Sexuality Disclosure in U.S. Gay, Bisexual, and Other Men Who Have Sex With Men: Impact on Healthcare-Related Stigmas and HIV Pre-exposure Prophylaxis Denial

Abstract

Introduction:

Stigma impairs access to health care by gay, bisexual, and other men who have sex with men (GBMSM). GBMSM who are open about their sexuality (“out”) are more resilient to stigma than GBMSM who are “not out.” Outness may influence healthcare utilization and prescription of HIV pre-exposure prophylaxis (PrEP) to HIV-negative GBMSM.

Methods:

Analyzing the 2018 American Men’s Internet Survey during 2019, the adjusted prevalence ratios (APRs) of healthcare stigmas and outness to healthcare providers were calculated. The effect of outness on annual healthcare visits and stigma was measured. PrEP seeking and denial by providers was quantified and stratified by outness.

Results:

Of 5,794 respondents, 3,402 (58.7%) were out to their provider. Out GBMSM were less likely to experience anticipated stigma (APR=0.75, 95% CI=0.72, 0.80) but more likely to experience enacted stigma or discrimination (APR=1.23, 95% CI=1.18, 1.28). In a subsample of out GBMSM, recently experienced discrimination was associated with higher healthcare utilization (APR=1.51, 95% CI=1.14, 1.51). Conversely, recent experienced discrimination was associated with lower healthcare utilization in not out GBMSM (APR=0.67, 95% CI=0.54, 0.82). Of 3,104 out GBMSM, 1,417 (45.7%) discussed PrEP with their providers, compared with 120/1,711 (7.0%) GBMSM not out (p<0.001). PrEP denials were less common among out (116/793, 14.6%) than not out (14/55, 25.5%) GBMSM (p=0.044).

Conclusions:

Healthcare provider-related stigmas impair healthcare engagement among not out GBMSM who were also more commonly denied PrEP. Ending the HIV epidemic necessitates creating safe environments for disclosure of sexual preferences and practices to facilitate access to HIV prevention.

INTRODUCTION

In the U.S., gay, bisexual, and other men who have sex with men (GBMSM) report higher rates of verbal harassment, poor treatment in healthcare settings, and physical and sexual violence than the general population.^1,2 GBMSM who anticipate or experience stigma engage in riskier sexual activities, have less access to health care, and are more likely to miss opportunities for HIV prevention.^3-8 “Outness,” or the extent to which one is open about one’s sexuality and sexual practices, plays a complex role in the relationship between stigma and health outcomes. Individuals who have a higher degree of outness regarding their sexuality experience higher levels of enacted stigma or discrimination.^2,9,10 However, they also have more social and emotional resources to cope with discriminatory experiences and are more likely to be offered appropriate prevention and treatment related to their sexuality.^10-13

At current HIV infection rates, GBMSM have a one in six lifetime risk of HIV infection and black GBMSM have a one in two lifetime risk of HIV infection.¹⁴ This highlights the urgent need to increase the uptake of pre-exposure prophylaxis (PrEP), which greatly reduces the risk of becoming infected with HIV.¹⁵ Although PrEP use has been steadily increasing, less than 10% of the estimated 800,000 U.S. GBMSM at increased risk for HIV infection were prescribed PrEP in 2017.^16,17 During screening for PrEP, GBMSM are asked their sexual histories and must make the decision whether to disclose their sexual preferences and practices including sex with men. Understanding the dynamics of how sexuality outness intersects with race, sexual behavior stigmas, healthcare utilization, and PrEP uptake is important for increasing its use among eligible GBMSM.

Prior surveys of healthcare providers found that most supported offering PrEP, but few actually prescribed it owing to lack of prescribing experience and concerns about costs and side effects.^18-20 Some providers cited concern for PrEP increasing risky sexual practices, a phenomenon known as risk compensation, as a reason for not prescribing PrEP.^21,22 There is some evidence that GBMSM on PrEP may increase their number of sexual partners and frequency of condomless sex, resulting in moderately increased incidences of gonorrhea or chlamydia.²³ Concerningly, some studies have suggested that providers were more likely to deny PrEP if the provider expressed higher levels of heterosexism, and providers may more likely to deny PrEP to black than white GBMSM because of concerns about risk compensation.^22,24 These findings suggest that provider-related intersectional stigmas by race and sexuality may compound existing sociodemographic barriers to accessing PrEP and reinforce disparities in PrEP uptake and ultimately HIV risks among black GBMSM.^25-27

This study seeks to understand the factors that predict GBMSM being “out” about their sexualities to their providers and the effect of outness on the relationship between healthcare-related stigmas and utilization. Finally, the study identifies how GBMSM’s sexuality outness affects PrEP seeking and denial.

