Introduction

Oral antiretroviral pre-exposure prophylaxis (PrEP) is safe and effective for HIV prevention [1]–[5]. The US Centers for Disease Control and Prevention recently released guidelines for clinical providers regarding PrEP implementation [6], augmenting prior interim guidance [7]–[12]. Among populations at risk for HIV infection, the availability of PrEP will depend in part on access to HIV testing and clinical providers [13]. PrEP prescribers in the US have included nurse practitioners, physician assistants, and physicians trained in family practice, internal medicine, emergency medicine, and infectious diseases [14], [15]. But uptake has not been rapid [15], so in order to identify PrEP implementation barriers, we need to understand where and how at-risk individuals encounter healthcare professionals. Where PrEP implementation guidelines recommend oversight by physicians and other prescribers, access to providers and ability to discuss PrEP with doctors may be key concerns among individuals considering PrEP use [16]–[20].

Ability to pay for PrEP prescriptions, clinical visits, and other components of PrEP care is another central concern. In the US, most individuals continue to obtain health insurance through employment benefits, with smaller proportions obtaining insurance through individual coverage and public insurance programs with defined eligibility criteria [21]. Although estimates vary, the approximate national average cost for ensuring a single individual through employer-based health insurance was $5,384 in 2012, of which the employee paid approximately $1,118 [22]; for individuals purchasing private insurance on the individual market, 2010 data suggest a monthly average premium of $215, yielding an annual cost of $2,580 [23]. In Rhode Island, where we conducted this study, these annual out-of-pocket premium expenses reached $1,335 for individuals insured through employers and $4,128 for those insured through the individual market [22], [23]. Access to both private and public insurance is now growing under the 2010 Patient Protection and Affordable Care Act, particularly in states that expanded the Medicaid program, which offers coverage to low-income individuals [24]. Several prior studies, however, have documented above-average proportions of uninsured individuals among populations at elevated risk for HIV infection, including men who have sex with men [25], [26]. The full cost of tenofovir with emtricitabine (TDF-FTC) for PrEP in the US is approximately $17,000 per year for the medication alone [27]; the total costs of PrEP per individual per year including laboratory services and professional fees approach approximately $18,000 (not including any treatment for side effects). Even among insured individuals, coverage for PrEP may be incomplete, and cost-sharing requirements such as co-pays or co-insurance may limit access [27], [28]. Although the manufacturer of TDF-FTC has instituted a medication assistance program, the effect of this program on access to PrEP is still under study [27].

MSM are a high-priority population for PrEP, given that MSM transmission is associated with the majority of new US infections [1], [29]–[32]. Research is beginning to explore healthcare access among MSM in light of specific PrEP implementation needs [33], as incomplete access to HIV testing and clinical care may inhibit PrEP uptake in the US setting [29], [34]. These impediments may add to other barriers to PrEP use, such as low perceived need or concerns about side effects, drug resistance, efficacy, adherence demands, and behavioral impacts [19], [35]. Suboptimal access to healthcare and HIV prevention services among MSM has been linked to distrust of providers, difficulty disclosing MSM status, minority race and ethnicity, poverty, young age, unknown HIV status, and experienced and internalized homophobia [36], [37]. Reluctance to disclose same-sex sexual behavior to providers may also create barriers to PrEP uptake [26], [38].

Among MSM, male sex workers (MSWs) have specific HIV prevention needs due to elevated numbers of partners, condomless sex, poverty, substance use, prevalence of other sexually transmitted infections (STIs), experiences of violence, difficulty negotiating condoms with clients, sex with male and female partners, low rates of behavior disclosure to clinicians, and difficulty accessing medical care [39]–[47]. Social marginalization and stigma exacerbate barriers to accessing social support, medical care, and HIV prevention services [40], [47]. Sex work itself is associated with additional risk factors, including substance use, unemployment, mental health problems, and childhood sexual abuse [43], [48], [49]. Risks for HIV and other STIs are particularly acute among street-based sex workers compared to MSWs who meet clients in other venues, such as the Internet or escort agencies; for example, compared to their Internet-based counterparts, street-based male sex workers have reported higher rates of HIV risk behavior, lower education, greater unemployment, lower fees for sexual services, more sex work to meet survival or substance use needs, and less negotiation power to refuse condomless sex [43]. Prevalence estimates of HIV among MSWs often exceed estimates among other MSM, including in North America, where studies have documented prevalence of 5–31% [47]. Sex work is illegal and highly socially stigmatized in most of the United States, including Rhode Island [50], [51], where we conducted the present study. Although there have been calls to investigate the potential implementation of PrEP among male sex workers [47], [52], little is known about PrEP acceptability among any subgroup of MSWs, including street-based sex workers.

