Authors: Gabriel G. Edwards (a.Department of Medicine, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA, United States; b.Center for HIV Identification, Prevention, and Treatment Services, University of California, Los Angeles, Los Angeles, CA, United States), Brandon Moghanian (a.Department of Medicine, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA, United States), Cathy J. Reback (b.Center for HIV Identification, Prevention, and Treatment Services, University of California, Los Angeles, Los Angeles, CA, United States; c.Friends Research Institute, Los Angeles, CA, United States; d.Center for Behavioral and Addiction Medicine, Department of Family Medicine, University of California, Los Angeles, United States), Katrina M. Schrode (e.Department of Psychiatry, Charles R Drew University, Los Angeles, CA, United States), Robert E. Weiss (b.Center for HIV Identification, Prevention, and Treatment Services, University of California, Los Angeles, Los Angeles, CA, United States; f.Department of Biostatistics, Fielding School of Public Health, University of California, Los Angeles, Los Angeles, CA, United States), Nina T. Harawa (a.Department of Medicine, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA, United States; b.Center for HIV Identification, Prevention, and Treatment Services, University of California, Los Angeles, Los Angeles, CA, United States; e.Department of Psychiatry, Charles R Drew University, Los Angeles, CA, United States)
Categories: Article, PrEP, men who have sex with men, sexual identity, incarceration, internalized homophobia
Source: AIDS care
Authors: Gabriel G. Edwards, Brandon Moghanian, Cathy J. Reback, Katrina M. Schrode, Robert E. Weiss, Nina T. Harawa
This study compares pre-exposure prophylaxis (PrEP) use among men who have sex with men (MSM) with a recent history of incarceration, across various factors known to contribute to HIV transmission risk, including sexual identity, race/ethnicity, sexual activity, incarceration history, injection drug use, and internalized homophobia. We analyzed baseline lifetime PrEP use (yes or no) of 170 male-identifying participants enrolled in a randomized-controlled trial in Los Angeles County between 2019 and 2022. Using logistic regression, we assessed the association of PrEP with sexual identity, socio-demographics, and potential confounders. Compared to gay/same-gender loving-identified participants, straight/heterosexual-identified (aOR=0.10, CI=0.02–0.49) and bi-pansexual-identified (0.39, 0.16–0.95) participants had reduced odds of PrEP use. Black/African American-identified participants (0.15, 0.03–0.78) had lower odds of PrEP use than White-identified participants. Participants reporting 3+ years cumulative lifetime incarceration had lower odds (0.28, 0.09–0.87) of PrEP use than participants reporting <6 months. Controlling for internalized homophobia rendered differences among sexual identity groups non-significant. A similar effect was not observed for race/ethnicity and lifetime incarceration. Internalized homophobia was an important driver of PrEP use differences among MSM along the lines of sexual identity but not along the lines of race/ethnicity or cumulative incarceration.
In 2021, men accounted for 79% of new HIV cases in the United States. The most common risk factor for HIV transmission is male-to-male sexual contact, present in 67% of all cases of people 13 and older.(“Diagnoses of HIV Infection in the United States and Dependent Areas, 2021,” 2023) Daily oral pre-exposure prophylaxis (PrEP) for HIV prevention was approved by the United States Federal Drug Administration in 2012. PrEP, when taken as prescribed, is highly effective at preventing HIV transmission among men who have sex with men (MSM).(Grant et al., 2010) Though PrEP use has increased among MSM,(Kamitani et al., 2020; Sun et al., 2022) only a minority of those with indications for PrEP take it.(Weiss, Prasad, Sanchez, Goodreau, & Jenness, 2021) Furthermore, disparities in PrEP use are well documented, with lower use among Black and Latino with indications for PrEP, including MSM in particular.(“HIV in the United States by Race and Ethnicity: PrEP Coverage,” 2023; Kamitani et al., 2020)
