Implementation of HIV pre-exposure prophylaxis for women of color: Perspectives from healthcare providers and staff from three clinical settings

Abstract

Women of color (WOC) account for 83% of new HIV infections among women in the United States. While pre-exposure prophylaxis (PrEP) is a safe, effective HIV prevention method for women, WOC are less likely to be prescribed PrEP than other populations. Guided by an implementation science research framework, we investigated the implementation of a PrEP initiative for WOC in a US city with high HIV incidence. Across three clinical sites, only three WOC were prescribed PrEP after one year. Analysis of qualitative interviews with clinic staff and providers identified time constraints, reluctance to prescribe PrEP, and discomfort with counseling as implementation barriers. Implementation facilitators included staff and leadership support for PrEP, alignment of PrEP services with organizational missions, and having a centralized PrEP Coordinator. By addressing these identified implementation barriers and facilitators, clinic staff and providers can ensure that WOC are provided with the full range of HIV prevention options.

Introduction

Women account for 19% of new HIV diagnoses in the United States. Women of color (WOC) are disproportionally affected by HIV, making up 83% of all new HIV cases in women (Centers for Disease Control & Prevention, 2019). In 2012, the US Food and Drug Administration approved a daily, oral tablet as pre-exposure prophylaxis (PrEP) to prevent HIV transmission (Food & Drug Administration, 2019).Previous research has elicited WOC’s perspectives of PrEP. WOC have voiced frustration about being unaware of PrEP (Auerbach et al., 2015; Fletcher et al., 2018; Goparaju et al., 2015) and have expressed concerns about potential side effects (Auerbach et al., 2015; Flash et al., 2014; Fletcher et al., 2018; Kwakwa et al., 2016), cost (Auerbach et al., 2015; Bond & Gunn, 2016; Flash et al., 2014; Goparaju et al., 2015), the daily pill burden (Bond & Gunn, 2016; Collier et al., 2017; Fletcher et al., 2018), and discussing HIV risk with providers (Bond & Gunn, 2016; Fletcher et al., 2018). WOC have also expressed mistrust of medical systems, providers, and pharmaceutical companies (Fletcher et al., 2018; Tekeste et al., 2019), while also identifying stigma surrounding PrEP and HIV (Auerbach et al., 2015; Fletcher et al., 2018). Despite these barriers to PrEP use, WOC have perceived important benefits of PrEP, including the potential for it to be a prevention method that they can control. WOC have also PrEP as valuable in providing protection within serodiscordant partnerships and being an “empowering” prevention option (Auerbach et al., 2015; Bond & Gunn, 2016; Collier et al., 2017; Fletcher et al., 2018; Sales & Sheth, 2019). Further, some studies have shown that WOC are more interested in and have greater intention to use PrEP than White women (Garfinkel et al., 2017; Tekeste et al., 2019).

Previous research examining healthcare providers perspectives of PrEP implementation has identified that providers have a variety of beliefs about who is best positioned to prescribe PrEP. Both HIV and non-HIV providers voice concern about the feasibility of providing PrEP in primary care settings due to time constraints and lack of specialized training (Hoffman et al., 2016; Krakower et al., 2014). Yet, some providers argue that PrEP should be provided in a primary care setting as HIV-negative individuals are more likely to access care at primary care centers (Bacon et al., 2017; Hoffman et al., 2016; Karris et al., 2014; Krakower et al., 2014). Krakower et al. (2014) describes this problem as the “purview paradox”: neither HIV specialists nor primary care providers (PCPs) believe PrEP prescribing to be within their purview (Krakower et al., 2014). Both HIV and non-HIV providers identified the lack of training as a barrier to PrEP implementation, requesting more training in adherence counseling, screening and selecting patients for PrEP, managing side effects, risk reduction counseling, and advising on PrEP and conception (Bacon et al., 2017; Finocchario-Kessler et al., 2016; Petroll et al., 2017; Smith et al., 2016; Terndrup et al., 2019). Providers have also expressed clinical concerns about prescribing PrEP such as the possibility of a patient developing drug resistance (Bacon et al., 2017; Blumenthal et al., 2015; Karris et al., 2014; Mullins et al., 2017; Psomas et al., 2019; Tripathi et al., 2012), and possible drug toxicities (Blumenthal et al., 2015; Karris et al., 2014; Krakower et al., 2014). Providers have stated the need for long-term monitoring of patients as a potential barrier to PrEP implementation, affecting their willingness to prescribe PrEP (Adams & Balderson, 2016; Blumenthal et al., 2015; Hoffman et al., 2016; Mullins et al., 2016). However, these studies did not examine healthcare providers’ perspectives of PrEP implementation specifically for WOC. National, patient-level data from US pharmacies reveals that providers have prescribed PrEP to significantly fewer African American and Hispanic women compared to White women, highlighting a gap in HIV prevention for WOC (Bush et al., 2016).