To improve PrEP implementation, an in-depth understanding of how MSM and MSWs access healthcare is needed. This study used qualitative methods to solicit detailed narratives from these populations exploring healthcare access, HIV/STI testing, unmet healthcare needs, and preferred PrEP providers. The MSWs recruited for this study were primarily street-based; that is, they met clients on the streets or in adult bookstores, and they worked independently rather than through brothels or agencies.

Methods

We conducted a two-stage qualitative study in Providence, RI. We identified initial themes using focus groups, and then obtained in-depth narratives in individual interviews. All participants met the following criteria: adult biological male; English-speaking; self-reported negative or unknown HIV status; and self-reported condomless anal sex with a man of positive or unknown HIV status in the past 6 months. None had participated in a PrEP efficacy trial. We recruited participants by direct outreach in entertainment venues, sex work venues, community-based organizations, and clinics, and by advertising in local media serving MSM. We collected data in private rooms in clinics. Participants could enroll in both stages, and a small number (approximately 10) did so. But because we anonymized individual interview procedures, we cannot comment with certainty on the number of overlapping participants.

Stage 1 enrolled 38 men in eight focus groups, with 4–6 men per group; three groups were designed to sample street-based MSWs (n = 16) and recruited participants in street-based sex work venues. All men in these groups were MSM who verbally disclosed sex work. Five groups sampled other MSM without focusing on street-based sex workers (n = 22). Five men in these groups disclosed sex work on our demographic questionnaire, and we counted them as sex workers for analyses; we do not know, however, the venues in which these men met their partners (e.g., on the street, in bookstores, on the Internet). These 5 men were recruited through online sources, a flier, and a referral from another study; the 17 MSM who did not report sex work reported learning about the study through online and print media, fliers in bathhouses, clinics, a needle exchange, and word of mouth. Methods for Stage 1 have been reported elsewhere [53]. Groups occurred in February–June 2012.

Stage 2 enrolled 56 men for semi-structured interviews from April 2013 to April 2014. Thirty-one men disclosed selling sex for money, drugs or other goods in the past 6 months, while 25 did not engage in sex work. Of the 31 MSWs, 24 were recruited in street-based sex work venues; the remaining 7 MSWs were recruited through online media, a needle exchange, friends, and a local newspaper. MSM who did not engage in sex work reported learning about the study through fliers at clinics and other venues serving MSM, online and print media, and word of mouth. We screened 92 individuals for focus groups and 110 individuals for interviews. No participant dropped out of the study, no repeat interviews were conducted, and we did not re-contact participants after data collection. We halted data collection after reaching data saturation for main themes in each subgroup (MSM and MSWs).

All participants completed written questionnaires capturing demographic and behavioral information, including information on sexual behavior, substance use, and the CAGE questionnaire for alcoholism [54]. The questionnaire asked about employment but did not specify whether sex work qualified as employment. Given the large percentage of MSWs reporting unemployment, as well as participants’ statements about engaging in sex work due to inability to find or maintain employment, we believe that men did not perceive sex work to be a form of employment for the purposes of this questionnaire. A primary facilitator led qualitative data collection in focus groups (KU), and a secondary facilitator observed group dynamics and took notes. Each interview was conducted by one facilitator (KU or CC), with debriefing to monitor data saturation. All facilitators had prior experience and training leading qualitative interviews.