METHODS

Study Population

The American Men’s Internet Survey (AMIS) is one of the largest online behavioral health surveys of GBMSM in the U.S. and has been conducted annually since 2013.^28,29 GBMSM are recruited through convenience sampling from banner ads on websites, social networks, and geospatial networking apps. Data for the AMIS 2018 cycle were collected between September 2018 to November 2018. Participants were eligible to participate if they were aged ≥15 years, male sex at birth, cis-gendered, resided in the U.S., and reported ever having sex with a man or a gay/bisexual sexual identity. Participants were asked questions regarding outness, stigma, and use of HIV prevention services. Participants were not asked every behavioral question to minimize survey burden, and the questions were administered randomly to maintain survey representativeness. Participants were included in this analysis if they completed the survey, reported male–male sex in the past 12 months, and reported being HIV-negative or unknown status.

Measures

All participants were asked their age, race/ethnicity, highest level of education, annual household income, and insurance status. Participants reported their ZIP code of residence, which was assigned to one of four Census regions and one of six urban–rural classifications using categories defined by the National Center for Health Statistics. ³⁰

Participants were asked whether they had told anyone they were attracted to or had sex with men. Those answering affirmatively were then asked whether they had disclosed their sexuality to their healthcare providers. Disclosing sexual activities with men to healthcare providers was assumed to apply to healthcare encounters discussing PrEP, whereas answering no to either questions meant that they were “not out” during PrEP encounters.

Anticipated healthcare stigma was measured by asking whether participants either felt afraid or avoided healthcare services because of fear someone may learn they had sex with men. Enacted stigma discrimination was measured by asking participants if they heard healthcare providers gossiping about them or had not been treated well in a healthcare setting because they had sex with men. Each form of stigma was also categorized by whether it occurred in the prior 6 months or ever. These metrics have been validated previously in the U.S. and internationally.³¹

Participants were asked whether they had visited a healthcare provider in the past 12 months. Participants answering affirmatively were then asked how many times they saw a healthcare provider not including hospitalizations, emergency room visits, or telephone calls within the past 12 months.

Participants were asked if they discussed PrEP with a healthcare provider in the past 12 months. Participants who answered affirmatively were then asked whether the participant or provider initiated the conversation about PrEP. Among those who requested PrEP, the participant was asked whether they started PrEP, decided not to start PrEP after learning more about it, or was denied PrEP by their healthcare provider. Participants who received PrEP or decided not to start it were classified as not being denied PrEP, whereas those desiring a prescription but not being offered it were classified as being denied PrEP. Among those who were denied PrEP, participants were asked whether they perceived discrimination due to race, sexuality, or SES. Participants reported the provider’s primary reason for PrEP denial and the responses were classified as: (1) not knowing how to prescribe PrEP, (2) deeming that the patient lacked an indication for PrEP or had a medical contraindication, or (3) providing a reason that was not based on Centers for Disease Control and Prevention or U.S. Food and Drug Administration guidelines.^15,32

Statistical Analysis

Participants were compared with respect to demographic characteristics and healthcare stigmas by outness to provider. Unadjusted log-binomial regressions were used to estimate differences in demographic and healthcare stigma factors by outness to provider. Then, stepwise multivariable log-binomial fitting models were used to select demographic covariates that significantly (p<0.05) affected the relationship between outness to provider and the type of stigma experienced. Each final log-binomial model included the selected covariates for the assessment of the adjusted relationship between outness to a provider and type of stigma reported.

