|Home | About | Journals | Submit | Contact Us | Français|
We examined the associations between specific types and sources of discrimination and mental health outcomes among U.S. racial/ethnic minority men who have sex with men (MSM) and how these associations vary by race/ethnicity.
A chain-referral sample of 403 African American, 393 Asian and Pacific Islander (API), and 400 Latino MSM recruited in Los Angeles County, CA completed a standardized questionnaire.
Past-year experiences of racism within the general community and perceived homophobia among heterosexual friends were positively associated with depression and anxiety. Past-year homophobia experienced within the general community was also positively associated with anxiety. These statistically significant associations did not vary across racial/ethnic groups. The positive association of perceived racism within the gay community with anxiety differed by race/ethnicity, and was statistically significant only for APIs. Perceived homophobia within the family was not associated with either depression or anxiety.
Higher levels of experiences of discrimination were associated with psychological distress among MSM of color. However, specific types and sources of discrimination were differentially linked to negative mental health outcomes among African American, API, and Latino MSM.
Accumulating data indicate that mental health problems are more prevalent among sexual minorities than among heterosexuals.1,2 A recent meta analysis of 25 epidemiological studies revealed that the lifetime prevalence of depression and anxiety disorders was at least 1.5 times higher among lesbians, gays, and bisexuals, while the lifetime risk for suicide attempts was 2.47 times greater in sexual minority groups.2 Research has linked a wide range of mental health outcomes among sexual minorities – including depression,3–10 anxiety,6,9,10 panic disorder,6 psychological distress,11–16 suicide attempts,11 and suicidal ideation11,17 – to the effects of discrimination. Furthermore, studies have shown widespread experiences of discrimination among sexual minorities.6,8,16,18,19 One population-based survey of U.S. adults found that more than three quarters (76%) of gay or bisexual respondents reported a lifetime experience of discrimination.6 In another population-based survey of U.S. adults, close to one quarter (21.4%) of lesbian, gay, or bisexual respondents reported experiencing discrimination in the past year.18
Although the disproportionate prevalence of mental health difficulties as well as experiences of social discrimination among lesbian, gay and bisexual people are well documented, the association between discrimination and mental health outcomes is not well understood. Among the gaps in our understanding are the following two points. First, it is unclear which types and sources of discrimination have an adverse effect on mental health among sexual minorities. Three of the 15 existing studies on the impact of discrimination on mental health outcomes utilized composite measures of overall perceived discrimination that combined various forms of discrimination experiences.6,10,15 Eleven studies examined specific types of discrimination such as anti-gay violence,5,9,11,12,14,20 racism,8,16 and homophobia,3,4,8,12,13,16 but made no specific references to their sources (e.g., family, heterosexual friends, a mainstream gay community, a larger general community). One study specified only a single source (i.e., White gay men) for one particular form of discrimination (i.e., perceived devaluation of Asian gay men7).
Second, it remains unknown whether prior study findings on the association between discrimination and poor mental health outcomes among sexual minorities are generalizable across racial/ethnic groups, as few studies of sexual minorities have directly compared the effects of discrimination on mental health outcomes between various racial/ethnic minority groups. Studies of discrimination and mental health among sexual minorities have either aggregated multiple racial and ethnic groups3–6,9,11–14,20, or been limited to one (i.e., African Americans,10 Asians,8 Latinos16) or two groups (i.e., Asians and Latinos15).
The present study addressed these gaps in the literature by examining the associations of specific types and sources of discrimination with mental health outcomes in a sample of African American, Asian or Pacific Islander (API), and Latino men who have sex with men (MSM). First, we examined the prevalence of discrimination based on race/ethnicity and sexual orientation. Second, we determined which types and sources of discrimination were associated with mental health outcomes. Finally, we assessed whether these associations varied by race/ethnicity.
Data came from the Ethnic Minority Men’s Health Study, designed to examine the impact of experiences of social discrimination, social networks, and sexual partnerships on sexual risk for HIV among African American, API, and Latino MSM in Los Angeles, CA. Study participants were recruited from May 2008 to October 2009 utilizing a chain-referral sampling methodology. Initial “seed” respondents were recruited through referrals from project staff at AIDS Project Los Angeles as well as outreach activities at MSM venues such as bars, dance clubs, and coffee shops. Men were eligible to participate in the study if they: (1) self-identified as African-American, API, or Latino; (2) were at least 18 years old; (3) were residents of Los Angeles County; (4) had had sex with at least one man in the last six months; and (5) did not participate in prior phases of this study involving qualitative interviews and survey instrument pretesting.
