Integrating Behavioral HIV Interventions into Biomedical Prevention Trials with Youth: Lessons from Chicago’s Project PrEPare

Abstract

On the heels of several trials demonstrating the efficacy of pre-exposure prophylaxis (PrEP) and the recent approval by the FDA of the supplemental indication for Truvada as PrEP, researchers, advocates, and community providers are calling for the investigation of implementation strategies that combine behavioral interventions with biomedical prevention. This paper describes the modification and integration of an evidence-based group-level intervention into a small PrEP pilot trial with young men who have sex with men (YMSM). The behavioral intervention as well as ongoing risk reduction counseling sessions were found to be highly acceptable among a sample of racially diverse YMSM.

BACKGROUND

HIV/AIDS is a public health crisis among young men who have sex with men (YMSM) in the United States, and especially among YMSM of color. Young black MSM experienced the largest increase in numbers of new diagnoses of HIV infection among all racial/ethnic groups—2,925 diagnoses in 2007 to 4,358 diagnoses in 2010 (CDC, 2012). Of all MSM aged 13 – 24 years diagnosed with HIV infection in 2009, an estimated 58% were black/African American, followed by Hispanics/Latinos (20%) and white (19%) (CDC, 2012). HIV prevalence among Black youth is nearly three times higher than among Latino youth and nearly eight times higher than white youth (CDC, 2012) Effective interventions to prevent HIV among YMSM are urgently needed.

One promising biomedical intervention recently approved by the U.S. Food and Drug Administration (FDA, July 2012) for primary HIV prevention is pre-exposure prophylaxis, or PrEP. In addition to reassuring safety data from smaller PrEP trials (CDC, 2011), several large efficacy trials have now shown that men and women at risk for HIV infection who took a daily tablet containing emtricitabine and tenofovir (FTC/TDF - Truvada®) experienced fewer HIV infections than those who received a placebo pill (Grant et al., 2010; Baeten et al., 2012; Thigpen et al., 2012). Thus, PrEP is a promising biomedical intervention that can be effective in preventing the acquisition of HIV.

For youth, “real world” effectiveness will depend on adherence to the PrEP regimen and the minimization of behavioral disinhibition. Adherence has been an issue in PrEP trials to date, with estimates ranging from 51% in the iPrEx trial to 82% in the Partners PrEP trial (Grant et al, 2010; Baeten et al., 2012). Adherence to medication as well as oral contraceptives is known to be a significant issue among adolescents and young adults (Hall et al., 2010; Lindsey et al., 2009; Rudy et al., 2009; Zindani et al., 2006). Another concern is that PrEP use may be associated with increased sexual risk behavior (Golub et al., 2010), especially among younger populations, though there has been no such evidence in PrEP clinical trials to date (Grant et al., 2010; Baeten et al., 2012; Thigpen et al., 2012; Van Damme et al., 2012). These issues highlight the need for behavioral strategies targeting adherence and risk reduction if PrEP is to be effective.

Behavioral interventions for young people have been proven to reduce HIV/STI risk by increasing condom use, reducing or delaying frequencies of sex, and increasing safer sex negotiation skills (Johnson, Scott-Sheldon, Huedo-Medina, & Carey, 2011). Several interventions have been certified by the CDC through their Dissemination of Effective Behavioral Interventions (DEBI) Program as having demonstrated ability to be scaled up for use in multiple communities. For youth, such evidence- based behavioral interventions are a critical tool in primary HIV prevention. The CDC’s Compendium of Evidence-Based HIV Behavioral Interventions is inclusive of 8 good-evidence antiretroviral medication adherence interventions that could be adapted for PrEP use (CDC, 2012).

Thus, combining behavioral and biomedical approaches may be critical to successful implementation of PrEP among youth populations. The use of combined behavioral/biomedical strategies to prevent HIV might also be the most developmentally-appropriate way to implement PrEP among youth. For adolescent and young adult MSM, the use of PrEP may be most useful in preventing HIV infection during the high risk years of youth. A growing body of neurobiological research and imaging studies suggest that adolescents may be especially prone to engage in risky behaviors, including sexual risk and substance abuse, due to patterns of adolescent brain development (Galvan et al., 2006; 2007). Thus, PrEP for young people may be best viewed as a time-limited strategy that can bridge the developmental period between sexual debut and adulthood.