Facilitators in both stages identified themselves as non-physician researchers funded by the National Institutes of Health, and without a financial interest in PrEP. We provided participants with information about tenofovir co-formulated with emtricitabine for PrEP (e.g., trial findings, FDA approval, side effects, dosage). Focus groups lasted approximately 2 hours, and interviews lasted 90 minutes. Each participant was compensated $75 total for his time and transportation costs. Focus groups and interviews were guided by written agendas; both included modules on PrEP knowledge, willingness to use PrEP, beliefs and intentions about behaviors associated with PrEP use, and comprehension of messages about PrEP efficacy. Individual interview agendas included separate sections discussing access to healthcare, HIV and STI testing, healthcare discrimination, healthcare needs, and barriers to accessing PrEP. Healthcare access was not a planned section of focus groups, but facilitators probed incidental comments about healthcare and analyzed emergent findings. Emergent findings are particularly robust because they arise from spontaneous comments (unprompted by facilitators). This paper reports results regarding access to healthcare and preferred PrEP providers. Findings on overall PrEP acceptability, PrEP messaging, and perceived healthcare discrimination have been presented in conferences [55]–[59] and are being reported in separate publications.

Sessions were audio-recorded, transcribed, and entered into the NVivo 9 program [60]. We developed thematic codes to analyze focus groups and individual interviews, and we applied codes to each verbatim line of text. Initial themes were derived deductively, based on the planned goals of the interview agenda [61]. In the focus group stage, these included barriers to accessing PrEP; in the interview stage, these themes included access to healthcare, unmet healthcare needs, HIV and STI testing behaviors, and access to PrEP. We then identified additional themes derived inductively from the data. In the focus groups, these included prior experiences with clinical care, health insurance barriers to accessing care, preferred PrEP providers, and reflections on the role of providers in PrEP prescription. In interviews, these themes included perceptions of provider-patient relationships (past, present, and ideal), and preferred PrEP providers. Focus groups were double-coded by two independent coders, with discrepancies resolved by discussion and consensus. Interview transcripts were coded by a single coder, and a subset were double-coded by one of two trained research assistants. We examined all coded text for each theme and compared text across interviews and groups to identify analytical findings, seeking both consensus and points of divergence. The findings of this paper are largely reported according to our deductive, question-based categories, using a positivist paradigm [62], [63]. This approach may also be characterized as “semantic” or explicit thematic analysis; given the descriptive goals of the current report, themes were based on the surface meaning of explicit participant statements [64].

Results

Results refer to men who engaged in sex work as “MSWs,” and men who did not as “other MSM,” although all participants are men who have sex with men. Descriptive characteristics appear in Table 1. Subgroups were roughly comparable in age and race, but MSWs reported lower incomes, less education, more unemployment, less stable housing, less access to health insurance and primary care providers, more uncertainty about HIV status, and more use of illicit and injection drugs. MSWs were also less likely to identify as gay and more likely to have had sex with women in the past 6 months.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Selected Sample Characteristics.

https://doi.org/10.1371/journal.pone.0112425.t001

In the remainder of these results, illustrative quotes are reported in text with anonymized research ID numbers. Numbers beginning “FG” represent participants from focus groups, and numbers beginning with “INT” represent participants from individual interviews. Because focus groups did not include a dedicated section on healthcare access, emergent findings from focus groups are represented by fewer quotes. Each section first considers common themes arising across subgroups, then unique findings for MSWs and other MSM. Men in both subgroups considered dental care to be a form of medical care, so findings on dental care are included here as well.

Discussion

We examined access to healthcare, HIV/STI testing, unmet healthcare needs, and preferred PrEP providers among MSWs (primarily street-based) and other MSM in Providence, RI. Similarities emerged across subgroups: MSWs and MSM both reported receiving care in ERs, mental health or substance use clinics, and clinics serving low-income clientele; HIV testing was more frequent than STI testing; unmet healthcare needs included acute and long-term medical care. Like other studies among MSM and sex workers [70], [71], we also identified a range of barriers to HIV and STI testing, including fear, concerns about confidentiality, and structural barriers such as cost. Consistent with prior studies, we found that cost may be a central barrier to PrEP uptake [72], [73], given that men in both subgroups reported an inability to access a wide range of other healthcare services due to cost. Unmet healthcare needs across these subgroups included not only HIV prevention services such as HIV and STI testing, but also more basic healthcare needs such as primary care, the management of chronic conditions, laboratory work, prescriptions, substance use treatment, mental health services, and dental care. Importantly, many individuals may perceive other basic healthcare needs as more urgent or important than access to PrEP; few mentioned HIV prevention services when discussing their healthcare needs. We also found that men in both groups had a range of preferred PrEP providers, including PCPs, specialists, and psychiatrists. Our finding that men in both groups preferred PCPs for PrEP implementation differed from a recent study in San Francisco, which found that MSM often preferred to separate sexual health from primary care [33], but early findings on PrEP implementation in San Francisco also report successful PrEP referrals from nonspecialist providers [31].

