A Qualitative Study of Medical Mistrust, Perceived Discrimination, and Risk Behavior Disclosure to Clinicians by U.S. Male Sex Workers and Other Men Who Have Sex with Men: Implications for Biomedical HIV Prevention

Abstract

Access to biomedical HIV prevention technologies such as pre-exposure prophylaxis (PrEP) requires individuals to disclose risk behavior to clinicians, but experiences of discrimination and medical mistrust may limit disclosure among male sex workers and other MSM. We explored experiences of perceived discrimination, medical mistrust, and behavior disclosure among male sex workers compared to other men who have sex with men (MSM). We conducted 56 interviews with MSM and compared findings about medical mistrust, discrimination, and disclosure for 31 men who engaged in sex work vs. 25 men who did not. MSM who engaged in sex work reported more medical mistrust and healthcare discrimination due to issues beyond MSM behavior/identity (e.g., homelessness, substance use, poverty). MSM who did not report sex work described disclosing sex with men to clinicians more often. Both subgroups reported low PrEP awareness, but willingness to disclose behavior to obtain PrEP. Medical mistrust and perceived discrimination create barriers for sexual behavior disclosure to clinicians, potentially impeding access to PrEP and other forms of biomedical HIV prevention. These barriers may be higher among male sex workers compared to other MSM, given overlapping stigmas including sex work, substance use, homelessness, and poverty. An intersectionality framework for understanding multiple stigmas can help to identify how these dynamics may limit access to biomedical HIV prevention among male sex workers, as well as suggesting strategies for addressing stigmas to improve the delivery of PrEP and other HIV prevention approaches in this population.

Emerging forms of biomedical HIV prevention, such as pre-exposure prophylaxis (PrEP), post-exposure prophylaxis (PEP), and treatment as prevention, require clinical providers to serve gatekeeping roles. Research has increasingly focused on the role of clinical gatekeepers now that pre-exposure prophylaxis (PrEP) is known to be safe and effective for preventing HIV.^1–6 In order for providers to verify PrEP eligibility along CDC guidelines, patients must disclose behaviors that establish “substantial risk of HIV infection.”⁵ For men who have sex with men (MSM)—a behavioral category that encompasses men who may or may not self-identify as gay or bisexual⁷—the CDC suggests identifying “any anal sex without condoms,” recent sexually transmitted infection (STI), and/or an ongoing relationship with an HIV-positive man as indications for PrEP.⁵ Behavior disclosure to providers is thus critical for implementing PrEP and other biomedical HIV prevention strategies.

Although MSM are a key population for biomedical HIV prevention,^1,8–10 difficulties communicating about behavior in healthcare settings may hinder implementation. Prior studies have found that mistrust, disclosure difficulties, and internalized homophobia may limit healthcare access and delay presentation for care among MSM.^11–14 A recent study of New York MSM found that although most had seen a clinical provider recently, 39 % did not disclose sexual activity with male partners, with lower rates among African American and Latino MSM.¹⁵ Medical mistrust,^16–19 discrimination experiences,^18–21 conspiracy beliefs,^19,22,23 and concerns about disclosure may be particularly significant barriers to PrEP access among MSM of color, as demonstrated by a recent PrEP acceptability study among African American MSM.²⁴ Treatment mistrust and HIV conspiracy beliefs are also linked to sexual risk-taking and nonadherence to antiretroviral treatment^22,23; it is unknown how mistrust may now influence PrEP use. As a complement to these findings, provider-based research also suggests that some clinicians experience discomfort discussing sexual health, lack time to discuss sexual behavior, or are uncertain how to approach sexual health with MSM patients.^14,25

Within the population of MSM, male sex workers (MSWs) experience heightened HIV risk due to sexual and substance use behaviors, socioeconomic disparities, STIs, social stigma, and limited opportunities for condom use with clients.^26–33 Studies of MSWs in North America have identified HIV prevalence ranging from 5 to 31 %, with estimates exceeding prevalence among MSM generally.³³ Many MSWs may not self-identify as gay/bisexual men or sex workers, which can limit the reach of HIV-related programming and services intended for these communities.³³ MSWs may also be less willing than other MSM to disclose behavior in healthcare settings; one San Francisco study found that 76.6 % of MSWs “never disclosed” sex work to clinicians.³⁴

Despite acknowledgement of medical mistrust, perceived discrimination, and disclosure difficulties among MSWs and other MSM, little research has considered differences in how MSWs experience these phenomena. Many comparative analyses of medical mistrust and disclosure among subgroups of MSM focus on differences by race and ethnicity,^15–24,35 rather than sex work. Earnshaw and colleagues have proposed a comprehensive model linking mistrust to HIV risk,³⁵ which suggests that intersecting stigmas (e.g., based on race, sexual orientation, sex work, or incarceration) combine to influence disparities in risk, screening, treatment, and survival. Following this framework, MSWs may experience overlapping stigmas that augment HIV risk and lower engagement in HIV-related services. This may increase their vulnerability as a subset of MSM, while decreasing access to biomedical HIV prevention. But to date, few studies have examined medical mistrust, perceived and anticipated discrimination, and disclosure in healthcare settings among MSWs compared to other MSM.