This paper aims to describe the integration of a readily available behavioral intervention into a biomedical prevention trial and to outline the adaptation process that occurred in order to embed the evidence-based intervention, Many Men, Many Voices (3MV; Wilton et al., 2009), into a PrEP pilot trial. The results from the PrEP randomized pilot trial have been published elsewhere (Hosek, Siberry, Bell et al., in press) and will not be discussed in this paper. Rather, this paper will focus on the process evaluation of 3MV including qualitative interviews with study participants.

METHODS

Study Overview

Project PrEPare, funded through the Adolescent Medicine Trials Network for HIV/AIDS Interventions (ATN), was designed to examine the feasibility and acceptability of a combined behavioral and biomedical intervention for HIV prevention among YMSM. Participants enrolled in the study first completed the 3MV behavioral intervention, then were randomly assigned to one of three study arms: 1) daily oral FTC/TDF, 2) daily oral placebo, or 3) no pill. Participants were seen every 4 weeks for 24 weeks. Study visits consisted of HIV/STI testing, brief medical examinations, risk reduction counseling and condom provision. All study procedures were approved by the Institutional Review Boards of the participating sites.

Many Men, Many Voices (3MV) was chosen due to the disproportionate impact of the HIV epidemic on YMSM of color. 3MV explores HIV risk within the context of dual identity for Black MSM, as both racial and sexual minorities. It is based on two behavioral change theories: Social Cognitive theory (Bandura, 1986) and the Trans-Theoretical Model (Prochaska & DiClemente, 1984). 3MV is delivered in a group seminar format, often over a weekend or other 2-day session. In a randomized controlled trial of 3MV, intervention participants reported greater reductions in unprotected anal intercourse (UAI), reductions in the number of sexual partners, and increases in HIV testing (Wilton, et al., 2009). In our implementation of the intervention, we also added an educational module on PrEP that was developed by the ATN’s PROTECT Working Group and reviewed by several youth community advisory boards (CAB) from ATN clinical sites. The purpose of this module was to assure participants’ understanding of PrEP prior to randomization, to introduce them to the need for daily pill-taking and to emphasize accurate reporting of adherence to the regimen.

Recruitment and Screening

Potential participants were approached by project staff at a variety of community-based venues (i.e. nightclubs, balls, college campuses, community-based organizations) and asked about their interest in the study. If interested, participants completed a brief eligibility screening via personal digital assistants (PDA), which uses a self-administered interview format that parallels computer assisted self-interviews. The software automatically determines whether the young man meets the eligibility requirements without staff input or assistance, thus increasing the participant’s privacy. Young men were eligible at screening if they were born male, between the ages of 18 – 22 years, reported at least one episode of unprotected anal intercourse (UAI) with a male in the last 12 months, and were HIV-negative by self-report.

Following the release of the iPrEx study results, the ATN DSMB recommended that all participants be immediately notified of the results of the iPrEx study and suggested that all participants currently enrolled should be unblinded, enrollment into the study be discontinued, all participants on the placebo and no pill arms be offered the option of switching to PrEP, and participants on the active PrEP arm continue as scheduled.

Participants

Sixty-eight young men were enrolled into the study and 58 were randomized. Ten participants were discontinued prior to randomization for various reasons (5 failed to adhere complete initial study evaluations, 4 discontinued because randomization was closed, and 1 withdrew consent/assent,). Participants primarily identified as Black (53%) and Latino (40%) (see Table 1).

Table 1.

Baseline Demographics

Age at Baseline (years)
Mean	19.97
Std	1.30
Median	20.00
Race (%)
Black/African American	31(53.45)
Native American/Alaskan Native	1(1.72)
White	4(6.90)
Other/Mixed Race	22(37.93)
Ethnicity (%)
Hispanic or Latino	23(39.66)
Non-Hispanic or Latino	35(60.34)
Education
Eighth grade or less	1(1.72)
GED	2(3.45)
High School Diploma	18(31.03)
Some College	30(51.72)
Current Employment
Unemployed	32(55.17)
Full time	6(10.34)
Part time	20(34.48)

Adaptation of 3MV

3MV consists of seven sessions that focus on factors that influence the behavior of black men who have sex with men, including cultural, social, and religious norms; interactions between HIV and other sexually transmitted diseases; sexual relationship dynamics; and the social and psychological influences that racism and homophobia have on HIV risk behaviors (CDC, 2011b). For the purposes of Project PrEPare, this intervention presented some inclusion limitations given that it was developed for use with black MSM. Thus, the need to adapt the intervention to allow for inclusions of other ethnic, racial and cultural backgrounds needed to be carefully considered.