Previous studies have examined MSM attitudes towards PrEP.(Kamitani et al., 2020; Sun et al., 2022) MSM is a label rooted in a behavior (sex with men) and not a sexual identity. Similarly, research on MSM often lumps together different sexual identity groups. Gay and bisexually identified MSM report that PrEP users are perceived by others as more likely to engage in high-risk behaviors, and this perception hinders the willingness of some individuals in these populations to initiate, adhere to, and disclose PrEP use.(Peterson, Nowotny, Dauria, Arnold, & Brinkley-Rubinstein, 2019; Sullivan, Mena, Elopre, & Siegler, 2019; Textor & Schlesinger, 2021) Gay and bisexual MSM who identify as Black or Latinx experience increased barriers to PrEP use due to sociostructural barriers such as high cost, medical mistrust, difficult-to-navigate health care systems, and lack of awareness or access to local, state, and national PrEP resources that support PrEP use and persistence.(Brantley et al., 2019; Nieto, Brooks, Landrian, Cabral, & Fehrenbacher, 2020; Sullivan et al., 2019) Internalized homophobia among MSM has been studied as a factor influencing PrEP use, with varying results. At least one study found internalized homophobia to be lower among PrEP users than among non-PrEP users.(Moeller, Seehuus, Wahl, & Gratch, 2020) Another study found higher levels of internalized homophobia and greater numbers of female sex partners to be positively associated with PrEP use.(Eaton, Matthews, et al., 2017)
Some MSM identify as heterosexual (HMSM) and tend to have sexual networks comprised of both men and women. HMSM with recent same-sex encounters have been found to engage in more HIV-related risk behaviors than HMSM without recent same-sex encounters.(Abdallah, Conserve, Burgess, Adegbite, & Oraka, 2020) HMSM, along with MSM who identify as bisexual, are more likely to report recent arrest compared to homosexual-identifying MSM.(Lim, Sullivan, Salazar, Spaulding, & DiNenno, 2011) HMSM have been found to report PrEP usage at lower rates than other MSM;(Abdallah et al., 2020) however, sexual identity and PrEP use among HMSM is relatively understudied.
For MSM who experience incarceration, the period following release is recognized as a high-risk period for HIV transmission. Relatively few studies have examined PrEP use among MSM who have recently left incarceration. Reentry populations are of particular importance to biomedical HIV prevention efforts because of the higher rates of incarceration experienced by MSM compared to other adult populations.(Meyer et al., 2017) Reentry following incarceration is associated with HIV risk behaviors, including polysubstance use, condomless sex, transactional sex, and injecting drugs with limited availability of sterile equipment.(Adams et al., 2011) People leaving incarceration may experience social disruptions during incarceration, such as the loss of a primary sexual partner and reduction of social support network, and these disruptions may increase sexual risk-taking following release.(Khan et al., 2011) Increased medical mistrust is associated with recent incarceration, potentially further complicating biomedical HIV prevention, which requires engaging with physicians and other health care providers.(Hoff et al., 2022)
The objective of this cross-sectional study was to assess whether sexual identity was associated with lifetime PrEP use among MSM reentry populations, and what factors might mediate the relationship, including sexual activity, demographics, incarceration history, injection drug use, and internalized homophobia. Recognizing the salience of both identity and behavior to our research question and our study inclusion criteria which focuses on behavior primarily, we use the term MSM when discussing our study sample. When citing others’ work, we use the terms employed by the authors of specific studies.
The MEPS Study was a randomized controlled trial that tested an intervention that aimed to increase PrEP use among individuals aged 18–49, assigned male at birth, recently incarcerated, with a history of substance use, who reported sex with a man or male-to-female transgender person in the 6 months before their most recent incarceration. Of the 220 completed baseline surveys, 32 did not identify as cisgender male; for example, female, non-binary, and gender-fluid. Additionally, 18 MEPS participants identified as cisgender male but reported a sexual identity that didn’t fall into straight/heterosexual, bi-/pansexual, or gay/same gender-loving. They are not included in this analysis because their experiences of sexual identity are likely to differ from those who identify as male. The remaining 170 male-identified participants are included in this analysis. The design of the study has been described elsewhere.(Edwards et al., 2020) Recruitment took place from November 2019 to December 2022. Sites of recruitment included the Los Angeles County Men’s Central Jail, facilities in Los Angeles County that offer substance use treatment, and the community.
Data analyzed in this paper come from baseline surveys. Variables analyzed were self-reported by participants, and included some that were analyzed as age, race, incarceration history, sexual activity, and injection drug use. These variables will be referred to collectively as “confounding variables” below.