Qualitative methods can help identify the circumstances shaping medical mistrust and disclosure among MSWs compared to MSM who do not engage in sex work. We used individual interviews to explore experienced and anticipated discrimination in healthcare settings, mistrust, and disclosure among MSWs and other MSM in light of PrEP implementation needs. PrEP is one of many biomedical HIV prevention technologies, and many MSM and MSWs may not be interested in or eligible for PrEP. But PrEP provides a timely example of biomedical HIV prevention, and lessons regarding PrEP acceptability and access may apply to the implementation of other HIV prevention technologies currently under development, such as topical rectal microbicides, long-acting injectable methods, and vaccines.³⁶ Each of these new methods will likely involve prescription or administration in healthcare settings, and PrEP can illuminate barriers to implementing HIV prevention through clinical care. All participants here were MSM, and we contrasted findings among participants who reported recent sex work and participants who did not.

Methods

This study took place in Providence, RI and was approved by the Yale Human Subjects Committee and the Miriam Hospital IRB. Methods have been reported elsewhere.³⁷ Data were collected as part of a qualitative PrEP acceptability study. We conducted 56 semi-structured individual interviews between April 2013 and April 2014 to obtain in-depth narratives about PrEP acceptability and access to PrEP. We included English-speaking cisgender adult men of self-reported negative or unknown HIV status, who reported condomless anal sex with a man of positive or unknown HIV status in the past 6 months. Individuals who did not meet these criteria were excluded, along with any individual who had participated in a PrEP efficacy trial. Thirty-one interviewees reported selling sex in the past 6 months and are analyzed as MSWs; the remaining 25 are analyzed as MSM who did not engage in sex work.

We recruited MSWs through in-person outreach and advertising in sex work venues and community-based organizations. We recruited other MSM through outreach and advertising in entertainment venues, clinics, and online and print media. Besides the 56 enrollees, 51 individuals were deemed ineligible during screening, two eligible individuals made appointments but did not attend interviews, and one eligible individual was interviewed, but then made statements that raised doubts about his eligibility. The 51 ineligible individuals were excluded for the following reasons, with all behaviors reported for the past 6 months: 28 reported that they were certain that all their male anal sex partners were HIV-negative individuals, eight had no male anal sex partners, seven always used condoms when having anal sex with male partners whose HIV status was positive or unknown, five had no male oral or anal sex partners, and three had not had sex. Each participant received $75.

We anonymized all procedures as follows: when we screened individuals for study eligibility, we assigned each person a research ID number and conducted screening using the ID number only. If an individual was eligible and wished to enroll in the study, we asked him to choose a pseudonym to schedule the interview. We retained a contact phone number for the participant until the interview occurred, and we then immediately destroyed the number. When interviews took place, we obtained verbal informed consent so as to avoid creating records that might identify participants. During interviews and analyses, we used participants’ research ID numbers only, rather than using any identifiers. We also obtained an NIH Certificate of Confidentiality to protect sensitive data regarding drug use and sex work.

Participants completed a 10-min written questionnaire reporting demographics and behaviors, followed by a 60-min semistructured, one-on-one interview with either the principal investigator [KU] or another interviewer [CC]. Interviewers identified themselves as non-physician researchers and provided information about tenofovir with emtricitabine as PrEP. Data collection occurred in private rooms in a clinic or needle exchange. Interview topics included healthcare and HIV/STI testing experiences, PrEP knowledge, willingness to use PrEP, attitudes about risk behavior during PrEP use, and preferences for PrEP information messaging.

Interviewers completed a verbal field note process after each interview to assess data saturation on main themes, which included a verbal summary of findings and themes for each agenda section, comparisons with past interviews, and comments on any unique results. These verbal field notes were audiorecorded, transcribed, and analyzed with interview findings. The principal investigator reviewed the audiotaped and transcribed notes throughout the study to monitor saturation, ending data collection when saturation was reached in both subsamples. Interviewers also discussed saturation and key findings together after each interview session. All interviews were audiorecorded, transcribed, and imported into NVivo 9.³⁸ We developed an initial thematic coding structure based on structural (question-based) codes, then added emergent codes (based on unexpected findings). The principal investigator initially began coding transcripts with two research assistants. Our original intention was to have the assistants work as independent coders, each coding a unique subset of transcripts. But in light of available training time and resources, we decided that multiple coders would instead be most helpful during an initial phase of coding to provoke discussion, reconsideration, and refinement of the thematic coding structure. The principal investigator and research assistants coded three transcripts together to establish familiarity with the draft coding structure, and then each research assistant double-coded four transcripts with the principal investigator. During coding sessions, the principal investigator and research assistants added emergent codes, considered different interpretations of themes, and resolved all coding disagreements through discussion, consensus, and refinement of the coding structure. At the completion of this process, the principal investigator discussed modifications of the coding structure with all coauthors, finalized the codebook, reviewed previously coded data for consistency, and recoded data where necessary. The principal investigator then coded all remaining interviews.

Once coding was complete, we reviewed coded text for each theme, sought points of consensus and divergence across interviews, and compared MSWs to other MSM. Because our goal was to contrast experiences between two subsamples, this analysis is descriptive and generally follows our deductive, question-based themes, using a positivist paradigm and reporting findings based on the surface meaning of participant statements.^39–41

Results

Table 1 reports sample characteristics; Table 2 reports illustrative quotes for key themes, along with the speaker’s awareness of PrEP, willingness to use PrEP, race, and ethnicity. (Although there were too few participants to disaggregate results based on race or ethnicity, we include this information in Table 2 to situate individual statements in the larger context of medical mistrust research.) We refer to men who engaged in recent sex work as “MSWs” and non-sex workers as “other MSM,” although all participants were men who have sex with men. Table 1 reflects demographic differences between subsamples; MSWs more frequently reported younger age, homelessness, unemployment, income below the federal poverty level, low levels of education, a lack of health insurance, lack of disclosure of MSM behavior to a primary care provider, unknown HIV status, heterosexual sexual orientation, attraction to women, drug use and injection drug use, and willingness to use PrEP. We have separately reported comparative findings on access to healthcare and HIV testing.³⁷ This analysis compares results regarding perceptions of healthcare discrimination, medical mistrust, disclosure of MSM behavior and identity in healthcare settings, and willingness to disclose risk behaviors to obtain PrEP. For each theme, we first consider common findings across subgroups, then describe unique findings for MSWs compared to other MSM.

