexual identity Family know respondent’s sexual identity and sex with men Integration into gay community Knowledge of HIV risk Peer communication regarding HIV prevention General social support 1.51 (.91) 133 (72.7%) 2.20 (.81) 12.6 (2.8) 4.43 (6.30) 4.40 (.92) 1.27 (.64) 21 (75.0%) 2.63 (.94) 13.8 (2.2) 4.74 (7.93) 3.83 (1.14) .082 .797 .009 .031 .820 .003 1.50 (.91) 28 (71.8%) 1.97 (.75) 11.6 (2.6) 2.71 (4.88) 4.38 (1.07) 1.47 (.88) 126 (73.3%) 2.32 (.84) 13.1 (2.7) 4.90 (6.82) 4.31 (.95) .827 .853 .015 .001 .060 .707 Peer support for sexual health Peer judgmentalism regarding sex Number of types of sexuality-related discrimination experienced 1.52 (1.10) 1.94 (.60) 1.62 (1.64) 1.63 (.74) 2.18 (.65) 2.33 (1.63) .573 .046 .028 1.39 (.84) 1.89 (.58) 1.80 (1.68) 1.56 (1.10) 1.99 (.62) 1.70 (1.65) .339 .320 .736 Note that all of the above listed statistics are from two-tailed, independent t tests Discussion This sample of young, well-educated MSM in Beirut reported high levels of any recent condomless anal sex, but less than one-fifth reported any condomless anal sex with HIV-positive or unknown status partners, and the vast majority had tested for HIV at least once. The data from this study revealed mixed findings regarding the associations between aspects of gay community integration and HIV protective behaviors. Integration into the gay community (i.e., spending more time with other MSM and at gay-friendly venues and on gayrelated social media outlets, and being open about their sexuality) was associated with not only getting tested for HIV but also having condomless anal sex with partners that posed a risk for HIV transmission (i.e., whose HIV status was believed to be positive or unknown), particularly given that preexposure prophylaxis (PrEP) remains difficult to access in Lebanon. These findings demonstrate that integration into the gay community can have a healthy influence as well as pose challenges to HIV prevention. Greater involvement in or exposure to the gay community increases access to other MSM, and opportunities to find sexual partners, which increases the chances of having at least one recent encounter of condomless anal sex. Greater Fig. 1 Knowledge of HIV risk knowledge partially mediates the relation between gay community integration and condomless anal sex with positive or unknown HIV status partners integration into the community can increase sources of support and contribute to feeling more comfortable with one’s sexual identity; however, it also exposes one to the vulnerabilities of the community and minority stress processes. These vulnerabilities include internalized stigma and judgmental communication regarding sex—a negative form of social influence that can have a negative impact on sexual health, as reflected in its positive correlation with high risk condomless anal sex in this study. Internalized stigma in a young MSM is likely to be further exacerbated by increased exposure to a gay community that expresses judgment regarding sex. This could serve to lower self-esteem and decrease motivation to engage in protective sex behavior. In a context (such as Beirut) where being gay is stigmatized, young gay men may internalize this stigma, and then project the internalized stigma on their peers [15, 21]. Similarly, increased involvement in the gay community can also increase exposure to sexuality-related discrimination and external stigma from the larger society, which may contribute to internalized shame and stigma. The role that integration into the gay community plays in knowledge of HIV risk may be key to understanding the nuances by which gay integration can have either a positive or negative influence on sexual health behavior. Knowledge of HIV risk was positively correlated with integration into the β= 0.75 (.32, 1.17) ** Knowledge of HIV risk Condomless anal sex with HIV+ or unknown status partner Gay community integraon Direct effect: β= 0.58 (.02, 1.13)** Indirect effect: β= 0.08 (.001, .26)** * p-value < .10; ** p-value < .05 β= 0.15 (-.02, 33)* 0.6 (0.3, 1.3) 1.6 (0.6, 4.4) 1.1 (0.5, 2.4) 1.3 (0.5, 3.1) 0.6 (0.3, 1.3) 2.1 (0.7, 5.8) 1.4 (0.6, 2.9) 1.2 (0.5, 2.7) 1.5 (0.6, 3.6) 0.4 (0.1, 1.4) 2.3 (0.9, 5.9) 0.8 (0.3, 2.3) OR odds ratio CI confidence interval b 1.6 (0.7, 3.8) 0.6 (0.2, 1.8) 2.7 (1.1, 6.8) 0.8 (0.3, 2.4) a 1.1 (1.0, 1.2) – 1.5 (0.5, 4.3) 0.4 (0.1, 1.3) 2.3 (0.9, 6.0) 0.8 (0.2, 2.7) 1.4 (0.6, 3.4) 0.6 (0.2, 2.0) 2.8 (1.1, 7.3) 0.7 (0.3, 2.2) 1.2 (1.01, 1.3) – – – – 1.7 (0.9, 3.3) – 1.1 (0.99, 1.3) 1.2 (0.98, 1.4) – – 1.1 (0.99, 1.3) – – – – – – 1.2 (0.9, 1.5) 1.1 (0.95, 1.3) 1.6 (1.01, 2.5) 1.6 (1.01, 2.5) 1.7 (1.1, 2.6) – 1.7 (0.97, 2.9) 1.7 (0.97, 2.9) 1.7 (1.01, 2.8) 1.7 (1.01, 3.0) 1.8 (1.02, 3.1) 1.8 (1.02, 3.1) OR (95% CI) Mediator: peer judgmentalism OR (95% CI) Mediator: HIV knowledge 1.8 (1.9, 3.5) – Integration into the gay community Direct effect Potential mediators Knowledge of HIV risk Peer judgmentalism towards sexual behavior Sexuality-related discrimination Indirect effect Covariates Age less than 25 years Any university education Employed In a committed relationship OR (95% CI) Mediator: HIV knowledge OR (95% CI) No mediator ORa (95% CIb) No mediator OR (95% CI) Mediator: sexuality-related discrimination Any HIV testing Any condomless anal sex with a partner whose HIV status is positive or unknown Table 3 Logistic regression analysis exploring potential mediators of the relationship between integration into the gay community and (1) condomless anal sex with partners whose HIV status is positive or unknown and (2) any HIV testing Int.J. Behav. Med. gay community, as greater integration increases access to information and services regarding HIV prevention that may be most easily obtained in the gay community. Greater knowledge of HIV risk was also correlated with having ever been tested for HIV, and it partially mediated the relationship between greater integration into the gay community and any history of HIV testing, suggesting the expected health benefit of greater knowledge. However, contrary to our hypothesis that increased knowledge would be protective and related to consistent condom use, greater knowledge of HIV risk was actually related to having had any condomless anal sex and it partially mediated the relationship between gay integration and condomless anal sex. These findings suggest that greater access to information about HIV may partly contribute to the negative impact of gay integration on condom use. One possible explanation for this counter intuitive relationship is that greater knowledge of HIV risk, including new ways of preventing HIV infection and the ability of HIV treatment to manage the disease as a chronic illness, may serve to lower concern about contracting HIV and the perceived need to use precautions to protect against HIV. Over the past two decades, civil society organizations in Lebanon, with the support of the National AIDS Program, have increased efforts to disseminate targeted messages for HIV prevention and increased the number of HIV testing sites. Knowledge of HIV, and exposure to voluntary counseling and testing, may be reducing the perceived threat of HIV. A number of studies of MSM in the USA have found positive beliefs about HIV treatment (treatment optimism) to be positively correlated with condomless anal sex and perceived peer norms related to condomless sex [26, 27]. Alternatively, greater knowledge of factors influencing HIV risk may serve to encourage some men to rely on methods other than condoms for reducing HIV risk, such as frequent HIV testing, knowing one’s HIV status and that of their partner(s), and use of PrEP (although this remains difficult and expensive to access in Lebanon), which could shift the discourse of unsafe sex into lower-risk condomless sex. A key limitation of this analysis is the cross-sectional nature of the data. The data provide evidence of association, but we cannot demonstrate or infer causation without longitudinal data. However, even with the longitudinal data that the study will collect it will remain difficult to examine causal mediation. In order to demonstrate causation, we would need to show that the processes of sexual identity development (e.g., integration into the gay community) and minority stressors were experienced prior to engagement in condomless anal sex or HIV testing. Additionally, we would want integration into the gay community to precede the mediator (e.g., acquiring knowledge of HIV risk), and this knowledge acquisition to precede the condomless anal sex or HIV testing, all of which would be very challenging to determine accurately. The relatively small sample size also limited our statistical power and Int.J. Behav. Med. this likely contributed to some of the mediators being only marginally associated with