In order to involve the community in the design of this study and to discuss the inclusion of all YMSM or only racial minority men, the protocol team brought together a group of community leaders and gatekeepers in Chicago. This group consisted of 13 male opinion leaders representing various ethnic and racial communities, amongst them African American, Latino, and Caucasian, as well as the LGBT community. A number of community-based agencies and advocacy groups who serve YMSM were represented as well. During the discussion, concerns were raised about the historically turbulent relationship that communities of color, particularly African American communities, have had with researchers over the years. The group recommended a recruitment strategy that would be inclusive of all YMSM, but with a specific focus on YMSM of color. Thus, in consideration of the disparate HIV prevalence rates among young MSM of color in the Chicago area, emphasis was placed on ensuring that members of this population were well represented in the study sample. Because of these decisions, 3MV was retained as the behavioral component of the combined intervention due to its cultural relevance, but modification was required in order to be inclusive of all study participants.

In order to modify the intervention in such a way that would maintain the cultural integrity as well as the core elements, the Project PrEPare team invited Kenneth Pettigrew, Director of Programs for the community-based organization Us Helping Us (Washington, DC) and Peter Freeman, Program Coordinator for Howard Brown Health Center (Chicago, IL) to assist. The Project PrEPare staff members were trained, both in Washington DC and in Chicago, to learn the appropriate techniques for facilitating the 3MV intervention and to prepare for a culturally responsive modification. A two-day work-session was scheduled with the principal investigators, project director, and research assistants in which each session of the intervention was critically examined for adaptation based on its focus on race/ethnicity issues. With the assistance of the consultants, the team determined that some exercises, primarily within Session 1, would require modification prior to implementation.

Within Session 1, the exercise entitled “Dual Identity” presented multiple challenges to the modification process. This session is designed to explore issues of dual identity as it relates to race/ethnicity and sexual orientation for Black MSM. Very early on, the team questioned whether the issues related to dual identity even existed among young MSM of other racial/ethnic groups, particularly white YMSM. In an effort to explore this issue, two CAB meetings comprised of YMSM from multiple racial/ethnic backgrounds were convened to obtain feedback on this particular session. CAB members were separated into groups based on their racial/ethnic identities, and the activities of the session were facilitated with each group separately. The entire group was then brought back together to discuss their separate group conversations.

Reports from both the group facilitators and the CAB members indicated that breaking off into small work groups based on race/ethnicity identification was both feasible and acceptable. However, the Caucasian CAB members, belonging to the dominant racial group in America, did not easily identify with the concept of “dual identity” specifically as it related to the intersection of racial identity and sexual orientation (i.e. being a White gay man). The group facilitators made the concept more relevant by guiding the discussion of dual identity to focus on what it means to be a man and its associated gender roles, what it means to be a gay man and its associated stereotypes, including but not limited to a discussion about the diversity of the LGBT community (i.e., the club scene, the leather scene, etc).

Bringing the conversation about dual identity back to the larger group also proved challenging. In particular, the young Black men in the group displayed significant discomfort with sharing their conversations about dual identity with men of other races. They explained the heightened stigma associated with being a Black gay or bisexual man, even within the larger LGBT community. This stigma is strongest when the disparity in HIV infection rates is discussed in the intervention, which resulted in feelings of undesirability by the young Black MSM who are more likely to be HIV positive than their white and Latino peers (CDC, 2012).

Based on feedback from the CAB members and discussions with the consultants, the Project PrEPare team decided that it would be best to separate the young men based on their racial/ethnic identity for this exercise. A brief break was also instituted immediately following the activity, with a return to the larger group for continued discussion on the impact of dual identity on self-concept, self-esteem, and self-efficacy, regardless of racial/ethnic identity. Each group was instructed to bring back only the “issues list” they created that was associated with being a gay man within their particular racial/ethnic group for discussion. Facilitators were instructed to focus the larger group discussion on the similarities between the different groups, focused primarily on discrimination and oppression based on sexual orientation, which set the stage for the conversation moving forward.