Background

Beginning in January 2017, three clinical sites in a major US city participated in a one-year initiative to increase PrEP uptake among WOC. This city has one of the top 15 highest HIV incidence rates in the United States (Centers for Disease Control & Prevention, 2018). Medicaid expansion under the Affordable Care Act was adopted and implemented by the city in 2014 (Advisory Board, 2020). About one-third of residents are low-income, and similarly, about one-third of residents receive health insurance through Medicaid (Kaiser Family Foundation, 2020). The city offers a drug assistance program for PrEP for both insured and uninsured residents. At the request of one of the participating sites, the site names and city of this initiative are not disclosed. Though this initiative focused on WOC, White women were not excluded from receiving PrEP services.

The clinical sites were a Title X-funded family planning clinic (site A), a 501(c)3 nonprofit pediatric emergency department (ED; site B), and a Federally-Qualified Health Center (FQHC) with primary care clinics (site C). Sites A and C focused on PrEP provision to adult WOC and site B focused on young WOC aged 13–24 years. Providers, clinical support staff, and health promotion staff at each site received training in providing PrEP information, counseling, and prescribing (providers only). At each site, WOC were provided information about PrEP through outreach performed by health education staff in the community or at the site (such as approaching patients in waiting rooms), or in conversation with site staff during medical appointments. Each site had established protocols for PrEP initiation, including, at a minimum, screening for PrEP eligibility and, if eligible, referral to a provider.

The goal of the initiative was to prescribe PrEP to over 350 WOC. After one year of implementation, the initiative fell far short of this goal, with only three WOC being prescribed PrEP over the course of the year. The purpose of this study was to systematically investigate clinical staff and providers’ perspectives on why this PrEP initiative failed to substantially increase PrEP prescribing to WOC and develop recommendations to inform service improvement and future PrEP implementation efforts.

Methods

Overview of design

The integrated-Promoting Action on Research Implementation in Health Services (i-PARIHS) framework guided the design and methods for this study (Harvey & Kitson, 2016). The i-PARIHS framework examines four core elements of implementation: innovation, recipients, context, and facilitation. Innovation refers to what is being implemented, in this case PrEP. Recipients are who is involved in the implementation (e.g., clinic staff, providers), and considers staff values, time, capacity, skills, and attitudes toward the innovation. Context is where the implementation occurs and includes the immediate environment where the innovation is implemented and the wider context (e.g., the health system). Facilitation is how implementation happens. Informed by the i-PARIHS framework, we developed two semi-structured interview guides (one guide for baseline interviews and one guide for follow-up interviews) that explored who was involved in implementation, perceived implementation barriers and facilitators, how staff and providers interacted with patients about PrEP, and when and how patients would come into contact with PrEP counseling (see Table 1 for questions within each i-PARIHS element). We selected the strategy of two rounds of interviews in order to examine implementation throughout the program lifespan, from the early stages through when the program was more developed.

Table 1.

I-PARIHS constructs with corresponding interview questions.

Innovation	• What have been the main challenges to implementing this initiative?
	• What has helped you implement this initiative?
	• What tools or materials do you use to implement the initiative?
	• How well does the PrEP for Women Initiative fit within your clinic?
Recipients	• How much of a priority is this initiative to you?
	• What are the necessary skills and training needed for your position to carry out this initiative?
	• What motivates you to implement this initiative?
Context	• How are staff involved in decisions that affect them?
	• What mechanisms are in place to support learning and evaluation for this initiative?
	• How does the wider health system, such as local or national priorities and policies, influence your implementation of initiative?
Facilitation	• How was the initiative developed? Who was involved in the launch of the initiative? How is training conducted and what is included?
	• What are the main components or activities of the initiative?
	• Who (what staff) is involved in the implementation? What are their corresponding responsibilities?