We also identified differences between MSWs and other MSM in their recent healthcare experiences, which may influence access to PrEP. MSWs were more likely to access care in ERs, substance use treatment programs, and correctional institutions. Routine or required intake procedures in these settings may account for the slightly larger proportion of MSWs reporting recent checkups and HIV tests. In contrast, MSM who did not engage in sex work were more likely to obtain STI testing and non-emergency care in ambulatory care settings. When discussing unmet healthcare needs, MSWs prioritized substance use treatment and mental health services, while other MSM highlighted dental care, eye care, and prescription coverage.

Although the focus of this research has been PrEP implementation, we note that PrEP may not be the optimal HIV prevention strategy for all MSM or MSWs. Men in these groups have many unmet health care needs and priorities, and PrEP may not meet their most pressing healthcare needs. Some MSM or MSWs may not want to use PrEP, some may be unable to use it due to side effects, or consultation with a provider may suggest that PrEP use is not indicated. For these men, access to other types of healthcare services may still be needed, and providing treatment for substance use and mental health disorders may have an independent impact on HIV risk. Men in these populations may prefer other ways of reducing their HIV risk, such as condom use, post-exposure prophylaxis, treatment as prevention, or other strategies; healthcare providers, health insurers, HIV prevention outreach personnel, and the general community should work to support access to a menu of HIV prevention options, as well as the flexibility for individuals to choose among those options.

Within the broader goal of encouraging access to many HIV prevention options, our findings have specific implications for PrEP implementation. We found extensive gaps in insurance coverage among street-based MSWs and MSM. These cost-related impediments to accessing healthcare occurred not only among men without health insurance, but also among insured men who could not afford co-pays or cost-sharing requirements of private health insurance plans. Lack of health insurance has been correlated with increased condomless sex among MSM [74], demonstrating both the need for PrEP and the need for affordable healthcare in this group. For men who wish to use PrEP, implementation solutions are needed to help individuals navigate the process of accessing health insurance, and among insured individuals, to access assistance with co-pay requirements (possibly through drug manufacturers’ medication assistance programs where available). Effective scale-up of PrEP may also require more general advocacy for expanded health insurance options for low-income individuals, including single childless adults. Although many states have now expanded their Medicaid programs to extend health insurance coverage in this group, individuals may need extensive help to gain awareness of these benefits, to apply for coverage, and to learn to use their coverage to access services after their insurance policies take effect.

Our findings also underscore the need for increased cultural competency and provider training in care for all MSM [38], [75] if PrEP implementation is to be successful. Ideally, PrEP should be part of a comprehensive care package for both MSWs and MSM, which would include substance use and mental health treatment, STI screening and treatment, hepatitis C screening, primary care, and long-term management of chronic conditions unrelated to HIV.

Addressing access barriers is critical for providing PrEP and other HIV prevention services to street-based MSWs. Barriers may include lack of access to primary care, lack of insurance coverage for visits, prescriptions, and lab work; and difficulty maintaining provider relationships due to substance use. But this study suggests that PrEP implementation efforts can capitalize on venues where MSWs already see providers. ERs, substance use and mental health clinics, correctional institutions, and HIV testing centers provide opportunities for PrEP education and referrals. MSWs often reported positive experiences in substance use clinics and mental health care, and they preferred to receive PrEP education in these settings. Although PrEP prescription in these settings may be rare and logistically challenging, men may accept referrals from trusted providers to other clinicians who are more familiar with PrEP. PrEP education for street-based MSWs should clarify that PrEP is not addictive; providers should also inform patients that TDF-FTC for PrEP has few interactions with other drugs, and that hepatitis C is not a contraindication to PrEP use.