For participants with healthcare utilization data, the association between anticipated healthcare stigma and seeing a healthcare provider in the past 12 months was measured using adjusted log-binomial regression using the same stepwise fitting process for demographic covariates. Among participants with one or more healthcare visits, the mean number of healthcare visits were compared against whether participants experienced healthcare discrimination. Prevalence ratios (PRs) comparing the mean number of annual healthcare visits between those who experienced and did not experience discrimination were calculated using adjusted Poisson regression with robust error variance and stepwise covariate fitting. Both analyses were stratified based on outness to their provider owing to the detection of effect modification.

The proportional differences in outness among GBMSM who discussed PrEP with their provider, requested it rather than had their provider mention it, and were denied PrEP were quantified using chi-square analysis. Finally, PRs were calculated using adjusted log-binomial regressions with stepwise fitting to assess the difference in demographic characteristics between participants denied and not denied PrEP. All analyses were conducted using in SAS, version 9.4 during the year 2019. The study was approved by the IRB of Emory University.

RESULTS

Of the 10,129 individuals who completed the AMIS 2018 survey, 5,794 (57.2%) were HIV-negative and answered questions on both healthcare stigma and outness to their provider. Among the 5,794 respondents included in the analysis, 3,711 (64.0%) were aged 15–34 years, 4,057 (70.0%) were white, 2,154 (37.2%) were from the South, 3,236 (55.9%) lived in a large metro or fringe area, 1,891 (32.6%) reported an annual household income of >$75,000, and 3,724 (64.3%) reported having private insurance (Table 1). Further, 3,402 (58.7%) were out to their provider, 1,722 (29.7%) had reported ever experiencing anticipatory healthcare stigma, and 502 (8.7%) had reported ever experiencing stigma in a healthcare setting. Outness to a healthcare provider was more prevalent among respondents who were older, white, resided in the West or Northeast, lived in a large metro or fringe area, had a college degree or higher, made ≥$75,000 annually, and had private health insurance. Respondents were less likely to be out to their provider if they reported anticipating healthcare stigma ever (adjusted PR [APR]=0.75, 95% CI=0.72, 0.80) or within the past 6 months (APR=0.44, 95% CI=0.39, 0.49). Conversely, respondents were more likely to be out to their provider if they reported having experienced healthcare discrimination ever (APR=1.23, 95% CI=1.18, 1.28) or within the past 6 months (APR=1.13, 95% CI=1.05, 1.23).

Table 1.

Log-binomial Regression of Sexuality Disclosure to a Healthcare Provider by Demographic Characteristics and History of Stigma (N=5,794), 2018 American Men’s Internet Survey