TABLE 1.

Selected sample characteristics

	MSM who reported recent sex work (n = 31)	MSM who did not report sex work (n = 25)
Median age (range)	27 (22–58)	39 (21–70)
Race
White	77.4 %	76.0 %
African American	19.4 %	12.0 %
Native American	3.2 %	4.0 %
Asian	0.0 %	4.0 %
Refused	0.0 %	4.0 %
Hispanic or Latinoa	9.7 %	24.0 %
Housing
Homeless	29.0 %	4.0 %
Staying with friends/family	38.7 %	24.0 %
Renting home/apartment	32.3 %	52.0 %
Owns home/apartment	0.0 %	20.0 %
Income <$12,000 per yearb	51.6 %	16.0 %
Education
Did not complete high school	29 %	20.0 %
High school or GED only	35.5 %	28.0 %
Some college	32.3 %	24.4 %
Completed college	3.2 %	28.0 %
Employment
Disabled	6.5 %	12.0 %
Unemployed	67.7 %	16.0 %
Full-time job	6.5 %	28.0 %
Part-time/seasonal job	19.4 %	28.0 %
Other	0.0 %	16.0 %
Health insurance
None	67.7 %	32.0 %
Insurance through public sources	12.9 %	36.0 %
Private insurance	19.4 %	32.0 %
Time of most recent checkup
Past 6 months	41.9 %	36.0 %
7–12 months ago	22.6 %	20.0 %
1–2 years ago	22.6 %	24.0 %
More than 2 years ago	3.2 %	4.0 %
Does not know	9.7 %	16.0 %
Has a primary care provider (PCP)	38.7 %	56.0 %
Has a PCP and has disclosed MSM behavior to the PCP	12.9 %	28.0 %
Most recent HIV test
Past 6 months	48.4 %	48.0 %
7–12 months ago	25.8 %	16.0 %
1–2 years ago	19.4 %	16.0 %
Longer than 2 years ago	3.2 %	8.0 %
Never tested	3.2 %	12.0 %
HIV status
Unknown	35.5 %	16.0 %
Negative	64.5 %	84.0 %
Received a positive STI diagnosis (other than HIV) in past 6 months	3.2 %	4.0 %
Sexual orientation
Gay/homosexual	12.9 %	32.0 %
Mostly gay	6.5 %	12.0 %
Bisexual	41.9 %	40.0 %
Mostly straight	19.4 %	8.0 %
Straight/heterosexual	12.9 %	4.0 %
Other	3.2 %	4.0 %
Did not know	3.2 %	0.0 %
Sexual attraction
Only males	9.7 %	28.0 %
Mostly males	12.9 %	28.0 %
Males and females equally	29.0 %	28.0 %
Mostly females	38.7 %	12.0 %
Only females	6.5 %	4.0 %
Did not know	3.2 %	0.0 %
Top/bottom during anal sex with men
Always top	54.8 %	40.0 %
Usually top, but sometimes bottom	16.1 %	28.0 %
Top and bottom equally	19.4 %	20.0 %
Usually bottom, but sometimes top	3.2 %	8.0 %
Always bottom	3.2 %	4.0 %
Nonresponse	3.2 %	0.0 %
Had sex with both men and women in past 6 months	80.6 %	52.0 %
Median total number of sex partners in past 6 months (range)
Total number of partners	8 (2–150)	5 (1–50)
Male oral sex partners	4 (1–149)	4 (1–50)
Male anal sex partners	2 (1–80)	2 (1–50)
Female oral sex partners	3 (0–30)	1 (0–13)
Female vaginal sex partners	3 (0–20)	0 (0–15)
Female anal sex partners	1 (0–5)	0 (0–20)
Sex under influence of alcohol in past 6 months	64.5 %	64.0 %
Sex under influence of drugs in past 6 months	80.6 %	52.0 %
Used drugs multiple times per week in past 6 months	67.7 %	28 %
Injection drug use in the past 6 months	51.6 %	4.0 %
Shared needles or works with others in the past 6 months (among those who reported injection)	75.0 % (of 16)	0 % (of 1)
Had heard of PrEP outside the study	19.4 %	16.0 %
Willing to use PrEP
Yes	64.5 %	48.0 %
Maybe	6.5 %	4.0 %
No	29.0 %	48.0 %

PCP primary care provider

^aWe followed NIH guidelines to collect data on Hispanic/Latino ethnicity separately from data on race

^bThe approximate Federal Poverty Line for an individual ranged from $11,170 in 2012 to $11,670 in 2014

TABLE 2.