Data collection

Baseline interviews were conducted between August 2017 and January 2018, beginning approximately eight months after the start of the implementation. Follow-up interviews were conducted approximately eight months after baseline interviews, between May and August 2018. A point of contact at each clinic provided a list of potential participants who were contacted via phone or e-mail to invite them to participate in the study. Follow-up interviews included available baseline participants as well as new staff. We stopped conducting interviews when we determined that saturation of themes had been reached (i.e., when we repeatedly heard similar things across interviews and determined that additional interviewing would not yield significant new insights into key topics of interest). We conducted 18 baseline interviews and 15 follow-up interviews (total n=24 unique participants). Interviews were conducted in-person or over the phone and lasted an average of 34 min (range: 25–68 min). Interviews were audio-recorded and transcribed. The Boston University Medical Center Institutional Review Board approved all study procedures.

Data analysis

The Framework Method guided the analysis through the following five steps: familiarization with the interview, codebook development, codebook application, data charting, and data interpretation (Ritchie & Lewis, 2014).This method, widely used in health research, allows data to be analyzed deductively, through pre-selected codes (e.g., implementation barriers and facilitators) and inductively, by exploring unexpected themes that may arise and developing additional codes (Gale et al., 2013). Two analysts reviewed the first four transcripts, drafted a codebook, and separately applied the codebook to the remaining transcripts. We used NVivo 11 qualitative data analysis software to organize and code the data. The analysts met weekly to review the coding process and ensure consistency and discuss and develop any new codes as needed, which were then added to the codebook and applied to previously coded transcripts. The codebook was not considered finalized until the last transcript was coded. The analysts explored the data through weekly discussion and the completion of analytic memos throughout the analysis process. The analysis led to the identification of key implementation barriers and facilitators.

Results

Among the 24 participants, 87.5% (n=21) were female. Nearly half (45.8%; n=11) were providers, with the remainder being clinical support services staff (29.2%; n=7) and health promotion staff (25.0%; n=6; see Table 2). Clinical support staff include nurses, nurse technicians, and clinic managers. Health promotion staff include staff whose primary responsibilities involve health education and outreach.

Table 2.

Characteristics of study participants by clinic (n = 24).

	Clinic A (n = 9)	Clinic B (n = 7)	Clinic C (n = 8)	Total
Gender	n	n	n	n (%)
Female	9	6	6	21 (87.5)
Male	0	1	2	3 (12.5)
Years worked at clinic
<1	3	1	2	6 (25.0)
≥1 and <3	5	0	4	9 (37.5)
≥3	1	5	2	8 (33.0)
Participant type
Clinical Support Services Staff	2	2	3	7 (29.2)
Health Promotion Staff	4	2	–	6 (25.0)
Providers	3	3	5	11(45.8)

Overview of qualitative themes

The semi-structured interviews provided an in-depth examination of participants’ perspectives of PrEP implementation for WOC. Across the interviews, eight key implementation barriers and six key implementation facilitators emerged. Implementation barriers included: 1) differences in understanding of the PrEP protocol; 2) reluctance to prescribe PrEP; 3) discomfort providing PrEP counseling; 4) concerns about protecting the confidentiality of minor-aged patients; 5) challenge of integrating PrEP into routine clinical practice due to time limitations; 6) perceived patient-level barriers to initiating PrEP that were not addressed in the implementation of the initiative; 7) managing different and changing priorities across departments; and 8) navigating the complex insurance landscape. Despite these significant implementation barriers, participants cited several implementation facilitators, including: 1) PrEP as congruent with their respective organizations’ goals and activities; 2) a centralized PrEP Coordinator supported implementation; 3) staff and leadership support for PrEP provided an enabling environment for implementation; 4) clinics’ relationships with local communities and other sites as helpful for implementation; 5) building on their existing point-of-care testing procedures aided PrEP implementation; and 6) the ED was a point of healthcare access for high-risk YWOC. These implementation barriers and facilitators are described below with illustrative quotes from different types of participants (P for provider, S for staff) at each clinical site (A, B, C), and detailed in Table 3.

Table 3.

Implementation barriers and facilitators, organized by the i-PARIHS constructs of innovation, recipients, and context.