Variable	Out to providers (n=3,402)	Not out to providers (n=2,392)	Unadjusted PR (95% CI)	Adjusted PR (95% CI)
Age, years, n (%)
15–24	1,067 (42.8)	1,429 (57.3)	ref
25–34	868 (71.4)	347 (28.6)	1.67 (1.58, 1.77)***
35–44	401 (70.6)	167 (29.4)	1.65 (1.54, 1.77)***
45–54	476 (70.7)	197 (29.3)	1.65 (1.55, 1.77)***
≥55	590 (70.1)	252 (29.9)	1.64 (1.54, 1.75)***
Race, n (%)
White	2,458 (60.6)	1,599 (39.4)	ref
Black	147 (49.8)	148 (50.2)	0.82 (0.73, 0.92)**
Hispanic	507 (54.3)	426 (45.7)	0.90 (0.84, 0.96)***
Other/mixed/unknown	290 (57.0)	219 (43.0)	0.94 (0.87, 1.02)
Region, n (%)
West	882 (62.2)	535 (37.8)	ref
Midwest	736 (56.9)	557 (43.1)	0.91 (0.86, 0.97)**
Northeast	554 (59.6)	376 (40.4)	0.96 (0.90, 1.02)
South	1,230 (57.1)	924 (42.9)	0.92 (0.87, 0.97)**
Urbanicity,a n (%)
Large metro/ fringe	2,018 (62.4)	1,218 (37.6)	ref
Medium/small metro	1,103 (56.5)	848 (43.5)	0.91 (0.86, 0.95)***
Micropolitan/non-core	281 (46.3)	326 (53.7)	0.74 (0.68, 0.81)***
Education, n (%)
College or higher	1,473 (74.0)	519 (26.1)	ref
Some college/Associates	690 (67.9)	326 (32.1)	0.92 (0.87, 0.97)***
High school or less	164 (61.0)	105 (39.0)	0.82 (0.75, 0.91)***
Unknown	1,075 (42.7)	1,442 (57.3)	0.58 (0.55, 0.61)***
Income, n (%)
≥$75,000	1,229 (65.0)	662 (35.0)	ref
$40,000–$74,999	885 (61.5)	555 (38.5)	0.95 (0.90, 0.99)
$20,000–$39,999	652 (59.5)	444 (40.5)	0.92 (0.86, 0.97)**
$0–$19,999	410 (55.1)	334 (44.9)	0.85 (0.79, 0.91)***
Unknown	226 (36.3)	397 (63.7)	0.56 (0.50, 0.62)***
Insurance, n (%)
Private only	2,347 (63.0)	1,377 (37.0)	ref
Public only	382 (56.3)	297 (43.7)	0.89 (0.83, 0.96)**
Other/multiple	290 (60.4)	190 (39.6)	0.96 (0.89, 1.03)
None	252 (56.6)	193 (43.4)	0.90 (0.83, 0.98)*
Unknown	131 (28.1)	335 (71.9)	0.45 (0.39, 0.52)***
Anticipated stigma ever,b n (%)
No	2,608 (64.0)	1,464 (36.0)	ref	ref
Yes	794 (46.1)	928 (53.9)	0.72 (0.68, 0.76)***	0.75 (0.72, 0.80)***
Anticipated stigma 6 months,c n (%)
No	3,204 (64.0)	1,802 (36.0)	ref	ref
Yes	198 (25.1)	590 (74.9)	0.39 (0.35, 0.44)***	0.44 (0.39, 0.49)***
Experienced discrimination ever,d n (%)
No	2,998 (56.7)	2,294 (43.4)	ref	ref
Yes	404 (80.5)	98 (19.5)	1.42 (1.35, 1.49)***	1.23 (1.18, 1.28)***
Experienced discrimination 6 months,e n (%)
No	3,298 (58.4)	2,346 (41.6)	ref	ref
Yes	104 (69.3)	46 (30.7)	1.19 (1.06, 1.32)**	1.13 (1.05, 1.23)**

Note: Boldface indicates statistical significance

^*p<0.05

^**p<0.01

^***p<0.001.

^aLarge metro/fringe: county within a metropolitan statistical area of ≥1 million inhabitants, medium/small metro: county within a metropolitan statistical area of <1 million inhabitants, micropolitan/non-core: non-metropolitan county.

^bAnticipate stigma ever covariates: race, region, urbanicity, education, income, and insurance.

^cAnticipated stigma past 6 months covariates: race, urbanicity, education, income, and insurance.

^dExperienced discrimination ever covariates: urbanicity, education, income, and insurance.

^eExperienced discrimination past 6 months covariates: region, urbanicity, education, income, and insurance.

PR, prevalence ratio.

Of the 5,794 respondents who answered healthcare stigma and outness questions, anticipated stigma ever and in the past 6 months was not associated with seeing a healthcare provider in the past year in both unadjusted and adjusted multivariable analyses. Among the 1,629 respondents who answered healthcare stigma and outness questions and had one or more healthcare visits in the past 12 months, 1,041 (63.9%) were out with their provider. Overall, respondents out to their provider had an average of 4.1 annual healthcare visits compared with 3.0 among those “not out” to their provider (PR=1.39, 95% CI=1.27, 1.51). Among respondents out to their provider, healthcare discrimination ever or in the past 6 months was associated with more frequent healthcare visits (ever: APR=1.21, 95% CI=1.04, 1.41; past 6 months: APR=1.51, 95% CI=1.14, 2.00) (Table 2). Conversely, among respondents not out to their provider, experiencing healthcare discrimination in the past 6 months was associated with fewer healthcare visits in the past 12 months (APR=0.67, 95% CI=0.54, 0.82) (Table 2).

Table 2.