Illustrative participant quotes

MSWs: MSM who reported recent sex work	MSM who did not report sex work
Experienced and anticipated healthcare discrimination
Discrimination based on non-MSM characteristics (substance use, incarceration, poverty, lack of education, homelessness) as well as MSM behavior or identity.   • INT143 (PrEP-unaware, a willing, b White, Non-Latino): The hospital treated me very poorly …. [The doctors] were probably looking at me as some homeless person that didn’t have a job. I mean, I worked for years. I’m just, you know, I’m down at the moment…. They looked at me like I was trash or something.   • INT144 (PrEP-unaware, willing, White, Non-Latino): It’s you’re not treated like a normal person [by the doctors in the correctional institution]. Like they think just ‘cause you’re a convict…because you’re in jail you’re an asshole…. [It’s in] the way they talk to you, the way they look at you…. Like, “Here, now, sit,” you know what I mean. Not like you know, “Can you come take a seat and discuss what, what, yeah, what’s going on.”  Some willingness to switch doctors after experiencing discrimination, but less common than MSM who did not report sex work.   • INT117 (PrEP-unaware, willing, African American, Non-Latino) I talk to [doctors] like I talk to you. I mean [sex] is nothing to be ashamed of ‘cause you are who you are …. It’s your health…. It was kinda awkward with him, with the doctor… I made him uncomfortable… where he had to basically [leave] and then come back five minutes later. I could see he left only because of where I was taking [the conversation]…. And that’s when I went and got another doctor…. If I feel as though you’re rushing me out the door or… sidestepping me… I will complain.  Concern about substance use-related stigma, including whether doctors believe participants feign pain to obtain medication.   • INT131 (PrEP-unaware, willing, White, Non-Latino): Sometimes I get a really nice doctor, that seems like they might want to, might care, might help, and sometimes you might get a doctor who’s like, “Oh this guy’s been here a few times, he’s just looking for drugs. Get him out the door”-type thing. ‘Cause you know, sometimes, you probably are [looking for drugs], but sometimes you’re actually hurt and it’s kind of like a boy who cried wolf type thing.	Discrimination based on MSM behavior or identity.   • INT116 (PrEP-unaware, willing, White, Latino): When I go to the emergency room every time they ask me what I am, I say I’m gay. The first thing they say is, “Do you want a HIV test?”, and … it’s kind of insulting…. Like I get mad.   • INT111 (PrEP-unaware, unwilling, White, Non-Latino): I had a doctor… who told me that I should stop [having sex with men] and that it’s very unhealthy or risky to my health … and like unnatural…. It didn’t put me off because I was like whatever. No matter what whether you’re a doctor or not, there’s people there that don’t, that just are disgusted by homosexuality…. It had to do with like whatever his upbringing or like, whatever religion he was or something.  Some concern about stigma related to drug use, but less common than MSM who reported sex work.   • Interviewer: What makes you comfortable in [the free clinic you go to]? INT147 (PrEP-unaware, unwilling, White, Latino): The people, the attitude, the way they treat you.…[R]espectfully, you know…. They don’t uh look down on ya. You know they don’t treat you different because you’re a drug addict or a homosexual or whatever. You’re just a human being.  Willingness to switch doctors after experiencing discrimination on the basis of MSM behavior or identity.   • INT115 (PrEP-aware, willing, White, Non-Latino): Um [I had] one older Indian doctor who told me to um avoid gay people like the plague …. Then I changed doctors … he seemed to be decent, but … I just felt like there was just an aspect of like, um, understanding the gay culture and… what we have to go through … and he didn’t understand that at all …. I [found a clinic advertising at] Pride [and] I was so happy that there was a, a place that I could go to where I could just openly explain, like, “Okay, this is my situation.”
Trust of healthcare providers
Medical mistrust was linked to the following factors: attributing substance use disorders to overprescription by doctors; perceptions that doctors are motivated by money; perception that doctors are unreceptive to patients’ self-diagnoses or expertise.   • INT101 (PrEP-unaware, unwilling, White, Non-Latino): I was addicted to OxyContin and prescription pills. I think [doctors] suck … [b]ecause I think they give, they give out a lot of things to people that don’t need them … They was offering way too much and I was getting whatever I wanted…. I really feel as if it’s all money, you know, it all revolves around money.  General trust in doctors.   • INT135 (PrEP-aware, unwilling, Native American, Non-Latino): No, [I’m not suspicious of doctors … I believe them] as long as it’s a medical opinion, sure…. I’ve never had a disagreement or … a problem with a doctor.  Negative perceptions of providers in correctional institutions.   • INT130 (PrEP-aware, unwilling, White, Non-Latino): Doctors in the [correctional institution] are hard. Assholes.… Because they deal with assholes all day …. I wouldn’t be a happy camper if I worked [there] either.… Like I said you know, “Give me your arms, ah fuck, I got another junkie.”…It’s not like you’re gonna have a conversation with this person.  Positive experiences with substance use treatment clinicians   • INT155 (PrEP-unaware, willing, White, Non-Latino): [Healthcare at the substance use clinic] was really good. everyone was uh, they really wanted to help you, you know and um they offered me a lot and they had like you know good, good meetings and just to like make us more aware of um like different situations and stuff so.	Medical mistrust was linked to the following factors: concerns about impersonal care, unnecessary prescriptions, or unsympathetic providers.   • INT124 (PrEP-unaware, willing, White, Non-Latino): [Doctors] don’t spend too much time with you…. And, uh, sometimes you feel like you’re on a conveyor belt, you know?…. Impersonal. That’s the word.  General trust in doctors.   • INT122 (PrEP-unaware, willing, White, Non-Latino): Doctors are there … not to judge you, just to help you … That, that’s all they’re out to do. So I mean I’m always as honest as I can. I mean I, I, if there’s something going on and I’m having trouble, I explain it to ‘em and they tell me what the, they’re there to help me.  Positive relationships with individual doctors, often PCPs or psychologists.   • INT148 (PrEP-unaware, willing, Native American, Non-Latino): I have a very good rapport with [my doctor]… he’s really a, a good [doctor], that’s also probably why I just stay with him … he knows most of my history…. his bedside manner [makes me comfortable], um, he, he listens.   • INT152 (PrEP-aware, unwilling, White, Non-Latino): I have pretty much complete trust in [my PCP]…. She listens to not just the physical side of it, but the mental if there’s something bothering me.