condomless anal sex and the indirect effects being marginally significant in our mediation analysis. Another limitation of our data is the representativeness of our sample. While long-chain peer referral recruitment is designed to penetrate all segments of a population, our sample lacked in representation of men who were less educated and not self-identified as gay; however, our sample had good diversity in terms of being well balanced on religious affiliation and the inclusion of refugees. There are also limitations related to measurement. The measure of condomless anal sex with partners whose HIV status was positive or unknown indicates engagement in sexual behavior that poses a risk for HIV transmission; however, we did not assess perception of viral load status among HIVpositive partners, so it is plausible that some reports of condomless sex with infected partners did not involve HIV risk (if the partner had an undetectable viral load). The item used to measure if anyone in the family is aware of the respondent’s sexual identity asked about the family’s knowledge of both the respondent’s sexual identity and sex with men, which are distinct from one another and thus make it impossible to know whether the respondent was answering with regard to one or the other or both. Several of our constructs were measured with single items, or items that had been developed by the study team and not validated in prior research, suggesting that our findings will need to be replicated with stronger measures in order to better establish the relationships studied in this analysis. Further, some of the scales (i.e., integration into the gay community and sexuality-related discrimination) had low internal reliability, which can attenuate observed correlations; the use of more reliable measures could alter the direction and magnitude of the associations found in our analysis. In conclusion, our findings highlight the potential sexual health benefits and vulnerabilities associated with increased integration into the gay community for young MSM living in a setting that is highly stigmatizing of homosexuality and how knowledge of HIV risk may play a key role in these relationships. The increased access to information about HIV risk and prevention that accompanies greater involvement in the gay community contributes to more HIV testing. However, greater involvement in the gay community and increased access to information about HIV risk and prevention resources were also associated with increased sexual risk taking, perhaps as a result of reduced concerns about the repercussions of HIV that may stem from knowledge of available, effective of HIV treatment, which in turn leads to increased sexual disinhibition. These findings imply the need to improve HIV prevention campaigns by adjusting sexual health messaging techniques in order to increase HIV knowledge and access to prevention tools while limiting their unintended tendency to increase sexual risk behavior. Funding Information Research reported in this publication was supported by the National Institute of Mental Health of the National Institutes of Health under Award Number R01MH107272 (PI: G. Wagner). Compliance with Ethical Standards Disclaimer The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Conflict of Interest The authors declare that they have no conflict of interest. Informed Consent All participants provided informed consent. Ethical Approval Ethics approval was provided by institution review boards at RAND and the Lebanese American University. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Wohl AR, Galvan FH, Myers HF, et al. Do social support, stress, disclosure and stigma influence retention in HIV care for Latino and African American men who have sex with men and women? AIDS Behav. 2010;15(6):1098–110. https://doi.org/10.1007/s10461-0109833-6. Ryan P. Integrated theory of health behavior change. Clin Nurse Spec. 2009;23(3):161–70. https://doi.org/10.1097/nur. 0b013e3181a423733. Fergus S, Lewis MA, Darbes L, Butterfield RM. HIV risk and protection among gay male couples: the role of gay community integration. Health Educ Behav. 2005;32:151–71. Meyer IH. Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: conceptual issues and research evidence. Psychol Bull. 2003;129(5):674–97. https://doi.org/10.1037/00332909.129.5.674. Meyer IH. Minority stress and mental health in gay men. J Health Soc Behav. 