Men who have sex with men (MSM) are disproportionately affected by HIV/AIDS in the United States. According to estimates from the Centers for Disease Control and Prevention (CDC), MSM comprised the majority (53%) of the estimated 56,300 new HIV infections in 2006.1 Among MSM, black and Latino MSM represented 54% of new infections in 2006 and 69% of new infections among MSM between the ages of 13 and 29.2
One factor contributing to the disproportionate infection rates among black and Latino MSM is undiagnosed HIV infection, which is proportionally higher among black and Latino MSM compared with white MSM. A CDC study conducted in 2004-2005 in 5 US cities found that black and Latino MSM infected with HIV were more likely to be unaware of their infection (67% and 48%, respectively) compared with infected white MSM (18%).3 The follow-up study conducted in 2008 in 21 US cities found that 59% of black MSM, 46% of Latino MSM, and 26% of white MSM infected with HIV were unaware of their serostatus.4 The high prevalence of black and Latino MSM who are HIV-positive unaware increases the potential for HIV transmission to uninfected sex partners.5-7
Only a few studies of MSM have examined correlates of being HIV-positive unaware in adjusted analyses.8-10 Compared with self-reported HIV-negative MSM, MacKellar et al8 found that HIV-positive unaware MSM were more likely to be older (ages 23-29 vs. 15-22), to be a racial minority (black, Latino or mixed race), to have a high school education or less, to test infrequently for HIV infection, and to report a previous sexually transmitted disease (STD) diagnosis. In addition, HIV-positive unaware MSM were more likely than uninfected MSM to have a greater number of lifetime sex partners, to engage in unprotected anal intercourse (UAI), and to perceive themselves to be at moderate-to-high risk of being HIV positive.8 Another study examined correlates of undiagnosed infection separately for MSM in Los Angeles and San Francisco. In the Los Angeles sample, HIV-positive unaware MSM were more likely than uninfected MSM to be black, to have a high school education or less, to use marijuana, to have been tested for HIV a year or more before, and to attend public sex environments.9 In the San Francisco sample, crack cocaine use and attendance at adult bookstores were associated with being HIV-positive unaware.9 A more recent study comparing HIV-positive unaware MSM to MSM who were previously diagnosed with HIV found that undiagnosed infection was associated with being black or Latino, being bisexually active, and having greater internalized homophobia.10
Despite the disproportionate burden of HIV infection among MSM of color, no published studies have examined correlates of being HIV-positive unaware separately for black or Latino MSM. Existing studies that examine correlates of undiagnosed HIV infection in adjusted analyses either pool all MSM irrespective of race or ethnicity9 or only provide unadjusted correlates of undiagnosed HIV infection.4 In this analysis, we examined an array of demographic, behavioral, and psychological variables for associations with being HIV-positive unaware relative to being HIV negative among black MSM and Latino MSM.
The Brothers y Hermanos study, conducted from May 2005 through April 2006, recruited 2235 MSM (1081 Latino MSM from Los Angeles and New York City and 1154 black MSM from New York City and Philadelphia). Eligibility criteria included (1) identifying as male, (2) being 18 years of age or older, (3) reporting sex (oral, anal sex, or mutual masturbation) with a man in the past 12 months, and (4) being a resident of one of the recruitment sites. Participation was open to men of any HIV status.
Participants were recruited using respondent-driven sampling.11 Project staff recruited eligible men (seeds) from community-based organizations or venues frequented by MSM. Seed selection was informed by referrals from community advisory board members and through visits by project staff to community organizations and public venues frequented by MSM. We purposely selected seeds that varied in HIV status and age. The seeds (numbers ranging from 16 to 36 including a second wave of seeds per site) recruited other eligible individuals (up to 3) from their social networks. New enrollees recruited the next wave of participants, with the process continuing until at least 500 men per research site were enrolled. Participants earned $15 for each eligible person that they recruited. Additionally, all participants received $50 in compensation for completing the study session. More information on study procedures can be found elsewhere.12,13
The study was conducted in project offices in office buildings, community-based organizations, and community health centers. Participants completed an audio computer-assisted self-interview (ACASI), available in English and in Spanish, which was followed by HIV testing. Men who did not disclose to project staff that they had previously been diagnosed with HIV infection were tested using a rapid oral fluid HIV antibody test (OraQuick Advance; OraSure Technologies, Inc, Bethlehem, PA). Participants who obtained a “preliminary positive” rapid test result and participants who had been previously diagnosed provided a blood specimen for confirmatory testing through Western blot assay. The study protocol was approved by the Institutional Review Boards at CDC and at each study site.