Disclosure of MSM and sex work behavior to providers
Positive disclosure experiences   • INT125 (PrEP-unaware, willing, African American, Latino): [I have disclosed to my doctor] several times… I get a little bit nervous … I’m a top, and when she’s thinking about anal, I’m like, “Do you mean me or do you mean the other person?”…. And she’s like, “Oh, that’s right, that’s right, you don’t do that”…. [She’s] very accepting.  Nondisclosure of sex with men and sex work; disclosure barriers; nondisclosure in group treatment.   • INT140 (PrEP-unaware, willing, African American, Non-Latino): [I don’t disclose MSM behavior at the hospital because] over there they talk a little loud…. “Oh, you’re being seen for gonorrhea, chlamydia,” all loud. I get so embarrassed with the people in the next room and everybody knows your business in there, you know? …. Even, God forbid, if I had HIV or AIDS, they, it would be on your record. And it’d be like they know who I am.   • INT132 (PrEP-aware, willing, White, Non-Latino): I can’t talk [about sex work with doctors], I don’t know. Listen, that’s my biggest problem with getting clean. [E]very treatment center I go to, I uh, I have to lie about my, my life…. I can’t sit in a crowd of people and say, “Yeah, I fucking, let 70 year old men fucking blow me every day” …. So I end up leaving … my issue doesn’t get resolved because I, I can’t even talk about it with anybody.  Many men in this subsample believed that MSM behavior is irrelevant to medical care, which was a barrier to disclosure.   • Interviewer: Do you tell doctors that you … have had sex with men, or would you tell them if they asked it? INT103 (PrEP-unaware, willing, White, Non-Latino): I mean, if it came up, I would, but I’m not just bring[ing] it up for nothing.   • INT135 (PrEP-aware, unwilling, Native American, Non-Latino): I don’t [talk about sex with men with doctors], why would I?  Experiences with indirect disclosure.   • INT142 (PrEP-unaware, willing, White, Non-Latino): I asked questions about sex with men you know … Not kinda like directing it towards me, like just speaking of it in general you know…. You know you try and beat around the bush you know so he didn’t, like, put two and two together.  Facilitators to disclosure included provider gender (some preferred male doctors) and the perception of professionalism.   • INT143 (PrEP-unaware, willing, White, Non-Latino): [Disclosure is] just it’s easier when you’re dealing with professionals. Like I used to come [to the needle exchange] for my syringes and my stuff for my drugs and I never once got a nasty look or you know…. They were just like, be safe.	Positive disclosure experiences.   • INT156 (PrEP-aware, unwilling, Asian, Non-Latino): Actually, it’s been pretty open… my primary care physician asked me right, right away whether I need anything related to, to HIV prevention… and I’ve always been very comfortable with, um, talking to him about, about this issue… I think he really asks everyone [about HIV]… because it was our first meeting and he didn’t really know anything about me.   • INT114 (PrEP-aware, maybe willing, White, Latino): Most clinics that I’ve been to they, they ask you um, I guess that’s part of the questionnaire… In the application um at [free clinic] it says are you straight, um gay, bisexual, lesbian and, and you just circle which one.  Barriers to disclosure included the belief that disclosure is irrelevant to care, cultural norms, and concern about discrimination. Many participants reported willingness to disclose in response to direct questions.   • INT118 (PrEP-unaware, unwilling, African American, Non-Latino): I have no problem talking with my doctor about sex, but it doesn’t come up… because I don’t need, uh, Viagra. If that came up, then we’d talk about it every time… I’d bring it up… [But] I don’t have that problem… We never talked about [sex with men]… because that’s my personal business… It might gross her out…. She could say, “I don’t want him as a patient anymore.”  Facilitators to disclosure included provider gender (some preferred female doctors), age, longstanding relationships; direct questions about sexuality.   • INT151 (PrEP-unaware, willing, White, Non-Latino): Some [doctors] are still squeamish about gay men. It’s hard to talk to gay men because it’s a huge threat to straight men …. So it’s hard to bring up gay issues with any kind of comfort. Personally I go to women doctors because I just find it easier to talk to them.   • INT119 (PrEP-unaware, willing, African American, Non-Latino): I have no problem at all [disclosing MSM identity] to doctors… If they ask I would just flat-out tell them… [but] I wouldn’t just go right out to doctors on the blue.
Disclosure willingness for obtaining PrEP
Facilitators and barriers to disclosing behavior in order to obtain PrEP.   • INT130 (PrEP-aware, unwilling, White, Non-Latino): I feel if I went to a detox and said, “Hey, you know, I did this and that and the other thing and what about this [PrEP]?” I feel that you know even if they didn’t know anything about it … they’d go and research it for me …. You wanna have a doctor that knows you [to get PrEP]. Then you can be more honest…. He’s there for your wellbeing … hopefully not just to take home a paycheck.   • INT131 (PrEP-unaware, willing, White, Non-Latino): If I could get [PrEP] from say like an ER doctor? Yeah [I would disclose]. If I had to go to my personal doctor, I probably wouldn’t ask [for PrEP]…. [M]y most recent personal doctor was also my, my ex-wife’s doctor, and my kid’s doctor, the same doctor, and yes, they’re not supposed to share anything, but I wouldn’t, I wouldn’t even take the chance. That’s how I lead the separate lives.  Indirect or limited disclosures to obtain PrEP.   • INT134 (PrEP-aware, willing, White, Non-Latino): Even girls, like girls can get [PrEP] too, right? Interviewer: Yeah…. Would you say that [sex with] girls was the reason why you need [PrEP]? INT134: Yeah, I’d probably say that, yeah.	Facilitators and barriers to disclosing behavior in order to obtain PrEP.   • INT123 (PrEP-aware, willing, White, Latino): To get prescribed [PrEP, I’d first talk to] my psychiatric doctor. He could probably point me to the right person that I could speak to…. And I could feel comfortable ‘cause I’m a person that I’m fairly nervous [talking about sexuality]…. I’m gonna show these papers [PrEP educational materials] to my doctor.  Indirect or limited disclosures to obtain PrEP.   • INT153 (PrEP-unaware, unwilling, White, Latino): [In order to get PrEP] I’d just, I, I’d let them know that, you know, me and my partner are no longer together, that I’ve been messing around a little bit more and I feel like just as a precaution I would like to take this pill…. I’d just say I increased sexuality, my sexuality was increased a little bit more than normal, so I’ve been out there a little bit more. Interviewer: Okay. So you’d kind of pick and choose what you would disclose. INT153: Yes.