For sexual identity, participants selected straight/heterosexual, homosexual/gay/same gender-loving, bisexual, queer, unsure or questioning, and other. Those selecting “queer”, “questioning”, “unsure”, or “other” were excluded from the analysis with one A small number (6) of respondents selected “other” and wrote “pansexual”, “open/no label”, and “open to everyone, and “attracted to everyone/everything”. They were combined with participants who identified with bisexual to obtain a bi-/pansexual category.
Age was categorized as 18–29, 30–39, and 40–49.
For race/ethnicity, choices were American Indian/Alaska Native, Asian, Black/African-American, Hispanic/Latino, Native Hawaiian/Other Pacific Islander, White, and Other. Participants selecting multiple categories were asked “which race/ethnicity, if any, do you identify with the most?” They were categorized and analyzed based on their response. Participants who responded “I identify with all my backgrounds equally” were classified as multi-racial, which made up the majority of the “other” category. Due to small sample sizes, those identifying as American Indian or Alaska Native, Asian, or Native Hawaiian/Other Pacific Islander were combined with the “other” category.
The internalized homophobia measure asked eight questions with answers choices 1-strongly agree, 2-agree somewhat, 3-neither agree nor disagree, 4-somewhat disagree, 5-strongly disagree, and refuse to answer. Questions were adapted from Herek et al. through research involving Black MSMW of varying sexual identities and similar socioeconomic status to this population (see table 1, supplemental material in Herek et al.).(del Pino et al., 2022; Herek, Cogan, Gillis, & Glunt, 1998) Responses were reverse-coded as necessary, and if responses for one or two questions were missing, responses were imputed as the average of the other responses. Participants with more than 3 skipped questions were treated as no response and were not included in the analysis. Responses were summed to calculate the total internalized homophobia score.
Lifetime PrEP use was scored yes=1/no=0 for responses to the question Have you ever used PrEP (pre-exposure prophylaxis)?
Length of lifetime incarceration was measured by the question During your entire lifetime, about how much time have you spent in a jail, prison, or a detention facility altogether? Please add up the total amount of time if you were held multiple times. Answer choices were Less than 1 week, 7–30 days, 1–5 months, 6–11 months, 1–2 years, 3–5 years, and 6 or more years. For analysis, we recategorized the variable into the categories < 6 months, 6 months to under 3 years, and 3 or more years.
For injection drug use, participants were asked if they had ever used any drug or other substance by injection, and responses were categories as 1=yes/0=no.
Sexual activity was a binary variable defined as anal or vaginal sex with a partner of any gender identity in the past 3 months. Participants were asked about sex with cisgender male, cisgender female, and transgender female partners separately.
We tested associations between sexual identity and all potential confounding factors using chi-squared tests. Sample size for some variables differs slightly due to missing responses. Missing data was <5% for all variables analyzed. We used linear regression to assess associations between internalized homophobia and sexual identity, and potential confounding factors. We used logistic regression to assess the associations of sexual identity and internalized homophobia with lifetime PrEP use, adjusting for the confounding factors that showed significant univariate associations with lifetime PrEP use (race/ethnicity and incarceration history). A multiple logistic regression with and without internalized homophobia was conducted to determine if internalized homophobia mediates the observed relationships. We limited confounders in the multiple regressions to those variables showing significant associations in the univariate analysis. For all regressions, we calculated pairwise contrasts for categorical variables with more than two levels. Individuals with missing data for any variable in the regression were not included in the model. Due to missing values for internalized homophobia, the sample size for the model when including this variable was reduced. We performed a sensitivity analysis in which the sample for models that did not include internalized homophobia was restricted to those that had an internalized homophobia score. All analyses were conducted using SAS v9.4 and a p-value <0.05 was used to determine significance.
Participant baseline characteristics are shown in Table 1. The sample included 170 MEPS enrollees who identified as straight/heterosexual, gay/same gender-loving, or bi-/pansexual men at the time of enrollment.