	Barriers	Clinic A	Clinic B	Clinic C	Facilitators	Clinic A	Clinic B	Clinic C
Innovation (degree of fit, compatibility, diffusion)	Differences in understanding of the PrEP protocol	■		■	Program fit within the organization’s goals and activities	■		■
	Reluctance to prescribe PrEP		■		Having a centralized PrEP Coordinator		■	■
Recipients (capacity, attitudes, beliefs, skills)	Staff and provider time limitations*	■	■	■	Staff and leadership support for PrEP services	■	■	■
	Staff discomfort providing PrEP counseling	■
	Staff and provider perception of patient-level barriers that need to be addressed in implementation	■	■	■
	Concern about protecting patients’ confidentiality		■
Context (organizational priorities, relationships)	Managing different and changing priorities across departments	■			Relationships with the local community	■		■
	Navigating the complex insurance landscape			■	Building from the point-of-care-testing program		■
					The ED as an access point for health care for YWOC		■

■Indicates theme was identified by clinic.

^*For clinic C, time limitation was identified as a barrier for providers only.

Implementation barriers

Though each site had their own established protocols for PrEP screening, participants at sites A and C described differences in understanding of their site-specific PrEP protocols. The protocol included both how staff were to screen patients for PrEP and what criteria staff used to assess PrEP eligibility. At site A, some participants reported that they used a sexual health history template in the Electronic Medical Record (EMR) to determine eligibility, while others appeared unaware that a template existed. A staff member stated, “We were never given an exact dialogue about how to conduct that conversation …. I just outline information that was given to me and answer any questions that [patients] may have” (S-A). Though the protocol stated that both clinical support staff and providers were supposed to conduct PrEP screening, in practice, this protocol was not always followed: “Put it like this, the [clinical support staff] weren’t really pushing it, the clinicians were more so” (S-A). Additionally, participants at site A reported wide variation in PrEP eligibility criteria. Some participants listed discrete criteria such as, “Women who either have multiple partners, or know that their partners are having multiple partners, or have had a recent bacterial STI” (P-A). However, others had broader views of PrEP eligibility criteria. For example, one staff member stated that she discussed PrEP with any patient “that just seem[ed] concerned about becoming HIV positive” (S-A).

At site C, despite the institution of a two-step PrEP screening protocol in which providers screened patients for PrEP eligibility using a sexual health history screener and then referred patients to the PrEP Coordinator or other clinical support staff, providers’ approaches to the PrEP protocol differed. Some providers followed the protocol closely and initiated sexual health conversations with their patients in order to assess risk factors for HIV: “It is primarily conversations during the taking of the sexual health history and social history questionnaire, and explaining what the benefits of PrEP and also the commitment that is PrEP and seeing if that is something they are interested in” (P-C). Other providers did not probe about risk factors but rather waited for patients to identify their own risk factors. Despite being trained in PrEP screening, one provider was unaware that a standardized protocol was in place, stating, “We just need to protocolize when we have that conversation to ensure that we are doing it for everybody. Because right now, “I don’t routinely screen every single patient I see, I use identifiable risk factors that come to my attention …. I think that I’m missing a significant portion of people who are eligible” (P-C).

Participants from the ED-based site expressed reluctance to prescribe PrEP as they were concerned that patients would not receive appropriate long-term PrEP care: “Because prescribing [PrEP] is not like prescribing Plan B, right, prescribing Plan B is like one medication order and we’re done. [PrEP] is a whole separate situation …. There needs to be some follow up on the ED end as well” (P-B). Another participant agreed, voicing concern that patients would not follow up with their primary care provider (PCP): “If we are starting them on the PrEP, then making sure that they have adequate follow up could be a barrier. Because we can start it but we can only give you but so much before you need to go to your doctor” (S-B). Some providers believed that long-term medications like PrEP should only be prescribed by a PCP, with one provider stating, “I’m not going to start prescribing a medication to treat you for long term, that’s something that a primary care doctor needs to follow you for and treat you for” (P-B).

Staff from the family planning clinic described discomfort providing PrEP counseling, feeling under-equipped to provide such counseling in a considerate, non-judgmental manner: “It’s really hard to talk to people about PrEP without making them feel uncomfortable, or disrespecting them in some way by suggesting that they are promiscuous or what have you” (S-A). One staff member reported that as a result of this discomfort, “We haven’t done a real push [for PrEP] because it’s been kind of sensitive” (S-A). Another staff member suggested that staff training should focus on what language to use in PrEP counseling: “I think that if we had a little bit more training on how to conduct that dialogue, or stay away from words that concern or feed into the stigma of HIV, or I guess more training on providing comfort on usage of PrEP” (S-A).