Process Evaluation of the Modified 3MV

In order to evaluate the acceptability of the program modifications as well as the inclusion of the 3MV intervention within the context of a larger biomedical trial, a process evaluation was conducted.

Process Evaluation Measures included the following a) Session Evaluation Form (SEF); (Harper et al., 2003), a brief questionnaire with 10 questions (see Table 2) on a 4-point Likert scale and 3 open-ended questions given to participants at the end of each session in order to assess overall satisfaction with the sessions as well as elements that they liked and disliked, b) Fidelity Logs completed during review of each digitally recorded session to document which elements of the intervention were delivered and to record unique issues that arose during the session, an c) Acceptability Questionnaire, administered by Audio Computer-Assisted Self-Interview (ACASI) at the final study visit, which queried participants on how they felt about attending the group-based 3MV intervention and how they felt about the ongoing risk-reduction conversations that occurred throughout the trial, and d) Qualitative Interviews completed with six participants (two from each arm of the study) who previously consented to the qualitative interview during their initial enrollment in Project PrEPare. Study staff from each site were asked to select youth for the qualitative interviews that covered a range of possible study experiences (e.g. those that were adherent to study visits vs. those who were not); the qualitative interviewers did not participate in the selection of participants to be interviewed. Each interview lasted about an hour and participants were asked to share their experiences with multiple aspects of Project PrEPare. Regarding 3MV, participants were asked about their overall impressions of the intervention, what they found to be most helpful and least helpful, and what activities stood out to them. All interviews took place in a confidential space at the lead investigator’s institution and were recorded using digital voice recorders.

Table 2.

Session Evaluation Form Responses

Item	Strongly Agree	Agree	Disagree	Strongly Disagree
I learned a lot from this workshop.	86%	12%	2%	-
I will be able to apply what I learned from this workshop in my life.	80%	20%	-	-
I was given an opportunity to participate and discuss information with others.	93%	7%	-	-
This workshop was well organized.	93%	7%	-	-
The topics in this workshop were interesting.	83%	17%	-	-
The presenter stimulated my interest in the material	85%	15%	-	-
The topics in this workshop were relevant to my life.	80%	20%	-	-
I found this workshop enjoyable	85%	15%	-	-
I would recommend this workshop to others.	93%	7%	-	-
I felt comfortable participating in this workshop.	93%	7%	-	-

These combined data serve to reflect the delivery of intervention components and participants’ engagement and comprehension. They also reflect the participants’ perceptions of the acceptability of the behavioral intervention alone compared to the intervention in the context of a biomedical trial. Data from the Session Evaluation Forms and Fidelity Logs were summarized to determine problems and successes of each intervention session. Descriptive data was generated from the responses to the Acceptability Questionnaire. Qualitative interviewers followed a semi-structured interview guide with suggested probes to elicit and clarify responses. After reviewing the digital recordings of the qualitative interviews individually, members of the study team generated codes which were discussed during team meetings and modified based on group consensus. From this, the protocol team developed a coding structure, (i.e., a hierarchical set of constructs that account for the phenomena seen in the data) and then further developed and refined the list of codes as additional transcripts were coded and discussed. The intent with this process data was to assess what worked and what did not work in the modified intervention as well as explore the perceived value of the behavioral intervention to the study participants.

RESULTS

Session Evaluation Forms

Data from the Session Evaluation Forms demonstrated high acceptability for the 3MV sessions. The quantitative responses (see Table 2) were favorable with participants finding the intervention educational, enjoyable, and relevant to their lives.

The open-ended question regarding what participants found to be most useful about the intervention yielded responses in two primary categories: Dual Identity (the focus of Session 1) and STI/HIV information. Those that found the Dual Identity exercises helpful wrote responses such as “I found it useful to know that subconsciously we still allow the negative stereotypes to affect us. It shows me how important it is to fight them,” and “the Self-Concept [exercise] because it made me realize who I am what to have in life and how I act on it.” Participants wrote responses such as “The STD/STI prevention - I didn’t know so much about it at first” and “The facts of HIV and STDs because it’s important to know the risk of unprotected sex.”