Fifty-five participants were aged 18–29 (32.4% of total), 71 were aged 30–39 (41.8%), and 44 were aged 40–49 (25.9%). Thirty-seven participants identified as straight/heterosexual (21.8% of total), 51 (30.0%) identified as gay/same gender-loving (SGL), and 82 (48.2%) identified as bi-/pansexual. Participants identifying as Hispanic or Latino, Black/African American, and White numbered 73 (42.9%), 42 (24.7%), and 38 (22.4%), respectively; seventeen (10.0%) identified with another or multiple identities. The number of participants reporting never having used PrEP in their lifetimes was 137 (80.6%), while 33 (19.4%) reported lifetime PrEP use. Nearly two-thirds (109 participants, 64.1%) reported sexual activity in the past three months, and less than one-third (49 participants, 29.3%) reported a history of injection drug use. Twenty-six participants (15.5%) reported less than 6 months of cumulative lifetime incarceration, with 58 (34.5%) reporting 6 months to 2 years, and 84 (50.0%) reporting 3 or more years. When compared across different sexual identity groups, only lifetime PrEP use varied significantly (χ2=13.53, p=0.0012) with gay/SGL having the greatest lifetime PrEP use (35.3%), then bi-/pansexual (15.9%) and straight/heterosexual (5.4%) the least.
Figure 1 gives mean internalized homophobia scores among sexual identity groups and potentially confounding variables. Internalized homophobia scores differed significantly by sexual identity group (p<0.0001); participants identifying as straight/heterosexual had the highest mean score (19.22; SE=1.10, SD=6.52), followed by bi-/pansexual (11.32; SE=0.96, SD=8.10), and gay/SGL (5.40; SE=0.95, SD=6.49). Internalized homophobia scores did not differ significantly among any of the confounding variables.
Table 2 gives the results of a multiple regression with internalized homophobia as the outcome. After adjusting for other confounders, sexual identity groups are significantly different in internalized homophobia. Compared to participants who identify as gay/SGL, straight/heterosexual-identifying participants score 13.83 points higher on the internalized homophobia scale (SE=1.69), while bi-/pansexual-identifying participants score 5.69 points higher (SE=1.43). No confounding variable showed significant associations with internalized homophobia.
See Supplemental Table 1 for results of the univariate logistic regression. Compared to Gay/SGL, we found significantly reduced odds of lifetime PrEP use for Straight/Heterosexual (OR=0.11, 0.02–0.49) and Bi-/pansexual (OR=0.35, 0.15–0.79), Black/African American race compared to White (OR 0.25, 0.06–0.99), and incarceration history of 3 or more years compared to <6 months (OR=0.29, 0.10–0.80). A rise of one point on the internalized homophobia scale was associated with a 10% reduction in the odds of reporting lifetime PrEP use (OR=0.90, 0.85–0.96).
The results of the multiple regression were similar to the univariate (see Table 3). The straight/heterosexual group had significantly reduced odds of lifetime PrEP use compared to the gay/SGL group (straight: aOR 0.10, 0.02–0.49; bi-/ aOR 0.39, 0.16– 0.95). The contrast from the regression showed that the straight/heterosexual group did not differ significantly from the bi-/pansexual group (aOR 0.26, 0.05–1.26). In a sensitivity analysis with the sample restricted to those having an internalized homophobia score, significant associations remained unchanged. Controlling for internalized homophobia rendered the differences in PrEP use non-significant among the three groups (χ^2^= 2.08, p=0.35).
Participants identifying as Black/African American were significantly less likely to have used PrEP than White-identifying participants before adjusting for internalized homophobia (aOR=0.15, 0.03–0.78). After adjusting, the trend persisted but was not quite significant (aOR=0.19, 0.04–1.03). Participants incarcerated for 3 years or more were significantly less likely to have used PrEP than men incarcerated for 6 months or less (aOR=0.28, 0.09–0.87). After controlling for internalized homophobia, the trend persisted but was no longer significant (aOR = 0.30, 0.09–1.06).
Lifetime PrEP use varied significantly among sexual identity groups, but did not appear to be driven by age, race/ethnicity, incarceration history, injection drug use, or sexual activity. Among male-identifying MEPS participants, those who identified as bi-/pansexual and straight/heterosexual were significantly less likely to report lifetime PrEP usage than those who identified as gay/SGL.
Internalized homophobia was an important driver of the difference between sexual identity groups in this analysis. After controlling for internalized homophobia, the χ^2^ statistic between groups became not significant.