Participants from the ED-based site also expressed concerns about protecting the confidentiality of patients under the age of 18 while considering PrEP. This issue intersected with logistical barriers to maintaining confidentiality: “There’s also issues of adolescent confidentiality and if the patient is in a random room, then you have to take them down the hall. And if you can’t take them down the hall and send their parent away, it’s a whole thing” (P-B). Ensuring confidentiality in the ED was described as more challenging than in the primary care setting: “It’s one thing when you have yourself together enough to go to a PCP appointment and the PCP can prepare the parent …. But when your 16-year-old goes to the ED for a sprained ankle, you do not see any of that coming” (P-B). Confidentiality concerns extended beyond the ED visit, including concerns about parents finding out that their child was prescribed PrEP from an insurance bill: “If this is a 16-year-old kid and they are here without mom, but then mom finds out that they’re on PrEP … how do we navigate it?” (S-B).

Participants across clinics the challenge of integrating PrEP into routine clinical practice due to time limitations. Staffing shortages limited the time providers had to focus on the initiative: “We’ve been short-staffed, and so my time has been pulled in a lot of different directions and away from this project” (P-A). The lack of time made it challenging for providers to include PrEP counseling during patient visits: “So we have to consider all of these things and how we’re going to use that 15-min visit for a physical, which is insane, you know what I mean? … I think that realistically there is no way that all of the providers are going to be focusing, are going to be offering PrEP to all those women and men who are eligible” (P-C). Providers at the ED described needing to use their limited time to address patients’ chief complaints: “If they come in for an ankle sprain, because I only have a certain amount of time to spend in the room with them, I won’t bring up that issue. I just don’t have time to kind of have a worthwhile discussion” (P-B).

Participants also perceived patient-level barriers to initiating PrEP that were not addressed in the implementation of the initiative. Participants perceived that patients had limited knowledge of HIV transmission, which made it more challenging to discuss PrEP. Participants had not expected that they would first need to provide basic education on how HIV is transmitted prior to informing women about how PrEP can be used as a prevention option: “Oftentimes I have to remind patients that the same way you become infected with chlamydia or gonorrhea [is how you become infected with HIV], and I think that that is a shock factor for a number of patients in terms of, what do you mean? What do you mean I can actually get HIV this way?” (P-C). Participants reported that some patients appeared worried that their insurance would not cover PrEP: “[I get] a lot of financial questions, if it’s going to get covered by insurance, if not, how will I be able to pay for it” (S-A). HIV-related stigma was also perceived as preventing patients from engaging in conversations about PrEP: “I feel like [patients] are scared when they hear that word [HIV]; they don’t even really want to take any steps or initiative to protecting their health” (S-A). A provider described a patient being unwilling to accept a PrEP prescription out of fear of being assigned the diagnostic code for “high risk sexual behavior”: “And that is when it turned around from, ‘Yeah I’m going to do this,’ to ‘No, I’m not that person. I’m not a high-risk person’” (P-C). Staff also believed that female patients had low perceived HIV risk that held them back from engaging in conversations about HIV prevention: “I don’t think women see [HIV] as an issue for them to concern themselves with” (S-C). For young WOC, participants were also concerned that a daily pill for HIV prevention would not be prioritized other, more urgent issues in their lives: “Because it’s like, you’re calling to tell me about a pill? And that’s the least of my worries right now. I need to worry about how I’m going to eat, how I’m going to make it to school ….” (S-B). Though participants described that some young WOC visited the ED alone, the frequency of parents/guardians accompanying children could present a challenge: “Either [the parents/guardians] are not engaged in care, they don’t want me in the room, or they’re like, I’m being insensitive …. So the guardians have actually been my hardest group to tackle throughout this whole thing” (S-B).

At the family planning clinic, participants reported that managing different and changing priorities across departments created important barriers to implementation of the PrEP initiative. Collaboration across departments could be difficult because of competing departmental priorities: “We’re on different priority and timelines, which is inevitable, and it doesn’t always align” (S-A). The burden of implementation was sometimes placed more heavily on one department over another: “It has been a difficult dance between [departments] … it is not always an equal balance” (S-A). Other issues needed to be prioritized over the initiative at times: “There is advocacy work that needs to be done. That is an urgent, in the moment need, so my attention and my investment needs to shift” (S-A).