^aPrEP-aware: participant had heard of PrEP before joining the study; PrEP-unaware: participant had not heard of PrEP before joining the study

^bWilling: participant reported willingness to use PrEP. Unwilling: participant reported that he would not want to use PrEP. Maybe willing: participant was uncertain about willingness to use PrEP

Experienced and Anticipated Healthcare Discrimination

Common Findings

Participants in both subgroups reported past discrimination experiences and anticipated discrimination in healthcare facilities on the basis of MSM behavior and identity. Past experiences were more frequently reported by MSM who did not engage in sex work, but this may be due to more frequent disclosure among men in this group. Although men in both groups reported willingness to change providers after experiencing discrimination, somewhat more MSM who did not engage in sex work described changing providers (perhaps linked to greater accessibility of health insurance in this subgroup).

MSWs

In keeping with an intersectionality framework,³⁵ MSWs reported several types of healthcare discrimination based on stigmas other than MSM behavior or identity. MSWs tended to attribute perceived discrimination to histories of substance use or mental illness, homelessness, race, unemployment, poverty, or incarceration. Providers’ actions were considered to provide evidence of bias, including rudeness, indifference, or dismissiveness; unwillingness to discuss topics directly or provide complete information; excessively long waiting times or short visits; rough physical treatment during procedures; denial of pain medication; and dismissiveness of a patient’s self-diagnosis or expertise.

A recurring concern among MSWs was that clinicians believed that they feigned pain and manipulated clinical interactions to obtain pain medication for misuse or sale, an anticipated stigma linked to substance use histories. Some men reported deliberately seeking pain medication in the past, but emphasized that they experienced legitimate pain on other occasions when clinicians refused to prescribe medications. These participants tended to classify doctors’ refusal to prescribe pain medications as a form of discrimination on the basis of substance use history, which was not described in the subgroup of MSM who did not engage in sex work.

A few MSWs also believed they had played a role in unpleasant interactions. Several described hostile or destructive behavior in medical settings, which they attributed to drug withdrawal, relapse, mental illness, pain, or frustration. Several also expressed regret at having relapsed while under care for substance use treatment, believing that relapses had harmed their relationships with providers.

Numerous men in the MSW group reported positive care experiences with providers at needle exchanges and substance use clinics; experiences with discrimination were more frequently described as occurring in emergency rooms and correctional facilities. Many believed that substance use treatment clinicians are helpful, that they sincerely appreciate patients’ efforts to complete treatment, and that they respect patients regardless of substance use or inability to pay.

Other MSM

Fewer MSM who did not engage in sex work reported healthcare discrimination due to characteristics other than MSM behavior or identity, suggesting that MSM who do not engage in sex work may experience fewer overlapping stigmas compared to MSWs. Several MSM described past or anticipated discrimination on the basis of substance use, but this was rare compared to MSWs. Negative treatment was most frequently linked to disclosure of same-sex sexual behavior or gay/bisexual identity. MSM identified several provider actions as proof of discrimination, including admonitions about sexual behavior, not asking follow-up questions after a behavioral disclosure, perceived discomfort, or unwillingness to discuss sexual health. Several participants reported interactions with PCPs who advised them to “avoid gay people” or MSM behavior due to health risks. One man also described feeling insulted when a provider’s first response to behavior disclosure was to offer an HIV test. MSM who had experienced discrimination, however, sometimes attributed experiences to individual providers’ prejudices against gay men, rather than a problem affecting the medical profession. Some also believed that discrimination against MSM in healthcare was decreasing, particularly among younger providers.

Trust and Mistrust of Providers

Common Findings

Both MSM and MSWs reported a desire for doctors to validate and acknowledge their expertise about their own health conditions, wishing that providers would be more receptive to patient self-diagnoses and requested courses of treatment. MSWs reported willingness to ask questions during medical care in every setting except correctional institutions. Other MSM also reported willingness to ask questions during care, including raising medical needs unprompted by the provider. Although only one participant reported asking his provider for PrEP, the willingness to ask for information and specific treatments in general medical care may be a facilitator for seeking PrEP. Men in both subgroups believed that most providers would research health questions in cases of uncertainty.