Measures for the Analysis
The ACASI collected standard demographic information, such as age, personal annual income, education, sexual identity, health insurance status, number of lifetime HIV tests, and STD history. We also assessed whether participants had a current health care provider and had disclosed their homosexual behavior to their current doctor or health care provider. In addition, participants estimated the size of their social network of MSM (number of MSM that they knew personally and had seen in the past 6 months in their respective recruitment city). For Latino MSM, we asked which language (English or Spanish) they primarily used to read, think, or speak.
Multiple questions assessed sexual risk behavior. Participants reported whether they had engaged in UAI in the past 3 months with main and casual male partners and UAI in the past 3 months with an HIV-positive or unknown status partner. The context of recent UAI was examined with questions regarding whether participants engaged in UAI at a private residence (past 3 months), at sex venues (past 3 months), whether the last episode of UAI (no timeframe) was with a black or a Latino male partner, and whether drugs and alcohol generally, or amyl nitrites specifically, were used before or during the most recent UAI episode.
The assessment also included psychological variables. We asked participants about their perceived likelihood of testing HIV positive on the study test or testing HIV positive in their lifetime (1 = extremely unlikely to 10 = extremely likely). We also asked men about assortative sexual mixing beliefs, specifically how much having sex with men of the same race/ethnicity reduce a man's risk of getting HIV (1 = not at all to 4 = a lot).
The primary outcome variable was being HIV-positive unaware versus HIV negative. We operationalized HIV-positive unaware respondents as those men who tested HIV positive in our study but self-reported being HIV negative in the ACASI, men who did not obtain the results of their last HIV test, or men who had never been tested for HIV infection. Participants who tested HIV negative during the study were categorized as uninfected. Men who reported on the ACASI that they had previously been diagnosed as HIV-positive (all of whom were confirmed as positive during the study) were excluded from this analysis.
All analyses were conducted separately for black and Latino MSM. Associations between being HIV-positive unaware and demographic, behavioral, and psychological variables were assessed with Pearson χ2 tests (using Fisher exact test where expected cell counts were <5) for categorical variables and Wilcoxon rank sum tests for continuous variables (eg, perceived risk). Correlates that reached P ≤ 0.10 in bivariate analyses were analyzed in a multiple logistic regression model using backward elimination of variables that did not reach P ≤ 0.05 in the multivariate model. Adjusted odds ratios and 95% confidence intervals are presented for the black MSM model and the Latino MSM model. All analyses were conducted with unweighted data because the intent was to examine associations in the sample rather than develop population estimates. Predictor variables were tested for evidence of collinearity and model fit was evaluated using the Hosmer-Lemeshow goodness-of-fit test. All analyses were conducted with SAS version 9.0 (SAS Institute Inc, Cary, NC).
A total of 1208 MSM (597 black and 611 Latino) were included in the analysis. Of these men, 1070 (492 black and 578 Latino) participants tested HIV negative on the study rapid test. One hundred thirty-eight men (105 black and 33 Latino) were HIV-positive unaware (18% of black MSM and 5% of Latino MSM, P < 0.001). Black MSM recruited in New York City were more likely to be HIV-positive unaware than black MSM recruited from Philadelphia (24% vs. 13%, P < 0.001). Latino MSM recruited from Los Angeles were more likely to be HIV-positive unaware than Latino MSM recruited from New York City (10% vs. 3%, respectively; P < 0.001).
Table 1 displays the unadjusted comparisons between HIV-positive unaware and uninfected participants. Among black MSM, HIV-positive unaware men were more likely than uninfected men to have moderately higher income ($5000 to $19000 vs. <$5000), to identify as gay, to have health insurance, to have disclosed their sexuality to their health care provider, to have ever been diagnosed with an STD, and to report fewer lifetime HIV tests. Among Latino MSM, HIV-positive unaware men were more likely than uninfected men to be older, to have been born outside of the United States, to have less than a high school education, and less likely to identify as gay. In addition, HIV-positive unaware Latino MSM were more likely than uninfected Latino MSM to prefer communicating in Spanish and to have a smaller network of MSM friends or acquaintances.