Participants from the FQHC reported that navigating the complex insurance landscape was challenging: “Because of the health care landscape that I practice in, [PrEP] is something that requires prior authorization often times, so there is a little bit of a delay” (P-C). Enrollment in drug payment assistance programs often required an extra clinic visit to complete paperwork, and coverage could be temporary: “We just learned today about a patient who when they were approved, they were only approved for 180 days and they have to reapply” (S-C). Though US citizenship is not a requirement for these programs, participants were concerned about the citizenship status of undocumented patients being revealed: “What do we do with someone that is undocumented and does not have insurance and wants to access it. What do we do about that?” (S-C).

Implementation facilitators

Participants viewed PrEP as congruent with their respective organizations’ goals and activities related to providing comprehensive health services to women. A staff member expressed that offering PrEP was, “directly aligned with [our clinic]. We have a mission to serve women … with comprehensive reproductive and sexual health information and services” (S-A). Participants viewed PrEP as an HIV prevention option that was a natural extension of their services: “I think in particular because of the nature of testing for STIs, and because we’re so focused on reproductive health, I think that we have the opportunity to capture women who are at risk [for HIV]” (P-A) and filling a gap as a service “that definitely was missing” (S-C).

Participants reported that having a centralized PrEP Coordinator supported implementation. The existence of a PrEP Coordinator helped address participants’ perceived burden of dealing with a new innovation: “Having the [PrEP Coordinator] here, and here to spearhead for that initial phase, I think it helps because it keeps the nurses from having to do so much” (S-B). The PrEP Coordinator was, “ … the information central so that we know where to go instead of just wondering who should we contact” (S-C). Providers sought the PrEP Coordinator’s advice when considering PrEP for particular patients: “I would just call him and say, hey this is a specific situation, what do you think, should they follow up sooner, things like that. It’s been good to have a role like that” (P-C).

Participants stated that staff and leadership support for PrEP provided an enabling environment for implementation. Leadership paved the way for the clinic to incorporate PrEP provision: “I think the first factor is the guidance and leadership of senior management around PrEP is really important” (S-A). Leaders’ endorsement of the initiative was described as, “ … very positive, very supportive about having programs like this available” (P-B). Leaders supported staff training by “facilitating the education component and making it a requirement for the clinical team” (P-C). Participants also described leaders as taking steps to support the initiative: “I believe the organization is investing in it, providing all the tools, all the information that we need, how to provide care for the patient and support for the providers” (S-C).

Similarly, participants also voiced their own support for PrEP provision, describing it as a critical HIV prevention tool for women: “I don’t think we’ve seen anything that can be so impactful as PrEP in a long time for women and for communities affected by HIV” (S-A). A provider described that he was motivated to discuss PrEP because he believed it would positively affect his patients’ health: “ … if we take five minutes to talk about PrEP, we have a very measurable and profound impact on risk reduction” (P-C). Participants expressed excitement toward providing another HIV prevention option for WOC: “I think that being able to expand the PrEP services, which have been primarily given to men who have sex with men [MSM], to women and specifically young women, I think it’s a really powerful tool” (S-B). Another provider commented that, overall, staff and providers were invested in the initiative: “Providers are engaged, and the clinic as a whole is engaged to learn and to offer PrEP. I think that’s the biggest piece, that we have a supportive environment that is willing to learn and engage clients with regards to the topic” (P-C).

Participants described how their clinics’ relationships with local communities and other sites offered opportunities as helpful for implementation. First, clinics leveraged their relationships with community members to conduct outreach: “A lot of the health educators are very well connected in the community, and people come up to them with events” (S-A). The FQHC utilized nursing students from a local university as outreach volunteers: “They become like ambassadors if you will, where they can go and now talk to other young women their age about PrEP and their options” (S-C). Second, relationships between initiative clinics were also perceived as helpful. In addition to partnering with other sites on outreach events, sites had monthly calls, which one staff member described as, “ … opportunities to share expertise with a team who are engaged and receptive to ideas” (S-C). Another staff member reported, “I relied on our partners … having strong partners has been helpful” (S-A).

