Diversity Network Committee Symposium

(2-023) DIVERSITY NETWORK COMMITTEE SYMPOSIUM, Seacliff B

HIV Prevention among Gay Men of Color

Chair:  Phillip W. Graham, PhD, RTI International

Discussant:  David Cordova, PhD, University of Michigan

Presenters:  George Ayala, PhD, The Global Forum on MSM & HIV (MSMGF), Keletso Makofane, PhD, The Global Forum on MSM & HIV (MSMGF), Glen-Milo Santos PhD, University of California, San Francisco and Department of Public Health, San Francisco, Sonya Arreola, PhD, RTI International

Understanding HIV Service Barriers and Facilitators among Men who have Sex Worldwide Utilizing Community-Based Survey Methods

George Ayala, The Global Forum on MSM & HIV (MSMGF), Sonya Arreola, Urban Health Program, RTI International, Jack Beck, The Global Forum on MSM & HIV (MSMGF), Tri D. Do, University of California San Francisco, Center for AIDS Prevention Studies, Pato Hebert, RTI International, Keletso Makofane, RTI International, Glen-Milo Santos, University of California, San Francisco and Department of Public Health, San Francisco, Patrick A. Wilson, Mailman School of Public Health, Columbia University

Gay men and other men who have sex with men (MSM) are disproportionately impacted by HIV.  Effectively addressing the sexual health needs of MSM is complicated by the persistent stigma, discrimination and violence they face.  Moreover, bio-medicalization of the HIV response has resulted in lopsided research and practice agendas that privilege individual-level clinical interventions over those that can target factors at the community, social, and structural levels. The role of community is often narrowly pre-determined, undervalued, or otherwise tokenized. In response to these challenges, the Global Forum on MSM & HIV (MSMGF) adopted a community-based participatory action research (CBPAR) approach in implementing the Global Men’s Health and Rights survey (GMHR).  We recruited a cross-sectional sample of MSM from 165 countries (n=5770) to examine factors associated with access to traditional HIV-related services. HIV prevalence in our sample was 18%. Condoms, HIV testing, and lubricants were highly accessible to 35%, 36%, and 21% of respondents, respectively. Less than half (42%) of HIV-positive respondents reported they had easy access to antiretroviral therapy. Access to condoms, lubricants, HIV testing, and HIV treatment was significantly lower in low income countries compared with high income countries. In multivariable regression models, homophobia, provider stigma and negative consequences for being out as MSM were significantly associated with reduced access to services. Conversely, community engagement and comfort with health service providers were associated with increased access. In addition to scaling up HIV prevention interventions tailored to the needs of MSM, it is imperative to both remove barriers to and support facilitators of HIV service access.  

Predictors of Survey Drop-out in a Global Survey of Diverse MSM

Keletso Makofane,Glen-Milo Santos, Sonya Arreola, Patrick Wilson, Jack Beck, George Ayala

Introduction: Internet-based HIV behavioral research promises to engage men who have sex with men (MSM) for whom homophobia limits the possibility of participating in in-person research. Previous studies of MSM have shown that 30-40% of respondents to online surveys drop out before answering the last question. Drop out is selective and thus threatens the validity of study findings by introducing selection bias.

Using the Gay Men’s Health and Rights Survey, a data set of 5,779 MSM from 165 countries conducted in 6 languages, we aimed to: a.) Measure the overall rate of survey drop-out, and  b.) Estimate relationships of survey language and country income with survey drop-out.

Methods: Survival Analysis methods were applied to investigate drop-out among respondents who did not report HIV positive status (N=5076). The outcome of interest was the number of questions answered before drop-out. Cox proportional hazards models were fitted to calculate adjusted hazard ratios (HR) for predictors of interest, including survey language and country income level. We adjusted for previously identified confounders – age, personal income, and education.

Results: Overall, 38% of respondents dropped out of the survey. Most (~90%) of these dropped out before answering half of the 142 questions. Adjusting for confounders, country income (p<0.001) and survey language (p<0.001) were significantly associated with drop-out. The hazard of drop-out was inversely related with country income level—compared to participants in high-Income countries, the hazard was higher among those from low-, lower middle-,  and upper middle-income countries (HRs= 1.81, 1.26, 1.15). Compared to respondents in English, Chinese and Georgian respondents had greater hazard rates for survey drop-out (HRs= 1.23, 1.72), while Spanish, French, and Russian respondents had lower hazard rates (HRs=0.92, 0.90, 0.65).