Sexual behaviors differed by men's HIV status (Table 1). HIV-positive unaware black MSM were more likely than uninfected black MSM to have recently engaged in UAI with a main partner, casual partner, and an HIV-positive or unknown status partner; to have used substances (alcohol, drugs, amyl nitrite) during the most recent episode of UAI, and to have recently had UAI at a sex venue, respondent's home or sexual partner's home. HIV-positive unaware black MSM were also more likely than uninfected black MSM to report having sex with a black or Latino partner during the most recent UAI episode or to have used substances (alcohol, drugs, amyl nitrite) during the last UAI episode. Sexual behaviors were also generally greater among HIV-positive unaware Latino MSM but did not significantly differ from uninfected Latino MSM across each sexual risk behavior as was observed for black MSM.
Findings for the psychological variables were uniform for the black and Latino MSM samples. For both groups, HIV-positive unaware men were more likely than uninfected men to believe that having sex with men of the same race/ethnicity reduces the risk of contracting HIV and were more likely to perceive themselves at risk of testing HIV positive on the study test or during their lifetime.
Table 2 presents the findings from the multivariate analysis. Among Latino MSM, being HIV-positive unaware was associated with bisexual (vs. gay) identity, UAI with a casual male partner in the past 3 months, higher perceived risk of testing HIV positive during the study and a greater belief that sex with men of the same race/ethnicity reduces one's risk of HIV transmission. Among black MSM, being HIV-positive unaware was associated with moderately higher income ($5000 to $19000 vs. <$5000), gay (vs. heterosexual or bisexual) identity, having health insurance, disclosure of sexuality to a current health care provider, fewer lifetime HIV tests, higher perceived risk of testing HIV positive during the study, and a greater belief that sex with men of the same race/ethnicity reduces one's risk of HIV transmission. Although recruitment city and MSM network size did not qualify for the final backward elimination models, these variables were forced into the models to control for potentially small effects (Table 2). Recruitment city and network size remained nonsignificant (P > 0.16) in both models and, except for 1 variable in the Latino model, all covariates from the original black MSM and Latino MSM models remained significant.
Beliefs about assortative sexual mixing (eg, choosing sex partners of the same race/ethnicity) were associated with being HIV-positive unaware in our samples of black and Latino MSM. Although most black and Latino MSM in our sample did not subscribe to these beliefs, men who were HIV-positive unaware were more likely than uninfected men to believe that sex with partners of the same race/ethnicity reduces risk of HIV infection. Prior research has established that black and Latino MSM, and MSM of other races/ethnicities, tend to select partners of the same race/ethnicity.14,15 Besides physical attraction, it is natural that black and Latino MSM feel safest or more comfortable with men from their own communities.16 However, our findings suggest that some black and Latino MSM believe that choosing male sex partners of the same race/ethnicity may be an effective HIV risk-reduction strategy. This is concerning because intraracial partner selection combined with high background prevalence of undiagnosed positives place uninfected black and Latino MSM at greater risk for HIV infection than MSM in other racial/ethnic communities.6 This is partially underscored by our results where we found, in unadjusted analyses, that HIV-positive unaware black and Latino MSM were more likely than uninfected MSM to report partners of the same race/ethnicity during the most recent UAI episode. HIV prevention programs targeting black or Latino MSM must emphasize the benefits of semiannual HIV testing and educate men about the greater prevalence of HIV in their communities and the risk that this poses to them.