At the ED, building on their existing point-of-care testing procedures aided PrEP implementation as it created processes such as alerts in the Electronic Medical Record (EMR) that notified staff of patients seeking STI testing, and therefore a potential candidate for PrEP:

“If any patient identifies with any STI concerns … that will go on the tracking board which would then be alerting me to go in and speak directly with that patient” (S-B). The testing program also created an alert that notified staff of a patient’s positive STI screening: “We launched a PowerForm for our HIV and STI screening, so now it’s built into our [EMR] … which is helping us identify some of the young women who are at risk” (S-B). Additionally, participants received monthly progress reports on the testing program (e.g., number of patients who received STI treatment), which demonstrated their role in reaching young women at risk for HIV: “Because of this screening program that we are doing, [the patient] was identified as having a STI and treated. [We] really get to see the benefit of what [we] are doing” (S-B).

ED-based participants also believed that offering PrEP made logical sense because the ED was a point of healthcare access for high-risk YWOC. Participants perceived that some young WOC utilized the ED more frequently than primary care: “I get the sense that a lot of the health care needs that they have get addressed through the ED as opposed to through their primary care doctor. So the benefit of running PrEP through the ED is that you are reaching the target population where they are interfacing with the health care community” (P-B). Participants reported that some young WOC had established relationships with ED staff: “We have lots of patients who come in and they have regular HIV and STI screens, more than once a year, and kind of have a good rapport with some of the staff” (S-B). A provider explained that the young women who came into the ED were the target population (at-risk WOC) for the initiative: “But I think [PrEP] will be only beneficial because we see a fairly high-risk population here …. There are plenty of women that are high risk and I think they can only benefit from [PrEP]” (P-B).

Discussion

Despite meeting its first two goals regarding numbers of staff trained and WOC reached with PrEP information, the PrEP initiative was not successful in reaching the outcome of prescribing PrEP for WOC at-risk, with only three WOC prescribed PrEP. Although clinical staff, providers, and site leadership were generally supportive of PrEP and viewed it as a “good fit” consistent with the missions of their organizations, significant challenges existed. Our investigation of staff and providers’ perspectives on the initiative, guided by the i-PARIHS implementation science framework, has provided important insights into PrEP delivery to WOC.

We identified important challenges with PrEP as a new innovation, including individual provider differences in understanding PrEP screening criteria and processes. Research on female PrEP patients has found that most were identified as PrEP candidates because they had a known HIV-infected sexual partner, suggesting that providers may not explore women’s risks past this single criterion (Blackstock et al., 2017; Seidman et al., 2016). Indeed, a recent study demonstrated that the use of the US Public Health Service’s 2017 screening guidelines overlooked women with other risk factors for HIV (e.g., a STI in the last six months) as well as those who were already motivated to take PrEP (Calabrese et al., 2019). Moving beyond screening for discrete eligibility criteria toward routinizing this innovation, that is, introducing PrEP to all women, could help improve PrEP access for WOC by positioning PrEP as part of regular care (Aaron et al., 2018; Auerbach et al., 2015; Calabrese et al., 2018; Collier et al., 2017; Edelstein et al., 2014; Flash et al., 2014; Goparaju et al., 2015; Seidman & Weber, 2016).

Though staff and providers were supportive of adding PrEP as a service and described having a PrEP Coordinator as helpful, these factors did not result in prescribing PrEP, aligning with findings from previous studies examining providers attitudes and willingness to prescribe PrEP (Blumenthal et al., 2015; Krakower et al., 2015; Smith et al., 2016). Effective PrEP Coordinators should be able to explain and demonstrate tasks (e.g., PrEP counseling) to the provider and staff recipients, as well as address challenges and adapt implementation as needed and specifically to the context (Harvey & Kitson, 2015; Kitson & Harvey, 2016; Laycock et al., 2018). For PrEP programs spanning multiple departments within a clinical site (i.e., Health Education and Clinical Care Departments), a designated coordinator from each department would be helpful to harmonize efforts. Appointing a coordinator to facilitate implementation has been useful with similar innovations (e.g., family planning, HIV testing) (Bokhour et al., 2015; Kitson & Harvey, 2016; Seidman et al., 2018). In order to increase staff comfort in PrEP counseling, brief, one-on-one training sessions have resulted in significant improvements in providers’ PrEP knowledge, screening, and prescribing (Edelstein et al., 2015). The PrEP Coordinator could use these trainings to address staff and providers’ specific concerns and increase delivery of PrEP.