MSWs

MSWs expressed mixed feelings about providers, particularly in hospital and correctional settings. Most men in this subgroup had a general confidence in doctors’ diagnosis and treatment skills, often accompanied by the reflection that they have little choice about trusting providers when they need care. But many MSWs also blamed providers for substance use disorders due to (perceived) mismanaged care and overprescription of pain or psychiatric medications. MSWs also reported that providers are “only in it for the money,” and some believed that providers use free samples to “hook them” on expensive medications. Some suggested that clinicians are unaware or dismissive of treatment barriers for low-income patients (e.g., transportation, housing instability, inability to pay), leading to unrealistic treatment expectations. MSWs also worried about breaches of confidentiality in hospitals and other facilities with non-clinical staff. Several MSWs suggested that correctional institution clinicians are exclusively focused on ensuring that sick men do not endanger the incarcerated population, rather than meeting individual patient needs.

In contrast to men who expressed mistrust, some MSWs also reported trusting relationships with providers, often in substance use and mental health treatment settings. Positive relationships with PCPs and psychiatrists were among these favorable experiences.

Other MSM

Provider mistrust was less frequently reported by other MSM. Some MSM reported “seeing doctors as people” with a range of competence and interpersonal skills. Many reported favorable experiences with individual doctors, and more MSM described having long-term and close relationships with providers. Men in this group more frequently mentioned family or acquaintances who were clinical providers; these social networks served as informal sources of medical expertise and may facilitate positive general impressions of providers. Compared to MSWs, more MSM discussed switching providers when they felt unsupported, uncertain about their care, or given inadequate time or information. This may also be due to the fact that MSM received routine and preventive care more frequently, rather than care in emergency rooms or correctional settings with less leeway for patient choice.

Disclosure of MSM and Sex Work Behavior to Providers

Common Findings

A common disclosure facilitator was the belief that disclosing MSM behavior or sex work is important to improve care, especially mental health treatment and requests for HIV/STI testing. Common disclosure barriers were anticipated discrimination, embarrassment, fear of confidentiality breaches, the perception that sex is irrelevant to care, and the desire to keep sex private. Many participants in both groups reported willingness to disclose MSM behavior if providers specifically asked about the gender of sexual partners, but they were unwilling to disclose spontaneously. For these men, providers’ failure to ask specifically about sex or partner gender was seen as proof that disclosures are irrelevant to medical care. Men did not report feeling offended by direct questions about sexual identity or behavior.

MSWs

Few MSWs reported having disclosed either sex work or MSM behavior to providers. When MSWs did disclose, these disclosures were more frequently described as occurring in settings related to mental health and substance use treatment. Some reported positive disclosure experiences, although several MSWs perceived physician discomfort. Some MSWs also reported “indirect” disclosure, such as by asking general questions about sex with men without disclosing any personal history. One unique barrier to disclosure was the receipt of care in group treatment settings, generally in substance use treatment. MSWs also identified a range of disclosure facilitators. Many preferred long-term relationships with providers, while several reported more comfort with unfamiliar or geographically distant doctors to avoid second-hand disclosure to partners or family. Other facilitators included provider gender (some preferred men; others did not state a preference), age (preferences differed), and visible evidence of professionalism (e.g., diplomas). MSWs reported more comfort disclosing injection drug use and heterosexual sex compared to disclosing MSM behavior and sex work.

Other MSM

Among other MSM, disclosure to clinical providers was more common but not universal. Men most often reported disclosing to PCPs, specialists, and mental health providers. Experiences were generally positive, although some reported changing providers. Disclosure facilitators included provider gender (some preferred women; others did not state a preference), having multiple visits over time with one provider, the belief that the provider has a sense of humor or openness to disclosure, having MSM-specific or HIV-specific health questions, having a provider that is one’s own age, and obtaining healthcare in a venue known to support gay patients. Disclosure barriers included feeling vulnerable during medical exams, anxiety related to thinking about HIV, and low perceived HIV risk.

Disclosure Willingness for Obtaining PrEP

Awareness and willingness to use PrEP are reported in aggregate in Table 1 and individually in Table 2. Although PrEP awareness was low, a majority of participants in both subsamples reported that they would be willing to use PrEP, although none had been offered PrEP in a healthcare setting. One participant was using PrEP, but he had actively sought out an MSM-friendly provider, requested PrEP, and asked his provider to obtain clinical guidelines. Participants in both groups reported a willingness to disclose MSM behavior or sex work if they knew that a provider could offer them PrEP or PEP. Learning about PrEP changed participants’ perception that MSM behavior and sex work are irrelevant to medical care, suggesting that PrEP may provide a motivation to disclose. Among some men in both groups, however, the availability of PrEP may prompt disclosure of overall HIV risk, not specific MSM behavior, in hopes that this is enough to meet PrEP eligibility criteria. Several MSWs and MSM reported that they would describe their sexual risks as arising from female partners alone, in hopes of obtaining PrEP without disclosing sex with men.

Discussion

This qualitative study found key differences between MSWs and MSM in disclosure, mistrust, and perceived discrimination in healthcare. Compared to MSM who did not engage in sex work, fewer MSWs reported disclosing MSM behavior, more expressed mistrust of providers, and more described experiencing discrimination on the basis of substance use, homelessness, race, and poverty. (Compared to other MSM, more MSWs also reported substance use, poverty, limited education, homelessness, and non-gay sexual orientation, which may contribute to differences in disclosure and perceptions.) Our findings are consistent with the Stigma and HIV Disparities Model, which suggests that overlapping stigmas may combine to disadvantage population subgroups such as MSWs through increased medical mistrust and reduced access to care.³⁵ More MSM who did not engage in sex work described disclosing MSM behavior in clinical settings, but disclosure was not universal. Men in both groups anticipated healthcare discrimination on the basis of MSM identity, echoing prior findings that internalized homophobia may limit access to services.¹¹ Our findings also agree with research suggesting that men can feel stigmatized when providers immediately associate MSM behavior with HIV risk.⁴² Our results can contribute to literature on building cultural competence among clinical providers serving MSM and other sexual minority populations.^43–53 We also support recent calls for increased clinical training and cultural competency in MSM healthcare.^14,54