Bivariate and Multivariable Poisson Regression of the Association Between Experienced Healthcare Discrimination and Healthcare Visits in the Past 12 Months by Outness to Provider (N=1,629), 2018 American Men’s Internet Survey

Variable	Mean visits (SD)	Unadjusted PR (95% CI)	Adjusted PR (95% CI)a
Out to providers (n=1,041)
Experienced discrimination, ever		1.26 (1.08, 1.46)**	1.21 (1.04, 1.41)**
No (n=922)	4.0 (3.3)
Yes (n=119)	5.1 (4.3)
Experienced discrimination, 6 months		1.53 (1.14, 2.07)**	1.51 (1.14, 2.00)**
No (n=1,013)	4.1 (3.4)
Yes (n=28)	6.3 (5.1)
Not out to providers (n=588)
Experienced discrimination, ever		1.20 (0.76, 1.88)	1.23 (0.76, 1.97)
No (n=570)	3.0 (2.7)
Yes (n=18)	3.6 (3.5)
Experienced discrimination, 6 months		0.57 (0.46, 0.70)***	0.67 (0.54, 0.82)***
No (n=581)	3.0 (2.7)
Yes (n=7)	1.7 (0.5)

Note: Boldface indicates statistical significance

^*p<0.05

^**p<0.01

^***p<0.001.

^aOut to provider covariates: age, race, insurance.

^bNot out to provider covariates: age, region, education, insurance.

PR, prevalence ratio.

Of the 4,815 respondents who saw a healthcare provider in the past 12 months and answered outness questions, 1,417 of 3,104 (45.7%) GBMSM out to their provider discussed PrEP with their provider compared with 120 of 1,711 (7.0%) of GBMSM not out to their provider (p<0.001). Of a randomized subsample of 1,105 respondents who discussed PrEP with their provider, 803 of 1,025 (78.3%) GBMSM who were out about their sexualities sought PrEP compared with 55 of 80 (68.8%) GBMSM not out to their providers (p=0.047). Among the 793 GBMSM out to their providers who sought PrEP, 586 (73.9%) received PrEP, 91 (11.5%) decided not to take PrEP, and 116 (14.6%) were denied a PrEP prescription. Of the 55 GBMSM who were not out to their providers and who sought PrEP, 37 (67.3%) received PrEP, four (7.3%) decided not to take PrEP, and 14 (25.5%) were denied PrEP (p=0.044 for PrEP denials by sexuality outness).

Of the 130 respondents who were denied a PrEP prescription, 57 (43.9%) reported perceiving any form of discrimination in the process of PrEP denial with 45 (34.6%) reporting sexuality-based discrimination, 14 (10.8%) reporting socioeconomic-based discrimination, and three (2.3%) reporting racial/ethnic-based discrimination. Additionally, 121 respondents were asked the primary reason their provider did not prescribe PrEP, and 43 (35.5%) noted their provider did not know how to prescribe PrEP, 29 (24.0%) reported reasons that were consistent with the existing guidelines, and 37 (30.6%) reported reasons that were inconsistent with current guidelines (Table 3). Of the 37 who reported provider reasons not supported by guidelines, 21 (56.8%) providers thought the patient could not afford PrEP, ten (27.0%) thought PrEP would encourage unsafe sex, three (8.1%) thought adolescents could not be prescribed PrEP, and three (8.1%) had an overt moral objection to homosexuality. Respondents who were younger, earned <$40,000 annually, lived in a micropolitan/non-core area, had public health insurance, and were not out to their provider were more likely to be denied PrEP (Table 4). However, stepwise multivariable analysis demonstrated that none of these covariates were statistically significant in a multivariable model.

Table 3.

Patient-reported Primary Reason Providers Denied PrEP (N=121), 2018 American Men’s Internet Survey

PrEP denial reason	n (%)
Provider did not know how to prescribe	43 (35.5)
Reasons supported by guidelines
Not at high risk for HIV	16 (13.2)
Medical contra-indication	13 (10.7)
Reasons not supported by guidelines
Thought patient would not be able to afford PrEP	21 (17.4)
PrEP would encourage unsafe sex	10 (8.3)
Thought adolescents could not receive PrEP	3 (2.5)
Moral objection to homosexuality	3 (2.5)
Other/Unsure	12 (9.9)

PrEP, pre-exposure prophylaxis.