Most participants answered the question about what they found to be least useful about the intervention with “nothing”. A couple of participants reported that the exercise on the “Freaky Things Other People Do” wasn’t useful because “because it didn’t help with being safe during sex, but it was funny” and “because it’s kinda like a tongue and cheek topic which can kinda be kinda draggy.”

The final open-ended question asked participants what they would change about the intervention. Many participants reported that they wouldn’t change anything (“absolutely nothing – it was fabulous!”). There were multiple comments around time with many participants wanting more time for the intervention (“more time/more days to learn”), some wanting less time (“the time length”) and a few wished it was scheduled later in the day (“The time was a bit early for my taste”).

Fidelity Logs

Digital recordings of the sessions were reviewed for fidelity to the intervention exercises and the core elements of 3MV. Two reviewers found 100% fidelity to intervention components along with high levels of engagement from participants.

Acceptability Questions

Along with other study elements, participants reported high acceptability for the group-based 3MV sessions as well as the ongoing counseling and conversations around sexual behavior (see Table 3).

Table 3.

Acceptability Questionnaire Responses

Item	Liked a lot	Liked	Did not like	Did not like at all
How well did you like participating in the 3MV group sessions?	71%	17%	9%	3%
How well did you like receiving risk reduction counseling at every session?	69%	28%		3%
How well did you like answering questions about your sexual behavior at every visit?	26%	68%	3%	3%

Qualitative Interviews

Overall Acceptability of 3MV

Consistent with the other evaluation measures, participants reported highly positive experiences with the behavioral intervention.

“Overall I enjoyed 3MV. For me it was like the most fun experience I’ve had with HIV awareness and STI awareness and behavioral prevention. That was the most influential program I’ve ever been in.” (Qualitative Interview #1)

“I think it was good to have that time put aside for me actually like being educated and the whole like sexually transmitted disease… I think that was something that was good and I was able to put my point of view out there, hear people’s point of views and see how it was on paper, because sometimes you have to see things on paper in order to really comprehend it so I think that was important.” (Qualitative Interview #2)

Generally, study participants reported 3MV as being very engaging and enjoyable. The mixed ethnic and racial group dynamics led to rich discussions and social interactions which some of these youth may not have experienced prior to their participation in 3MV. During later study visits, participants often asked research staff if they could continue attending future 3MV sessions.

What was most helpful for you?

Many of the participants also cited the topics from Session 1 around sexual orientation, stereotypes, identity and gender roles as being useful.

“The one discussion I do remember was the one about sexuality like in our own community and how it’s different, how people take the roles of like heterosexuality. They kind of want to put them into homosexual relationships and stuff like that, like taking the role of a man and taking the role of a woman and stuff like that. That’s one of the discussions I remember.” (Qualitative Interview #3)

“The thing I found most helpful and the most relevant were the gender roles…I just found that part to be so interesting because even in straight relationships there are roles out there and it just happened that those are placed on the gay relationships and it’s so not true. I feel this huge misunderstanding about gay people is that as my mom says eloquently ‘there’s someone taking the dick and someone giving it and the one taking it is a flaming homo and it’s just inserting it’ and that really bothers me. It’s funny because I have a boyfriend who doesn’t live in Chicago and he will be in town in a few weeks and I want him to go out to dinner with my mom. I want her to meet him and I think I said to her, “When you tell me which one of us is the role of the boy and who gets on the top.” (Qualitative Interview #4)

Participants enjoyed and found it helpful to participate in a guided discussion with their peers about topics like sexual orientation, stereotypes, identity and gender roles and expressed comfort with the disclosure of their personal experience with peers from other ethnic/racial backgrounds.

All the qualitative interview participants felt that the level of knowledge that they gained around HIV and STI transmission was helpful.

“I found everything to be pretty helpful. I learned a lot, I mean a lot as far as the process of contracting HIV as a whole and what happens, what happens when a membrane is breaking and all that, the technical aspect which was really really useful, the risk reduction. I practice it now and I’m a little bit more careful now that I’ve learned a lot from it.” (Qualitative Interview #5)

“The discussions about…now I’m a little foggy on the names that were given, the pink parts and the T-cells. So I’ve learned not to use harsh soaps or anything like that or mouthwash after you’ve got done having oral sex.” (Qualitative Interview #6)

The discussions around HIV and STIs during 3MV served to prepare the participants for in-depth individual discussions around these topics with the study staff as they continued their participation in the study.