After providing written informed consent, eligible seed participants completed a one-hour, standardized questionnaire using audio computer-assisted self-interview (ACASI) technology. Following this, each seed participant received $50 for compensation and was invited to recruit up to three other MSM friends or acquaintances. He was given three “recruitment coupons” to pass out to these contacts as an invitation to participate in the study. For every redeemed coupon, the referring participant received $10.
Once accepting the coupon from their seed participant, potential recruits contacted the study staff by telephone. During this telephone call they were screened for eligibility and, if eligible, were scheduled to complete the study protocol at our project site. All potential participants had to come to their appointment with a recruitment coupon. On the completion of the ACASI-based survey, enrolled participants (“Wave One recruits”) were given three recruitment coupons each and instructed to give these coupons to their MSM friends or acquaintances (“Wave Two recruits”). This recruitment and enrollment process continued until we reached our target sample size of approximately 400 for each ethnic group.
We developed a culturally sensitive and psychometrically valid quantitative survey instrument by utilizing qualitative data collected from six focus groups (n=50) and 35 individual in-depth interviews for the previous phase of this study and through extensive pretesting with a sample of 168 MSM of color (56 African Americans, 56 APIs, and 56 Latinos). First, we modified existing measures (e.g., racism, homophobia)16 based on qualitative interview transcript data to ensure that measures used were consistent with our target population’s subjective experiences and phraseology. Initial drafts of the measures were then pre-tested on 18 participants (6 per ethnic group), using experimentally validated pretest procedures.21 After revising the draft measures based on respondent feedback, the revised measures were administered via ACASI to 150 participants (50 per ethnic group) to assess their psychometric properties. Data collected from these 150 participants were analyzed and used to make final refinements to the instrument.
Respondents were asked about their race/ethnicity, age, level of education, nativity, sexual orientation, marital status, any lifetime history of incarceration (juvenile or adult), and HIV serostatus.
We assessed subjective experiences of racism from two different sources: the general and gay communities. Past-year experiences of racism within the general community were measured using a four-item summative index, on which respondents indicated the number of different ways they had been mistreated due to their race/ethnicity within the general community in the past year (e.g., stopped or harassed by the police; physically threatened). Perceived racism within the gay community was measured with a four-item scale (e.g., I’ve felt white gay men have acted as if they’re better than me because of my race or ethnicity; I’ve felt ignored or invisible where white gay men hang out because of my race or ethnicity; Cronbach’s alpha = 0.86). Response to each item was scored on a four-point Likert scale (1 = “strongly disagree,” 2 = “somewhat disagree,” 3 = “somewhat agree,” 4 = “strongly agree”). Response scores were averaged to create the perceived racism within the gay community scale.
We assessed subjective experiences of homophobia from three different sources: the general community, heterosexual friends, and the family. Past-year experiences of homophobia within the general community was measured with a four-item summative index that represented the number of ways respondents had experienced various forms of such mistreatment (e.g., stopped or harassed by the police; physically threatened). Perceived homophobia among heterosexual friends and perceived homophobia within the family were measured with, respectively, a three-item scale (e.g., I have lost heterosexual friends because of my sexual orientation; Cronbach’s alpha = 0.87), and a four-item scale (e.g., Family members have pressured me at different times to get married to a woman; Cronbach’s alpha = 0.82), respectively. Each item used a four-point Likert scale ranging from 1 (strongly disagree) to 4 (strongly agree). Scores were averaged across each set of items to form the perceived homophobia among heterosexual friends and homophobia within the family scales.
We assessed two mental health outcomes: depression and anxiety. Depression was measured by using the seven-item, short-form version of the Center for Epidemiological Studies Depression Scale (CES-D).22 This scale asked about the number of days respondents experienced each of the seven depressive symptoms in the previous week (e.g., feeling lonely, feeling sad; Cronbach’s α=.91) on a four-point response set (1 = less than one day; 2 = 1–2 days; 3 = 3–4 days; 4 = 5–7 days). Anxiety was measured with the six-item Anxiety subscale of the Brief Symptom Inventory.23 This scale asked about the level of discomfort respondents felt about each of six anxious states in the previous week (e.g., feeling fearful, spells of terror or panic; Cronbach’s α=.93) on a five-point response set (0 = no discomfort; 1 = a little bit of discomfort; 2 = moderate discomfort; 3 = quite a bit of discomfort; and 4 = extreme discomfort). Response scores were averaged to create the depression and anxiety scales.