MSWs in this study reported lapses in care and difficulty visiting healthcare providers due to substance use, relapse, fear of discovering bodily damage due to substance use, and problems with transportation to appointments. Access to substance use treatment may be a higher priority for many street-based MSWs than access to HIV prevention services, including PrEP; although either may be beneficial alone, combining the implementation of PrEP with substance use treatment may have powerful benefits for this population, particularly if substance use treatment facilitates engagement with the care system as a whole. PrEP clinicians might also use strategies known to improve engagement and retention in HIV care, such as reminder calls and texts, contacting emergency contacts after missed appointments, working with community agencies or peer counselors to reengage individuals who have left care, or implementing supportive services such as transportation assistance [76].

For professionals interested in implementing HIV prevention services, including PrEP, among MSM who do not engage in sex work, barriers to uptake and long-term follow-up are lower but not absent. Lack of insurance coverage, lack of prescription drug coverage, and nondisclosure of risk behavior may still complicate PrEP uptake. Many men in the sample expected their PCPs to know about and actively offer them PrEP after a disclosure of same-sex attraction or sexual behavior. PrEP uptake may therefore be enhanced if providers discuss PrEP with patients in their practices who have already disclosed gay/bisexual identity or MSM behavior. This finding complements a recent study of HIV care providers, which found that patients’ requests for PrEP are a primary motivator for providers to prescribe PrEP [77]. If patients and providers are each waiting for the other to begin a conversation about PrEP–or an even more basic level, to begin a conversation about HIV and other STIs–this may create a barrier to access. Education about HIV prevention options and strategies for effective provider-patient communication are needed for both providers and MSM, in order to encourage both parties to initiate discussions about HIV prevention during clinical visits.

This study is among the first qualitative inquiries into healthcare access, HIV testing, and PrEP provider preferences among street-based MSWs and other MSM. Qualitative methods are ideal for identifying heterogeneous and detailed experiences, including subtle differences between subgroups, and our two-stage approach allowed us to probe unsolicited data about healthcare experiences and preferences in a diverse population of MSM and MSWs. We are among the first to explore barriers to PrEP use among street-based MSWs in the US, a marginalized population at increased HIV risk. Our large sample yielded nuanced data and wide-ranging experiences.

Our findings also have limitations. Because we used anonymous interview procedures, we do not know how many individuals enrolled in both stages of this study; we believe this number to be small based on facilitators’ familiarity with participants. Compared to other PrEP acceptability studies among US MSM [16], [19], [78]–[84], our sample was more likely to be low-income or disabled, non-gay-identified, and white and non-Latino, which may limit generalizability of findings. All data were self-reported and may be subject to recall, self-report, and social desirability biases. Our results may not generalize to other groups of MSM or MSWs, and findings from the MSW subgroup may not apply to MSWs who meet partners in other venues besides streets and adult bookstores, such as through the Internet or brothels.

Future research can build on these findings by using population-based and longitudinal methods to quantify healthcare utilization and testing behaviors, subgroup differences, associations between healthcare access and actual PrEP uptake, and changes in health-seeking or healthcare needs over time. Further studies could also examine the prevalence of testing “by proxy” for STIs and HIV (e.g., by donating blood or extrapolating from a partner’s test results). To facilitate optimal implementation of HIV prevention strategies including PrEP, quantitative methods might also examine men’s preferences for different types of providers and self-reported efforts to obtain PrEP and other services. Research should also obtain qualitative and quantitative data on how providers and patients communicate about PrEP after discussing sexual histories. Settings for this work should include ERs, substance use clinics, mental health clinics, correctional institutions, and HIV testing centers. Integrating these venues into PrEP implementation efforts can advance access for MSM, including men who engage in street-based sex work.

Acknowledgments

We are grateful to the study participants, Project Weber, Miriam Community Access, the Yale Center for Interdisciplinary Research on AIDS, the Lifespan/Tufts/Brown Center for AIDS Research, Melissa Guillen, Genevieve Ilg, Bobby Ducharme, and Dr. Caroline Kuo for help during the implementation of this study. We are also grateful to Dr. Douglas Krakower for helpful commentary on an earlier version of this manuscript.