Conclusions: The proportion of respondents who dropped out before the end of the survey was consistent with previous research. Differences in hazard rate by country income could be driven by internet access and connection speeds, and differences by language could be driven by different reading comprehension levels demanded by translations of the original English survey instrument. Global surveys should be designed to be taken using low-speed internet or using multiple modalities, and should be piloted in all languages to ensure consistent reading comprehension levels.

The overall hazard for drop out was markedly higher for earlier questions in the survey than for later questions. This finding suggests that beyond a certain point, increasing the length of a survey only marginally decreases the proportion of respondents who will complete it.

Syndemic Conditions Associated with Increased HIV Risk in Global Sample of MSM

Glenn-Milo Santos Keletso Makofane, Tri Do, Sonya Arreola, Patrick Wilson, Jack Beck, Thomas Pyun, Pato Hebert, George Ayala

We evaluated the relationship between syndemic conditions—defined as a cluster of interconnected psychosocial health conditions—and sexual risk behaviors and HIV prevalence among MSM in the 2012 Gay Men’s Health and Rights survey. In this survey, we assessed the syndemic conditions of depression, substance use, violence, and external homophobia, adjusting for age and region. We observed a significant dose-response relationship between increasing number of syndemic conditions and having unprotected anal intercourse (UAI) (p<0.001). Compared to those who did not report syndemic conditions, the adjusted odds ratio (AORs) of UAI among those with 1, 2, or 3 or greater conditions were 1.4 (95% CI 1.2-1.7), 1.7 (1.4-2.1), and 2.3 (1.6-3.3), respectively. We observed a parallel dose-response association with number of conditions and HIV-status (p<0.001). Compared to those without syndemic conditions, the AORs for being HIV-positive among those with 1, 2, or 3 or greater conditions were 1.8 (1.4-2.2), 2.2 (1.8-2.9), and 2.5 (1.7-3.8), respectively. In this sample, we found evidence of intertwining syndemics that may be synergistically operating to drive the HIV epidemic among MSM. To effectively curb HIV and advance MSM health, it is imperative to concurrently address these multiple conditions and ameliorate social inequities. 

Structural, Community and Individual Factors Related to Sexual Happiness: Sexual Health among a Global Community of MSM

Sonya Arreola, Keletso Makofane, Glen-Milo Santos, Jack Beck, Pato Hebert, Patrick A. Wilson, George Ayala

Background: Sexual health of gay and other men who have sex with men has predominately focused on predictors of biological health outcomes, including HIV.  Although sexual happiness is an important indicator of sexual health research on sexual happiness is limited.

Methods: In April 2012, a convenience sample of 5779 self-identified gay/MSM was recruited to complete a 30-minute global online Gay Men’s Health and Rights survey. Factor analyses of sexual happiness items revealed two subscales: satisfaction with quantity, and satisfaction with quality of sex. We used multivariable linear regression to explore relationships between structural-, community- and individual-level variables, and the quantity and quality of sexual happiness. All scales ranged from 1 to 5. 

Results: Homophobia (ß= -0.12 95%CI: [-0.17; -0.08]), internalized homophobia (ß= -0.29 [-0.34; -0.25]), and psychological distress (ß= -0.40 [-0.44; -0.36]) were negatively related with the quantity subscale of sexual happiness; whereas community engagement (ß= 0.27 [0.21; 0.33]) was positively related with the quantity subscale of sexual happiness. Similarly, homophobia (ß= -0.05 [-0.09; -0.02]), internalized homophobia (ß= -0.21 [-0.24; -0.18]), and psychological distress (ß= -0.25 [-0.30; -0.21]) were negatively related with the quality subscale of sexual happiness; whereas community engagement (ß= 0.15 [0.10; 0.20]) was positively related with the quality subscale of sexual happiness.

Conclusions: Consistent with their negative impacts on other indicators of sexual health such as HIV risk, homophobia, internalized homophobia and psychological distress appear to have a negative effect on sexual happiness; and community engagement appears to enhance sexual happiness. These findings indicate that further research should consider the subjective experiences of men’s sexual happiness as an indicator of their sexual health.