Table 4.

PrEP Denial by Patient Demographic and Behavioral Characteristics (N=848), 2018 American Men’s Internet Survey

Variable	Denied (n=130)	Not denied/Decided not to take (n=718)	Unadjusted PR (95% CI)
Age, years, n (%)
15–24	48 (20.7)	185 (79.3)	ref
25–34	31 (13.6)	197 (86.4)	0.66 (0.43, 0.99)
35–44	13 (11.2)	103 (88.8)	0.54 (0.31, 0.96)*
45–54	22 (16.4)	112 (83.6)	0.79 (0.50, 1.25)
≥55	16 (11.6)	122 (88.4)	0.56 (0.33, 0.95)
Race, n (%)
White	95 (15.3)	528 (84.8)	ref
Black	6 (18.2)	27 (81.8)	1.19 (0.56, 2.52)
Hispanic	16 (13.5)	103 (86.6)	0.88 (0.54, 1.44)
Other/mixed/unknown	13 (17.8)	60 (82.2)	1.17 (0.69, 1.98)
Region, n (%)
West	34 (14.1)	208 (86.0)	ref
Midwest	32 (18.3)	143 (81.7)	1.30 (0.84, 2.02)
Northeast	17 (12.6)	118 (87.4)	0.90 (0.52, 1.54)
South	47 (15.9)	249 (84.1)	1.13 (0.75, 1.70)
Urbanicity, n (%)a
Large metro/fringe	73 (13.6)	465 (86.4)	ref
Medium/small metro	42 (16.4)	214 (83.6)	1.21 (0.85, 1.71)
Micropolitan/non-core	15 (27.8)	39 (72.2)	2.04 (1.26, 3.30)**
Education, n (%)
College or higher	47 (11.5)	361 (88.5)	ref
Some college/Associates	29 (16.6)	146 (83.4)	1.44 (0.94, 2.21)
High school or less	6 (20.0)	24 (80.0)	1.74 (0.81, 3.72)
Unknown	48 (20.4)	187 (79.6)	1.77 (1.23, 2.56)**
Income, n (%)
>$75,000	30 (10.0)	270 (90.0)	ref
$40,000–$74,999	37 (15.2)	206 (84.8)	1.52 (0.97, 2.39)
$20,000–$39,999	34 (21.4)	125 (78.6)	2.14 (1.36, 3.36)**
$0–$19,999	21 (20.0)	84 (80.0)	2.00 (1.20, 3.34)**
Unknown	8 (19.5)	33 (80.5)	1.95 (0.96, 3.96)
Insurance, n (%)
Private only	82 (13.4)	529 (86.6)	ref
Public only	23 (23.7)	74 (76.3)	1.77 (1.17, 2.66)**
Other/multiple	13 (16.9)	64 (83.1)	1.26 (0.74, 2.15)
None	6 (12.8)	41 (87.2)	0.95 (0.44, 2.06)
Unknown	6 (37.5)	10 (62.5)	2.79 (1.44, 5.43)**
Out to providers, n (%)
Not out	14 (25.5)	41 (74.6)	ref
Out	116 (14.6)	677 (85.4)	0.57 (0.35, 0.93)*

Note: Boldface indicates statistical significance

^*p<0.05

^**p<0.01

^***p<0.001.

^aLarge metro/fringe: county within a metropolitan statistical area of ≥1 million inhabitants, medium/small metro: county within a metropolitan statistical area of <1 million inhabitants, micropolitan/non-core: non-metropolitan county.

PR, prevalence ratio, PrEP, pre-exposure prophylaxis.

DISCUSSION

This study found that GBMSM who experienced fear or avoidance of health care out of concern of their sexuality being disclosed were less likely to be out to their providers. Conversely, those who were out to their providers were more likely to have experienced discrimination in healthcare settings. Outness to a healthcare provider was also identified as an effect modifier of the relationship between recent experienced healthcare discrimination and healthcare utilization. Those who were out to their providers but had experienced recent healthcare discrimination saw their providers more frequently than those who had not experienced healthcare discrimination, whereas those who were not out to their providers and experienced recent healthcare discrimination saw their providers less frequently.