As in other studies,4,8 perceived risk for HIV infection was related to being HIV-positive unaware. HIV-positive unaware black and Latino MSM perceived themselves to be at higher risk for testing HIV positive during the study compared with uninfected MSM. These associations are consistent, given the greater behavioral risk profile among HIV-positive unaware men compared with uninfected men in our sample. Unlike previous studies,8 however, few sexual behavior variables in our sample were associated with undiagnosed HIV infection in adjusted analyses. Of the various sexual risk variables that were assessed in this study, only recent UAI with a casual male partner was independently associated with undiagnosed HIV infection among Latino MSM. Despite statistically significant associations between undiagnosed HIV infection across every sexual risk variable in bivariate analyses among HIV-positive unaware black MSM, none of these variables contributed to the final multivariate model. One reason may be that our sexual risk variables captured recent behaviors (past 3 months) rather than prior sexual risk behaviors that may have contributed to HIV infection. An alternative explanation is that sexual risk variables generally are poor predictors of HIV infection for black MSM. Evidence from multiple racial comparative studies have found that sexual risk is not associated with disproportionate HIV prevalence8 or HIV incidence17,18 among black MSM compared to other MSM. Our study extends previous findings and demonstrates that sexual risk may not contribute to a greater odds of being HIV-positive unaware (vs. being uninfected) among black MSM.
Our analysis also highlights possible missed opportunities in HIV testing for black MSM. Being HIV-positive unaware among black MSM in our sample was associated with having a moderately higher income, having health insurance access and a greater likelihood of disclosing one's sexuality to a health care provider. Despite greater access to care, black MSM with undiagnosed HIV infection reported fewer lifetime HIV tests than uninfected black MSM. Because some HIV-positive unaware black MSM in our sample perceived themselves to be at higher risk for acquiring HIV than uninfected black MSM, it is plausible that stigma, shame or fear may have discouraged HIV-positive unaware men from asking their health providers to be tested for HIV. Data from our study also suggest that health care providers may not uniformly offer HIV tests to at-risk individuals. Although some studies of MSM report that disclosing one's sexuality to a health care provider is associated with a greater likelihood of being offered an HIV test during a healthcare visit,19-20 other studies have found contrasting results. A small prospective study of MSM reported that few participants received an HIV test, despite engaging in recent sexual risk behavior and disclosing these sexual risks to their healthcare providers.21 Similarly, a recent publication of MSM who reported that their health care provider was aware of their homosexuality found that 40% were not offered an HIV test.22 Although we did not ask participants if they had discussed HIV testing with their health care providers, our data highlight possible missed opportunities to counsel and test undiagnosed black MSM who had disclosed their sexuality to a provider in localities (Philadelphia, PA; and New York, NY) where surveillance data have documented high rates of infection among black MSM.23,24 The US Preventative Services Task Force's clinical guidelines encourage HIV testing for populations at greater risk for HIV infection.25 Despite these guidelines, studies a decade apart continue to report that black MSM tend to be diagnosed later in their disease progression (within a year of an AIDS diagnosis) relative to MSM of other races/ethnicities.26,27
Missed opportunities to diagnose black MSM in our sample who already access health care not only translates into poorer health outcomes for these men, but also elevates the risk of HIV transmission to uninfected partners. A supplemental analysis revealed HIV transmission risks among the 44 HIV-positive unaware black MSM who had disclosed their sexuality to a current health care provider. Undiagnosed black MSM who disclosed their sexuality to their provider reported unprotected insertive anal intercourse with 9 HIV-negative male partners in the previous 3 months and unprotected receptive anal intercourse with 14 HIV-negative male partners during the same period. Receiving an HIV-positive diagnosis from a health care provider may have limited the possibility of such exposures to (presumably) uninfected partners.
Another noteworthy finding in our data is the association between sexual identity and being HIV-positive unaware. Among black MSM, a gay (vs. heterosexual/bisexual) identity was associated with a greater odds of undiagnosed HIV infection. This is consistent with prior studies showing that HIV infection is more likely among gay than nongay identified MSM.28,29 Interestingly, among the Latino MSM in our study, a bisexual (vs. gay) identity was associated with a greater odds of undiagnosed HIV infection. A prior study also found an increased risk of undiagnosed infection among behaviorally bisexual MSM pooled across race/ethnicity.11 One possible explanation for our Latino finding involves language barriers in HIV prevention messaging. In a supplementary analysis of our data, we found that bisexual-identified Latino MSM had a somewhat stronger preference for Spanish language than gay-identified Latino MSM (P = 0.04). HIV prevention information communicated in English may miss these bisexual/heterosexual men, leaving them at higher risk for HIV infection. Another possibility is that these men may have been less likely to attend to HIV prevention messages directed toward gay-identified Latino men. These hypotheses merit attention in future research.