There are compelling reasons for researchers to consider bisexually identified men separately from other MSM. Bisexual men may experience marginalization from both the heterosexual and gay communities, and have reported experiencing greater stigma around PrEP use than their gay/SGL counterparts.(Watson et al., 2023) Black/Hispanic bisexual individuals have been viewed more negatively than White bisexual individuals.(Feinstein et al., 2022) Bisexual men were less likely to disclose their sexual identity than gay men, an association attributed to lower connectedness to LGBT communities which may impact their exposure to PrEP and willingness to incorporate a biomedical HIV prevention strategy into their lives.(Keene, Heath, & Bouris, 2022) In a qualitative study of young MSM in Chicago, bisexual individuals were less likely than their gay counterparts to know anyone on PrEP, less likely to have spoken with their healthcare provider about PrEP, and less likely to report feeling comfortable telling someone if they began taking PrEP.(Phillips II, Raman, Felt, Han, & Mustanski, 2019) A 2016 report on PrEP disparities in California found that gay-identified survey respondents were much more likely to be aware of PrEP (OR=4.55) than respondents identifying as bisexual.(Pulsipher CA et al., 2016) A qualitative study of Black and Latino MSM found that bisexual-identifying participants lacking interest in accessing PrEP services cited PrEP’s association with gay-identifying men.(Jaramillo et al., 2022) Lower PrEP familiarity and use has been observed among bisexual-identifying men compared to gay-identifying men.(Grov, Rendina, Whitfield, Ventuneac, & Parsons, 2016) Bisexual-identifying men who use PrEP and are in relationships with women may fear that their PrEP use might unintentionally disclose their sexual practices to their partner.(Elopre et al., 2021) A 2014 review of sexual health interventions for men who have sex with men and women (MSMW) found that risk-reduction interventions alone are likely insufficient to improve sexual health.
This study highlights HMSM, a group that has received comparatively little attention, with many studies focused on young, college-aged men,(Carrillo & Hoffman, 2017) as well as men who publicly present as heterosexual but secretly have sex with men, a group of men once referred to as “down low.”(Ford, Whetten, Hall, Kaufman, & Thrasher, 2007; McCune Jr, 2014) HMSM appear in reports on PrEP use in small numbers; National HIV Behavioral Surveillance (NHBS) System data collected from 4,052 MSM in 20 urban areas in 2017 included just 20 heterosexual-identifying individuals.(Finlayson et al., 2019) One estimate of the proportion of straight/heterosexual-identified men who report ever engaging in sex with men was 2.6%,(Mishel, 2019) implying that this group may be insufficiently reached by current NHBS methods which focus on social venues with high proportions of MSM. One challenge to surveying this group may be the apparent discordance between an identity seen as referring to sex only with a different gender identity, and behaviors involving sex with multiple gender identities. This discordance exists across race/ethnicity groups but has been found to be highest among White and Black/African American men.(Ross, Essien, Williams, & FernÁNdez-Esquer, 2003) Notably, our findings demonstrate that HMSM report higher levels of internalized homophobia compared to men who identify as bi-/pansexual and gay/SGL. Future reporting should highlight data collected from straight/heterosexual-identified and bi-/pansexual-identified MSM as distinct from gay/SGL-identified MSM. In addition to comparing factors such as internalized homophobia, future studies should consider factors such as socioeconomic class and education levels.
We also observed significant differences in lifetime PrEP use for participants who identified as Black/African American compared to other racial/ethnic groups, as well as among participants reporting 3 or more years cumulative lifetime incarceration compared to participants who reported less than 6 months. After controlling for internalized homophobia, the difference became borderline non-significant for Black/African American individuals. For cumulative lifetime incarceration, the adjusted odds ratios remained significant even after controlling for internalized homophobia.
The finding of lower lifetime PrEP use among Black MSM aligns with the literature. In the 2023 HIV Surveillance Report to the United States Centers for Disease Control, 11% of Black/African American individuals who could benefit from PrEP were covered by it, compared to 21% of Hispanic/Latino individuals and 78% White individuals.(Monitoring selected national HIV prevention and care objectives by using HIV surveillance data—United States and 6 dependent areas, 2021, 2023) Eligible Black MSM are less likely to begin taking PrEP even when they reside in geographic areas with mechanisms to deliver PrEP free of cost regardless of insurance status.(Mulatu et al., 2022) Despite heightened PrEP awareness among all groups,(Finlayson et al., 2019) Black/African American MSM, in particular, exhibit a significantly lower rate of use.