In particular, these findings may help providers facilitate behavioral disclosures needed to implement biomedical HIV prevention among MSM and MSWs. Providers might begin by educating patients about how behavioral disclosures can improve care, which may dispel the impression that sexual behavior is irrelevant to health. For example, providers can discuss PrEP, PEP, rectal and pharyngeal STI testing,⁵⁵ or other health needs that are experienced more frequently by MSM.^56,57 After identifying how disclosure can improve care, providers may then ask specifically about partner genders and transactional sex, rather than relying on spontaneous disclosure. Asking about sexual orientation or identity alone is insufficient, given disparities between sexual orientation and behavior reported in this and other studies.⁷ Our findings suggest that men seeking PrEP may emphasize other risks (e.g., sex with women or injection drug use) or ask general questions about MSM behavior rather than disclosing actively. Asking follow-up questions about sexual behaviors may be helpful. Emerging technologies such as tablet- or computer-based methods may also help solicit information,^25,58 and online resources may build skills and comfort in discussing sexual health with patients (e.g., www.lgbthealtheducation.org). Some men in this study reported feeling stigmatized when providers immediately offered HIV testing after disclosures; providers might wait several minutes after disclosure before offering testing, or explain that testing is offered routinely to all patients. Our results also highlight the need for complementary outreach preparing men to talk with providers about HIV, including processes of disclosure, managing discrimination, and communicating with providers who are unfamiliar with new prevention strategies. Several organizations have prepared PrEP education brochures with information for both providers and users,^59,60 and patients can consult online directories of LGBT-friendly medical providers.⁶¹

Comparing subsamples on the basis of transactional sex identified several ways in which male sex workers may experience heightened barriers to disclosure and receipt of biomedical HIV prevention services, particularly mistrust and anticipated discrimination on multiple bases. Pursuant to an intersectionality framework, Earnshaw and colleagues have suggested approaching stigma reduction through a focus on resilience, including economic empowerment, community capacity, the development of common identities between patients and providers, social support, and interventions based on coping behaviors.³⁵ All of these approaches may be useful for reducing HIV risk among MSWs. For providers interested in delivering biomedical HIV prevention services to MSWs, it may help to receive training on how overlapping stigmas may complicate disclosures, and to emphasize shared goals and potential shared identities with MSW patients.

Our approach has several strengths. Our methodology captured a wide range of experiences, and we accessed marginalized population of street-based MSWs to provide a novel in-depth comparison of MSW and MSM in the PrEP context. Our findings are also limited. We collected data in clinic and needle exchange settings, which may bias findings if individuals who were uncomfortable in clinics chose not to participate. Participants tended to be socioeconomically disadvantaged and less gay-identified than samples in other MSM studies, all were English-speaking, and most were white and non-Latino. Future research with a more diverse population of MSWs—including MSWs who meet partners in non-street-based settings, such as online²⁹—is needed to understand how stigmas intersect for men who engage in transactional sex. Our results in the male sex work subsample may be most transferable to other populations of street-based MSWs in urban U.S. areas, particularly populations that also experience concurrent disadvantages such as substance use and homelessness.

Our comparative approach also has some limitations. Comparisons between MSM and MSWs reflect not only differences due to sex work itself but also demographic differences that may be associated with sex work engagement. Our methods for recruiting MSWs and MSM also differed slightly. Most MSWs were recruited through in-person outreach in sex work venues (80.6 %), followed by media advertising (16.1 %) and word of mouth (3.2 %). MSM who did not engage in sex work were most commonly recruited through advertising in online and print media (80.0 %), followed by in-person outreach in entertainment venues (12.0 %), and word of mouth (8.0 %). This may introduce additional differences between the subgroups. For example, MSWs may have been reluctant to screen for our study without in-person assurances regarding confidentiality and legitimacy, particularly given that the study was associated with a local hospital. Providing these assurances during recruitment may have influenced not only sampling but also participants’ willingness to share information during interviews. Compared to MSWs, MSM who did not engage in sex work may have been more comfortable calling the study phone line or responding to study advertisements (which disclosed an affiliation with the hospital); men in this group may also be better served by local media targeting the MSM population, and they may have better access to working phones. We cannot rule out the possibility that some men in the non-sex-work subsample had engaged in sex work without disclosing it. A number of MSWs, however, mentioned they were comfortable disclosing to us for several reasons: they trusted our community contacts who assisted in recruitment, they would not have a recurring relationship with interviewers, data collection was anonymous and protected by a Certificate of Confidentiality, and interviewers were specialists in HIV and sex work.

Future studies should use quantitative methods to expand these findings, and to evaluate strategies for improving communication between providers and MSM patients, including MSWs. Further research may also consider conspiracy beliefs. We did not probe these beliefs here, but during our prior focus group study among MSM and MSWs,⁶² several MSW participants suggested that HIV had been designed to eliminate substance users. Conspiracy beliefs can contribute to medical mistrust, HIV risk, and nonadherence to antiretroviral drugs^22,23,63,64; further research is needed to understand how conspiracy beliefs may affect PrEP use among MSWs. Addressing medical mistrust, disclosure, and cultural competency in healthcare for MSWs and other MSM can help expand biomedical HIV prevention access in these high-priority populations.