The study has several limitations. First, the data are based upon self-reported risk behavior and there may be discrepancies with actual level of risk. Second, the data were obtained through social network-based sampling and may not represent the attitudes, behavior, or HIV status of black or Latino MSM in the metropolitan areas from which the samples were drawn. Our sample was primarily drawn from low-income (<$20,000 per year) MSM networks that may not be representative of black and Latino MSM in New York City, Los Angeles, or Philadelphia. Nevertheless, utilizing a network-based approach recruitment approach is integral to the development of future intervention studies for MSM of color because data suggest that sexual networks place men in these communities at higher risk for HIV infection.7 Third, our sample included small subsamples of HIV-positive unaware men and our findings—especially for Latino MSM—should be interpreted with caution. Fourth, some sexual risk variables only captured characteristics of the last sexual encounter and may not reflect enduring patterns of behavior. Last, our study is cross-sectional and causal inferences cannot be made. Future studies with stronger designs (eg, cohort studies) are needed to confirm and extend our findings.
It is estimated that undiagnosed HIV-positive individuals are responsible for approximately half of all sexually transmitted HIV infections in the United States.30 Black and Latino MSM are at disproportionate risk for HIV infection and more likely to be HIV-positive unaware compared with MSM of other races/ethnicities in the United States.3,4 Interventions that prompt health care providers to offer HIV tests to MSM are urgently needed, as are interventions that build upon previous qualitative research of clinical encounter experiences among MSM of color.31-33 Although prevention messages should advise black and Latino MSM about the influence of a high background prevalence and the relatively greater odds of exposure to HIV during UAI within assortative partnerships, discouraging intraracial mixing is neither practical nor desirable. Ultimately, scalable interventions that meaningfully reduce the likelihood of infection and that utilize existing sexual and social networks to fundamentally alter the risk environment for black and Latino MSM are needed. In the meantime, frequent HIV testing, timely and appropriate linkage to care, and provision of antiretroviral therapies for those diagnosed with HIV and those at high risk for HIV34,35 may be the best available methods to reduce background HIV prevalence and undiagnosed HIV infection among black and Latino MSM communities in the United States.
The authors thank the study participants, the collaborating community-based partners, and the Brothers y Hermanos research team.
1. Hall IH, Ruiguang S, Rhodes P, et al. Estimation of HIV
incidence in the United States. JAMA
2. Subpopulation estimates from the HIV
incidence surveillance system—United States, 2006. MMWR Morb Mortal Wkly Rep
3. Centers for Disease Control and Prevention. HIV
infection, and HIV
testing among men who have sex with men (MSM
)—five U.S. cities, June 2004-April 2005. MMWR Morb Mortal Wkly Rep
4. Centers for Disease Control and Prevention. Prevalence and awareness of HIV
infection among men who have sex with men—21 cities, United States, 2008. MMWR Morb Mortal Wkly Rep
5. MacKellar DA, Valleroy LA, Secura GM, et al. Unrecognized HIV
infection, risk behaviors and perceptions of risk among young men who have sex with men. J Acquir Immune Defic Syndr
6. Millett GA, Peterson JL, Wolitski R, et al. Greater risk for HIV
infection of black
men who have sex with men. A critical literature review. Am J Public Health
7. Millett GA, Flores SA, Peterson JP, et al. Explaining disparities in HIV
infection among black
and white men who have sex with men: a meta-analysis of HIV
risk behaviors. AIDS
8. Mackellar DA, Valleroy LA, Secura GM, et al. Perceptions of lifetime risk and actual risk for acquiring HIV
among young men who have sex with men. AIDS Behav
9. Raymond HF, Bingham T, McFarland W. Locating unrecognized HIV
infections among men who have sex with men: San Francisco and Los Angeles. AIDS Educ Prev
10. Young SD, Shoptaw S, Weiss RE, et al. Predictors of unrecognized HIV