The GBMSM who were out to their providers tended to be older, white, reside in the West or Northeast, live in a large metro or peri-urban area, have a college degree or higher, and made ≥$75,000 annually. This suggests that GBMSM may feel more comfortable being out if they possess greater social privilege, live in environments tolerant of same-sex relationships, and have access to providers accustomed to caring for sexual minorities. GBMSM able to be out may have a greater ability to develop supportive communities, build higher self-esteem to be resilient against stigma, and receive appropriate sexual health counseling and preventative services. ^10-12,33,34 Unfortunately, many GBMSM may be unable to be out owing to heterosexist stigma in their communities, fear of exclusion, or even threat of violence.^1,35-37

The decision for a patient to disclose their sexual identity to their provider is an intimate one and care should be taken to preserve this trust. The finding that not out GBMSM who experienced discrimination based on their assumed sexuality had lower healthcare utilization is concerning and may suggest that being involuntarily outed in a healthcare setting may dissuade GBMSM from continued engagement in healthcare. It is incumbent on healthcare administrators and providers to ensure that GBMSM receive equitable and culturally and clinically competent care. Strategies to mitigate stigmas can be structural and include removing stigmatizing policies, hiring staff with diverse sexual minority statuses and racial/ethnic backgrounds, involving staff in anti-bias training, restructuring facilities to enhance patient confidentiality, and implementing policies to ensure equitable treatment of all staff and patients.^38-40 Provider-level interventions can include teaching providers about the effect of stigma on health, building appropriate skills to serve the racial/ethnically diverse GBMSM, and directly engaging with them to develop empathy and break down stereotypes.^38-40

Regarding PrEP, the study found that seeking PrEP was more common among out GBMSM and that not out GBMSM were more likely to be denied PrEP. This could potentially be due to providers not identifying an indication to prescribe PrEP to not out GBMSM. This observation, in addition to the finding that more than a third of patients denied PrEP reported discrimination by sexuality, suggests that patients may be challenged by the choice to either keep their sexuality hidden but risk being denied PrEP or disclose their sexuality and possibly subject themselves to discrimination.

Two thirds of PrEP denials were related to providers either not knowing how to prescribe PrEP or having misconceptions about it. Providers not knowing how to prescribe PrEP was the largest barrier to prescribing PrEP.^21,41,42 Providers also perceived PrEP as expensive, but out-of-pocket costs are low or zero for most patients because of insurance and medication assistance programs.^42,44 Providers also expressed concern about PrEP risk compensation and increased risk of sexually transmitted infections.^23,45 This is not a reason to withhold HIV prevention and withholding PrEP may exacerbate existing HIV disparities, especially for black GBMSM who are more likely to be denied PrEP because of risk compensation concerns.^22,46 Black GBMSM remain at elevated risk for HIV even though they do not engage in risk behaviors at higher rates, highlighting the importance of ensuring their access to PrEP.^47,48 Finally, PrEP is approved in adolescents who weigh at least 35 kilograms/77 pounds.³² Dissemination of the most up-to-date science of PrEP could potentially reduce some of these PrEP denials.

Limitations

This study has limitations to consider. AMIS is an online convenience sample and respondents tended to be more educated, white, and have higher incomes compared with the general U.S. male population. Therefore, deeper exploration of how racism and classism intersect with stigma and outness was limited by sample size. AMIS is cross-sectional and cannot elucidate the exact relationship between outness to a provider, healthcare stigma, and PrEP. For instance, GBMSM may be selective in providers they are out to or seek PrEP from a different provider than a primary care provider. Additional research focusing on outness with a specific provider and its effect on the patient–provider relationship is needed to establish causality.

CONCLUSIONS

Sexual behavior stigmas in healthcare settings adversely affects the healthcare access GBMSM and their patient–provider relationships. To end the HIV epidemic, healthcare systems in the US as well as around the world need to engender safe environments that build trust facilitating GBMSM to disclose their sexuality and access risk-appropriate HIV prevention. Structural and individual-level interventions developed in partnership with racially and ethnically diverse GBMSM can make healthcare systems more welcoming and improve the clinical outcomes of the population served.