Descriptive analyses included means and proportions of demographic, discrimination, and mental health variables, both in the overall sample and within each racial/ethnic group. Multivariate analyses of mental health began with the full set of demographic and discrimination variables listed in Tables 1 and and2,2, and the interactions between each discrimination variable and race/ethnicity. Backward elimination removed all main effects with p > 0.20 and interaction terms with p > 0.05. To account for clustering of respondents in the same referral chain we used the GEE modeling framework24 for the multivariate analyses and to obtain the p-values for the descriptive statistics.
We enrolled a total of 1196 participants. Of these participants, 453 were seeds, 722 were recruited by seeds (170 in wave one, 126 in wave two, 105 in wave three, 61 in wave four, 56 in wave five, 49 in wave six, 40 in wave seven, 38 in wave eight, 39 in wave nine, 14 in wave ten, and 24 in waves 11 to 14), and 21 had no information about their recruitment status. The mean, median, minimum, and maximum number of participants recruited by the seeds was 1, 0, 0, and 125, respectively.
The sample consisted of an approximately equal number of African American (N=403), API (N=393), and Latino men (N=400). As Table 1 shows, sample characteristics differed by race and ethnicity. African Americans tended to be older than APIs and Latinos. APIs were more likely to have a college degree than were African Americans and Latinos. APIs and Latinos were more likely to be foreign born than were African Americans. African Americans were more likely to self-identify as bisexual than were APIs and Latinos. African Americans were more likely to have ever been married to a woman than were APIs and Latinos. African Americans and Latinos were more likely to have a lifetime history of incarceration than were APIs. African Americans had the highest HIV infection rate, followed by Latinos and APIs. The level of depression was similar across three racial/ethnic groups. However, the level of anxiety was the highest among Latinos, followed by APIs and African Americans.
There were significant racial/ethnic group differences in experiences of racism and homophobia (Table 2). African American and Latino MSM were more likely to report past-year experiences of racism within the general community than were API MSM. By contrast, API MSM were more likely to report racism within the gay community than were African American and Latino MSM. Past-year experiences of homophobia within the general community were most commonly reported by Latino MSM, followed by African American and API MSM. We found no statistically significant racial/ethnic differences in perceived homophobia among heterosexual friends and within the family.
Table 3 reports results of bivariate and multivariate regression analyses of the associations between discrimination experiences and depression and anxiety. Bivariate results showed that all five discrimination measures were associated with depression and anxiety. However, in multivariate analysis using backward elimination, only past-year experiences of racism within the general community and perceived homophobia among heterosexual friends remained as significant factors associated with depression. Depression was positively associated with past-year experiences of racism within the general community (b = 0.11, 95% CI = 0.07, 0.16, p < 0.001) and perceived homophobia among heterosexual friends (b = 0.11, 95% CI = 0.06, 0.16, p < 0.01). None of the interaction terms between race/ethnicity and the five discrimination measures was retained in the final multivariate model.
The multivariate model of the anxiety outcome suggested significant, positive main effects of past-year experiences of racism within the general community, past-year experiences of homophobia within the general community, and perceived homophobia among heterosexual friends (b = 0.08, 95% CI = 0.02, 0.13, p < 0.01; b = 0.06, 95% CI = 0.01, 0.11, p < 0.05; and b = 0.11, 95% CI = 0.06, 0.17, p < 0.001; respectively). In addition, the interaction of race/ethnicity with perceived racism within the mainstream gay community was statistically significant (p < 0.01). The positive association between anxiety and perceived racism within the gay community was found among APIs (b = 0.19, 95% CI = 0.08, 0.30, p < 0.001), but not among African Americans (b = −0.02, 95% CI = −0.10, 0.06, p = 0.5873) or Latinos (b = −0.02, 95% CI = −0.13, 0.09, p = 0.7240).
Experiences of discrimination based upon race/ethnicity and sexual orientation were highly prevalent in our sample of MSM of color in Los Angeles, CA. We also found similarities and differences in experiences of discrimination by race/ethnicity. The report of experiencing homophobia among heterosexual friends and within the family was similar among African American, API, and Latino MSM. However, the report of experiencing racism within the general community was more common among African Americans and Latinos than among APIs, while the reverse trend was observed with more APIs reporting experiences of racism within the mainstream gay community as compared to African Americans and Latinos. The report of experiencing homophobia within the general community was highest among Latinos.