Several factors may explain this lack of use. A systematic review of barriers and facilitators to the PrEP care continuum among Black MSM in the United States found that medical mistrust was a frequently identified barrier among qualitative studies reviewed.(Russ, Zhang, & Liu, 2021) Black MSM also have been found to associate PrEP use with promiscuity to a greater degree than White MSM, a belief associated with lower PrEP use.(Eaton, Kalichman, et al., 2017) Black MSM who perceive healthcare-related discrimination are less likely to be aware of PrEP, whereas those who disclose same-sex activity to healthcare providers are more likely to be aware.(Maksut, Eaton, Siembida, Fabius, & Bradley, 2018) Black MSM often have highly interconnected sexual networks due to a confluence of factors including population size, personal preferences, and sexual racism.(Newcomb, Ryan, Garofalo, & Mustanski, 2015; Raymond & McFarland, 2009) One potential driver of increased PrEP interest among younger Black MSM compared to their younger White MSM counterparts is their awareness of the higher HIV prevalence in their communities,(Eaton, Kalichman, et al., 2017) suggesting a pathway to PrEP engagement for the former.
Few carceral settings offer PrEP or other options for sexual risk reduction in the United States, though PrEP has been implemented in these setting in other countries.(Lindsay et al., 2023) We did not find significant differences in cumulative years incarcerated among participants of differing racial/ethnic or sexual identity. Given the wealth of quantitative and qualitative data in the literature highlighting incarceration as a risk factor for HIV, strategies to incorporate PrEP into carceral settings, including during reentry planning and offering long-acting injectable PrEP, are a promising avenue for improving use and potentially decreasing HIV incidence.(Edwards et al., 2023; Shabazz, 2015)
These findings must be interpreted considering the study’s limitations. The MEPS eligibility criteria did not include sexual identity and instead focused on sexual behavior (“reports sexual intercourse with a male or male-to-female transgender women in the six months prior to jail entry”). This potentially excluded some sexual and gender minorities who did not report sexual intercourse during the defined time period. Study participants were recruited from facilities that were generally LGBTQI+ friendly, which might have increased their likelihood of PrEP use and impact the generalizability of the results. The study was conducted in California where, compared to other states, PrEP is more available to patients regardless of insurance or immigration status, which could increase the likelihood of lifetime PrEP use in the study population.(“Medi-Cal Expansion Provided 286,000 Undocumented Californians With Comprehensive Health Care,” 2022; “SB159: California Pharmacists Initiation of PrEP and PEP in a pharmacy,” 2022)
Despite these limitations, the study setting and inclusion criteria allowed us to examine factors that may still limit PrEP use in a setting where several structural barriers have been overcome and to do so among a population with recent indications for PrEP. The insights about MSM and differences observed when stratifying identification by race/ethnicity and sexual orientation can be used to refine interventions and ensure they reach diverse subsets of MSM. Interventions to increase PrEP uptake among Black and Latino MSM should address the underlying causes of medical mistrust, including healthcare-related discrimination, as well as leveraging existing social networks as social support from others with similar intersectional identities.(Earnshaw, Bogart, Dovidio, & Williams, 2015)
For MSM who identify as heterosexual or bisexual, additional efforts are needed to address underlying sociocultural factors (biphobia and norms around masculinity) which can form the foundation for internalized homophobia, including when directed from other corners of the LGBTQ+ community. Community-wide efforts, then, are needed to eliminate these factors at the society level. Furthermore, PrEP access for MSM of all sexual orientations is negatively impacted by poverty, racism, and uninsurance.(Doblecki-Lewis et al., 2017; Lee et al., 2023; Serota, Rosenberg, Thorne, Sullivan, & Kelley, 2019) These factors affect MSMW norms, along with an improved understanding of how varying sexual identities shape this group’s attitudes and behaviors.(Jeffries, 2014) In addition to expanding outreach approaches to increase the number of HMSM reached by surveys such as NHBS so that trends in their PrEP uptake over time can be observed, interventions for HMSM must take into account members’ increased levels of internalized homophobia and draw distinctions between sexual practice and sexual identity.