infection among poor and ethnic men who have sex with men in Los Angeles. AIDS Behav
11. Heckathorn DD. Respondent-driven sampling: a new approach to the study of hidden populations. Soc Problems
12. Marks G, Millett GA, Bingham T, et al. Understanding differences in HIV
sexual transmission among Latino
men who have sex with men: the Brothers y Hermanos Study. AIDS Behav
13. Lauby JL, Millett GA, LaPollo AB, et al. Sexual risk behaviors of HIV
-negative, and serostatus-unknown black
men who have sex with men and women. Arch Sex Behav
14. Berry M, Fisher RH, McFarland W. Same race and older partner selection may explain higher HIV
prevalence among black
men who have sex with men. AIDS
15. Bingham TA, Harawa NT, Johnson DF, et al. The effect of partner characteristics on HIV
infection among African American men who have sex with men in the Young Men's Survey, Los Angeles, 1999-2000. AIDS Educ Prev
16. Wilson PA, Valera P, Ventuneac A, et al. Race-based sexual stereotyping and sexual partnering among men who have sex with men who use the Internet to identify other men for bareback sex. J Sex Res
17. Koblin BA, Husnik MJ, Colfax G, et al. Risk factors for HIV
infection among men who have sex with men. AIDS
18. Samuel M, Winkelstein W Jr. Prevalence of human immunodeficiency virus infection in ethnic minority homosexual/bisexual men. JAMA
19. Bernstein KT, Liu K, Begier EM, et al. Same-sex attraction disclosure to health care providers among New York City men who have sex with men: implications for HIV
testing approaches. Arch Intern Med
20. Wall KM, Khosropour CM, Sullivan PS. Offering of HIV
screening to men who have sex with men by their health care providers and associated factors. J Int Assoc Physicians AIDS Care
21. Mimiaga MJ, Goldhammer H, Belanoff C, et al. Men who have sex with men: perceptions about sexual risk, HIV
and sexually transmitted disease testing, and provider communication. Sex Transm Dis
22. Petroll AE, Mosack KE. Physician awareness of sexual orientation and preventive health recommendations to men who have sex with men. Sex Transm Dis
23. New York City Department of Health and Mental Hygiene. Press release: new HIV
diagnoses rising among young men who have sex with men. September 11, 2007. Available at: www.nyc.gov/html/pr2007/pr079-07.shtml
. Accessed January 20, 2011.
24. Philadelphia Department of Public Health. AIDS Activities Coordinating Office. Annual Surveillance Report. 2008. Available at: www.phila.gov/health/pdfs/Epi_Update_2008.pdf
. Accessed January 20, 2011.
25. U.S. Preventative Services Task Force. Screening for HIV
: recommendation statement. Ann Intern Med
26. Wortley PM, Chu SY, Diaz T, et al. HIV
testing patterns: where, why, and when were persons with AIDS tested for HIV
27. Hall I, Byers RH, Ling Q, et al. Racial/ethnic and age disparities in HIV
prevalence and disease progression among men who have sex with men in the United States. Am J Public Health
28. Ramirez-Valles J, Garcia D, Campbell RT, et al. HIV
infection, sexual risk behavior, and substance use among Latino
gay and bisexual men and transgender persons. Am J Public Health
29. Torian LV, Makki HA, Menzies IB, et al. HIV
infection in men who have sex with men, New York City department of health sexually transmitted disease clinics, 1990-1999: a decade of serosurveillance finds that racial disparities and associations between HIV
and gonorrhea persist. Sex Transm Dis
30. Marks G, Crepaz N, Janssen RS. Estimating sexual transmission of HIV
from persons aware and unaware that they are infected with the virus in the USA. AIDS
31. Siegel K, Raveis V. Perceptions of access to HIV
-related information, care and services among minority men. Qual Health Res
32. Wheeler DP. Working with positive men: HIV
prevention with black
men who have sex with men. AIDS Educ Prev
33. Malebranche DJ, Peterson JL, Fullilove RE, et al. Race and sexual identity: perceptions about medical culture and healthcare among men who have sex with men. J Natl Med Assoc
34. Grant RM, Lama JR, Anderson PL, et al. Preexposure chemoprophlaxis for HIV
prevention in men who have sex with men. N Engl J Med
35. Centers for Disease Control and Prevention. Antiretroviral postexposure prophylaxis after sexual, injection-drug use, or other nonoccupational exposure to HIV
in the United States. MMWR Morb Mortal Wkly Rep