Prior studies have documented the significant association between racism and mental health outcomes among U.S. API and Latino MSM.7,8,15,16 Our data indicate that experienced racism within the general community may have generalized adverse effects on psychological well-being among not only API and Latino MSM but also African American MSM. Our multivariate analysis found that past-year experiences of racism within the general community were positively associated with both depression and anxiety, but these associations were not moderated by race/ethnicity. Regardless of their race/ethnicity, those who were more likely to experience racism within the general community reported more symptoms of both depression and anxiety.
By contrast, racism within the mainstream gay community appeared to have negative psychological consequences only for API MSM in our sample. Our multivariate analysis found a significant positive association between perceived racism within the gay community and symptoms of anxiety, but this association was moderated by race/ethnicity. Perceived racism within the gay community was positively associated with anxiety among API MSM, but not among African American and Latino MSM. This finding is not surprising, given that a prior study of U.S. Asian gay men found higher levels of depression among those who were more likely to perceive devaluation of Asian gay men by White gay men.7 One explanation for the racial/ethnic difference with respect to the impact of racism within the gay community on psychological well-being may be related to a way in which racism manifests itself within the gay community. Our previous qualitative study of MSM of color in Los Angeles found that there were differential values placed on various racial/ethnic groups as potential sexual partners. In this tacit race/ethnicity-based sexual hierarchy, Whites are most preferred, followed by Latino and African American men, with API men as least preferred.25 Although all three groups of MSM of color experience some form of racism within the gay community, and some forms of race-based sexual objectification, APIs may face particular difficulties in finding sexual partners among those men whom they find desirable.26–28 These specific experiences of rejection or being devalued may lead to psychological distress.
In this study, we examined whether three different sources of homophobia – the general community, heterosexual friends, and the family – all affected psychological well-being among MSM of color. Our multivariate analyses found that past-year experiences of homophobia within the general community were positively associated with anxiety symptoms, whereas perceived homophobia among heterosexual friends was positively associated with both depressive and anxiety symptoms. Homophobia within the family, on other hand, was not associated with either depression or anxiety. We also examined whether race/ethnicity moderated the significant associations of homophobia within the general community and among heterosexual friends with mental health outcomes. We found no moderation by race/ethnicity. Taken together, our data indicate that homophobia experienced within the general community and among heterosexual friends both appear to have a negative effect on psychological well-being of all African American, API, and Latino MSM, but homophobia within the family seems to have little influence on their mental health. The reason why we found no effects for homophobia within the family may be that the psychological distress caused by family rejection of homosexuality is compensated by the benefits offered by the family such as emotional and financial support around matters not related to sexuality. Another possible explanation may be that those experiencing the greatest levels of family rejection cease contacts with their family, and so reduce its impact. These hypotheses should be examined in future studies.
Anti-gay stigma, harassment, and violence have been shown to have a variety of deleterious mental health consequences among sexual minorities.5,9,11,12,14,20 Consistent with these findings, our bivariate analyses found the significant positive association of past-year experiences of homophobia within the general community with depression and anxiety. However, when the multivariate model adjusted for demographic characteristics and included all measures of racism and homophobia, our measure of homophobia within the general community (assessing anti-gay harassment and threats) was significantly associated with anxiety but not with depression. In this model, the significant level for the association between past-year experiences of homophobia within the general community and anxiety was reduced from a p-value less than 0.001 to a p-value less than 0.05. By comparison, the significant association of perceived homophobia among heterosexual friends with depression and anxiety found in bivariate analyses remained unchanged in multivariate models. The magnitude of the association of perceived homophobia among heterosexual friends with depression and anxiety was greater than that of the association of past-year experiences of homophobia within the general community with depression and anxiety. These results suggest that for MSM of color, perceived homophobic disapproval and rejection by an immediate circle of heterosexual friends is more detrimental to psychological well-being than experiences of sexual orientation-related harassment and threats by members of the general community.