1995;36(1):38. https://doi.org/10.2307/2137286. Bogart LM, Wagner GJ, Galvan FH, Landrine H, Klein DJ, Sticklor LA. Perceived discrimination and mental health symptoms among Black men with HIV. Cult Divers Ethn Minor Psychol. 2011;17(3): 295–302. https://doi.org/10.1037/a0024056. Tao J, Wang L, Kipp AM, et al. Relationship of stigma and depression among newly HIV-diagnosed Chinese men who have sex with men. AIDS Behav. 2016;21(1):292–9. https://doi.org/10.1007/ s10461-016-1477-8. Preston DB, D’Augelli AR, Kassab CD, Starks MT. The relationship of stigma to the sexual risk behavior of rural men who have sex with men. AIDS Educ Prev. 2007;19(3):218–30. https://doi.org/10. 1521/aeap.2007.19.3.218. Hays RB, Paul J, Ekstrand M, Kegeles SM, Stall R, Coates TJ. Actual versus perceived HIV status, sexual behaviors and predictors of unprotected sex among young gay and bisexual men who identify as HIV-negative, HIV-positive and untested. AIDS. 1997;11(12):1495–502. https://doi.org/10.1097/00002030199712000-00014. Harper GW. Sex isn’t that simple: culture and context in HIV prevention interventions for gay and bisexual male adolescents. Am Psychol. 2007;62(8):806–19. https://doi.org/10.1037/0003-066x. 62.8.806. Walker JNJ, Longmire-Avital B, Golub S. Racial and sexual identities as potential buffers to risky sexual behavior for Black gay and Int.J. Behav. Med. 12. 13. 14. 15. 16. 17. 18. bisexual emerging adult men. Health Psychol. 2015;34(8):841–6. https://doi.org/10.1037/hea0000187. Mutchler MG, McDavitt BW, Tran TN, et al. This is who we are: building community for HIV prevention with young gay and bisexual men in Beirut, Lebanon. Cult Health Sex. 2017;20(6):690–703. https://doi.org/10.1080/13691058.2017.1371334. Kippax S, Connell RW, Dowsett GW, Crawford J. Sustaining safe sex: gay communities respond to AIDS. London: Falmer Press; 1993. Fergus S, Lewis MA, Darbes LA, Kral AH. Social support moderates the relationship between gay community integration and sexual risk behavior among gay male couples. Health Educ Behav. 2009;36(5):846–59. McDavitt B, Mutchler MG. “Dude, you’re such a slut!” barriers and facilitators of sexual communication among young gay men and their best friends. J Adolesc Res. 2014;29(4):464–98. https://doi. org/10.1177/0743558414528974. United Nations. The Joint United Nations Programme on HIV/ AIDS. Regional report: Middle East and North Africa. Geneva, Switzerland. 2016. Mahfoud Z, Afifi R, Ramia S, et al. HIV/AIDS among female sex workers, injecting drug users and men who have sex with men in Lebanon: results of the first biobehavioral surveys. AIDS. 2010;24(Suppl 2):S45–54. https://doi.org/10.1097/01.aids. 0000386733.02425.98. Heimer R, Barbour R, Khouri D, et al. HIV risk, prevalence, and access to care among men who have sex with men in Lebanon. AIDS Res Hum Retrovir. 2017;33(11):1149–54. https://doi.org/ 10.1089/aid.2016.0326. 19. O'Brien K, Wortman CB, Kessler RC, Joseph JG. Social relationships of men at risk for AIDS. Soc Sci Med. 1993;36(9):1161–7. https://doi.org/10.1016/0277-9536(93)90236-w. 20. Bingham T. Questionnaire from MyLife MyStyle Evaluation Study. Los Angeles: County Health Department; 2011. 21. Mutchler MG, McDavitt B. ‘Gay boy talk’ meets ‘girl talk’: HIV risk assessment assumptions in young gay men’s sexual health communication with best friends. Health Educ Res. 2010;26(3): 489–505. https://doi.org/10.1093/her/cyq069. 22. Kegeles SM, Hays RB, Coates TJ. The Mpowerment Project: a community-level HIV prevention intervention for young gay men. Am J Publ Health. 1996;86(8_Pt_1):1129–36. https://doi.org/10. 2105/ajph.86.8_pt_1.1129. 23. Hayes AF, Preacher KJ. Statistical mediation analysis with a multicategorical independent variable. Br J Math Stat Psychol. 2013;67(3):451–70. https://doi.org/10.1111/bmsp.12028. 24. Lohr SL. Sampling: design and analysis. Second ed. Boston: Brooks / Cole; 2010. 25. Institute SAS. The SAS System Version 9.4. [Computer Software]. 2013. 26. Crepaz N, Hart TA, Marks G. Highly active antiretroviral therapy and sexual risk behavior. JAMA. 2004;292(2):224. https://doi.org/ 10.1001/jama.292.2.224. 27. Peterson JL, Bakeman R. Impact of beliefs about HIV treatment and peer condom norms on risky sexual behavior among gay and bisexual men. J Commun Psychol. 2005;34(1):37–46. https://doi. org/10.1002/jcop.20082. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.