Participants described PrEP as a potentially powerful HIV prevention tool for women but perceived significant patient-level barriers to initiating PrEP, including stigma and fear, factors that increase WOC’s vulnerability to HIV (Aaron et al., 2018; Calabrese et al., 2018; Sheth et al., 2016; The Henry, 2019). Research on delivering PrEP to women encourages shared decision-making approaches to explore these vulnerabilities (Calabrese et al., 2017; Seidman et al., 2018). Shared decision-making is a collaborative process in which providers offer multiple, evidence-based options, patients explain their experiences and desires, and together providers and patients create plans (Aaron et al., 2018; Makoul & Clayman, 2006; Seidman et al., 2018). The shared decision-making approach allows for the exploration of behavior, interpersonal, community, and structural-level barriers to care that are specific to the experiences of WOC (Aaron et al., 2018; HIVE, 2019; Seidman et al., 2018). This approach could uncover factors that would not be identified through traditional risk assessment checklists, while helping to clarify or correct staff perceptions of potential barriers for patients (Aaron et al., 2018; HIVE, 2019; Seidman et al., 2018; Seidman & Weber, 2016). WOC engaged in shared decision-making for selecting contraception have reported higher satisfaction with their care than those who used either a provider-driven or patient-driven approach (Dehlendorf et al., 2017).

As WOC’s preferences may differ from those of other populations (e.g., MSM), more research examining how to tailor PrEP care to WOC is needed. Some research is already underway. For example, a study at the University of Rochester Medical Center is examining multilevel determinants of PrEP care continuum outcomes for WOC (University of Rochester Medical Center, 2019). Preliminary findings from an open pilot trial of Black women in Miami found that a culturally tailored intervention using motivational interviewing techniques increased PrEP uptake, knowledge, and motivations to use PrEP (Dale, 2020). Recently reported evidence demonstrates that Black cisgender women in Chicago preferred to receive PrEP information from their regular primary care provider or other healthcare provider (Hirschhorn et al., 2020). Understanding how WOC prefer to receive PrEP information (e.g., from providers, trained lay persons, peer educators, or other sources, such as patient decision support tools) will better serve this population. Separate research should focus on young WOC preferences for PrEP services, as these may differ from adult WOC. Findings from this study suggest that young WOC may have unique barriers to PrEP uptake, such as resistance from parents/guardians. Additionally, providing PrEP services to young WOC under the age of 18 requires thoughtful consideration of how to protect their confidentiality, both within clinical settings (e.g., PrEP counseling) and beyond (e.g., insurance billing practices).

Limitations

Findings presented here represent a range of perspectives (e.g., health promotion staff, clinical support staff, and providers) in three different care settings that provide useful insights into the implementation of PrEP for WOC, filling a gap in the current literature. However, there are several limitations that warrant considerations. First, case studies and qualitative data are not necessarily generalizable to numerous other settings but rather provide an in-depth exploration of topics within specific contexts (Yin, 1999). Second, retention of participants in this study was challenging. Ideally, those who participated in baseline interviews would have also participated in follow-up interviews to provide insight into how their views and experiences changed throughout the initiative. We were flexible with participants by conducting interviews in-person or over the phone, and conducting interviews outside of typical business hours. However, due to staff turnover across clinics and their limited availability, follow-up interviews were not possible with all participants. Finally, the research team’s own experiences, social identities, and knowledge may have introduced bias into the research. To reduce bias, we followed a formal process for data collection and analysis informed by a theoretical framework. We practiced reflexivity by recognizing how our own experiences influenced the research process (Baker, 2011; Kuper et al., 2008), by 1) presenting the study aims, research plans, and dissemination plan to participants for feedback; 2) developing neutrally-worded interview guides; 3) completing analytic memos that provided space to reflect about the research process; and 4) involving two research team members in data analysis.

Conclusion

Little research has studied real world efforts to deliver PrEP to WOC. Though the clinical sites included in our study seemed well-positioned to provide PrEP to WOC, significant implementation barriers impeded their ability to actually prescribe PrEP. Opportunities exist to address staff- and provider-identified barriers, including hiring a PrEP Coordinator to lead implementation, developing standardized screening and eligibility criteria, providing ongoing feedback and targeted training to staff regarding implementing a standardized PrEP protocol, and incorporating a shared decision-making approach in PrEP counseling to ensure that WOC receive care sensitive to their needs and experiences.