We should note three study limitations. First, the study results have limited generalizability. Participant recruitment took place in Los Angeles County which has a relative high concentration of API and Latino residents as well as an established mainstream gay community. The experiences of MSM of color might differ in other areas without such large racial/ethnic and sexual minority communities. Also, the sample was overrepresented by men who were HIV positive because many of the initial seeds used for recruitment were HIV-positive men who used services of our collaborating agency (i.e., AIDS Project Los Angeles). HIV infection rates reported by our study participants were higher than those estimated by the Los Angeles County Department of Public Health, HIV Epidemiology Program.29 More studies with MSM of color are needed to validate our study findings. Second, self-report measures have their limitations with respect to accuracy. We might have underestimated particularly the prevalence of past-year experiences of racism and homophobia within the general community because some participants might have difficulty in accurately recalling all the events that had happened during a one-year period. Third, the cross-sectional nature of our study does not enable us to draw causal inferences for the effects of experienced discrimination on mental health outcomes. Because we are uncertain about the temporal order of our measures, it is possible that depressed and/or anxious individuals were more likely to report experiences of discrimination. Future studies should use a longitudinal design to examine the causal links between experienced discrimination and negative psychological consequences.
Despite these limitations, our data have several implications for further research and intervention development with respect to specific types and sources of discrimination differentially linked to negative mental health outcomes among African American, API, and Latino MSM.
Further research can both tell us much more about both mediating and moderating factors between such stresses and such mental health outcomes. For example, how do different sources of racism influence internalized racism for racial/ethnic minority MSM? How do different sources of sexual orientation-based discrimination influence internalized homophobia for racial/ethnic minority MSM? Do different patterns of internalized stigmatization by racial/ethnic minority MSM differentially affect their awareness of sources of social and institutional support? Do they affect the utilization of those sources of support? How might the capacity to tap various sources of support or affirmation influence self-esteem and resilience? Are there resources within the communities of racial/ethnic minority MSM which could be developed or expanded to enhance the resilience of racial/ethnic minority MSM to address this problem on a larger institutional or community level (e.g., Davis, Cook & Cohen, 200530)?
While the present study did not focus on specific clinical dynamics, it suggests some practical considerations with respect to the evaluation of and scope of services for racial/ethnic minority MSM reporting emotional distress to health providers. It suggests the value of expanding the exploration of potential sources of stress to encompass an examination of issues of race-based and sexual orientation-based discrimination and victimization. It would appear to be helpful to assess such experiences not only on a general level, but with respect to family, friendship networks, and contacts with the mainstream gay community. Sources of support in managing such stressors would also be part of such an assessment. Moving from assessment of the individual’s history to his understanding of these experiences, and the links between these experiences and his own particular symptoms might be a first step of an intervention. As noted in the discussion above of research needs, it may be valuable to explore and address the ways in which these forms of stigmatization may result in internalized racism and homophobia. In this regard, assessment might be said to be shaped by systemic psychotherapy theory in broader respects,31 and consistent with minority stress theory.1,12 Further research will help to shape specific intervention strategies which might draw upon existing intrapersonal, social, and community resources.
Mental health services still carry a certain stigma32; thus it may potentially be helpful to consider offering services not explicitly identified as psychological in nature (such as group discussions) for those who have experienced discrimination or victimization due to race or sexual orientation. Providing the opportunity to talk about these matters with others who have undergone comparable experiences provides the opportunity to gain support, examine the impact of such experiences, and potentially seek out further support services if necessary.
In addition, community-based organizations that serve gay, lesbian, bisexual and transgender populations may wish to consider the value of interventions which both raise awareness about the damaging effects of discrimination on racial/ethnic minority MSM and work to shift social norms. Educational and social marketing campaigns have heightened awareness about homophobia, not only in the general community but in communities of color (e.g., the New York State Black Gay Network’s “We Are Part of You” campaign; see www.wearepartofyou.org). Creating comparable campaigns that target racism within the mainstream gay community and foster a sense of inclusiveness have the potential for significant impact on the well-being of gay and bisexual racial/ethnic minority men.
This study was supported by the U.S. National Institute of Mental Health grant R01MH069119.
Contributor StatementK. Choi developed the research question, participated in the data analysis and interpretation, and wrote the draft of the manuscript. J. Paul, R. Boylan, and S.E. Gregorich participated in the data analysis and interpretation and contributed to the writing of the manuscript. G. Ayala reviewed and provided feedback on the draft manuscript.
Human Participant Protection
The study procedures were approved by the Committee for Human Research of the University of California, San Francisco and by the Institutional Review Board of AIDS Project Los Angeles.