What did you find least helpful?

Participants were asked to discuss what they found least helpful about the intervention. Several participants stated that they wouldn’t change anything, but others were able to identify areas for improvement.

“The icebreaker exercise, because I think by us not really being too familiar…it’s kind of hard to break the barrier and feel comfortable enough asking people certain questions. You don’t want to get too personal but you don’t want to be Oprah Winfrey or anything. It’s a trick method. You have to do it in a way that it’s there for you so you can just kind of lay it out there. I think that was probably it because we had to ask a lot of questions and I was like oh brother, asking them all this stuff because I don’t want them looking like I’m trying to get too personal with them. That probably could have been changed a little different, like instead of giving us more voice it could have been more voice on the paper so they could feel comfortable like I could feel comfortable.” (Qualitative Interview #2)

“I think it was okay as it was, but one thing I would have liked more was the part about learning about the diseases. If you guys would have spent a bit more on that I think that would have been helpful.” (Qualitative Interview #3)

“The discussions about two things – the discussions about syphilis because I already knew about syphilis and it was a small part of the discussion as a whole, and the discussion of relationships and the affects they have on our behavior, sexual and personal and how do you say it - monogamous. Monogamous relationships and open relationships, the reason being is because there’s a lot that…I guess having an open dialogue helps with the understanding of it on a personal level, but you can’t really generalize that topic.” (Qualitative Interview #5)

3MV was facilitated by the study team members primarily on the weekends and included whichever participants were ready for that step in the study process. The demographics of each group varied in age, ethnicity, sexual orientation, current gender, and education, amongst other factors. This diversity in the group composition led to very unique experiences for each cohort, hence the variability in the participants’ feedback.

DISCUSSION

The incorporation of an evidence-based group-level behavioral intervention into a PrEP feasibility trial involved careful consideration of the needs of the target population as well as thoughtful consultation with community members and intervention experts in order to insure that the modification was relevant to study participants. The resulting intervention was found to be highly acceptable to the youth and potentially beneficial to overall study recruitment and high retention rates (98.5%). However, no data that could specifically demonstrate the relationship between 3MV acceptance and study recruitment and retention were collected. Ongoing risk reduction conversations and counseling sessions throughout the study were also well-received.

Within the current public discourse concerning real world implementation of biomedical prevention strategies such as PrEP, questions abound with respect to the “right mix” of intervention (biomedical and behavioral) necessary to produce the greatest preventative effect. At the forefront of this discussion is the acceptability and feasibility of utilizing presently available evidence-based behavioral interventions, which have limitations including only moderate levels of efficacy and few available resources to bring multi-session approaches to scale (Sullivan et al., 2012). Many of these interventions, however, have been designed to address the sociocultural drivers of HIV infection among the high-risk populations for which they are intended, a characteristic believed to play a significant role in sustaining behavioral change over time.

Project PrEPare sheds some light on the potential impact of incorporating a group-level behavioral intervention into the implementation of a biomedical prevention strategy. While many participants and even some study staff initially perceived the time commitment involved with attending the behavioral intervention as a potential barrier to recruitment, word of mouth regarding the positive experience of the weekend ultimately became a draw for some participants. As previously mentioned, participants reported that they enjoyed the time spent bonding with their peers and the wealth of information that they received from the curriculum. Moreover, engagement with participants over the course of a weekend allowed for the study staff to establish rapport that aided in the high overall retention in the study. Participants reported that they appreciated the opportunity to regularly meet with a team of non-judgmental healthcare providers (i.e., mental health specialists, sexual health educators, adolescent medical practitioners) about various aspects of their lives. Participants also felt comfortable sharing multiple sources of contact information, including social media handles, which allowed for consistent and discrete access to them (Purnell et al., 2012).

Real-world providers must consider the subsequent effects of group-level behavioral interventions on client/staff relations when implementing biomedical interventions. Interim guidelines from the CDC and subsequent patient instructions from the FDA support that adherence to the PrEP regimen as well as frequent HIV/STI screening is critical for optimal success of this biomedical approach (CDC, 2011; FDA 2012). Utilizing a behavioral intervention such as 3MV in real world settings could support the required medication adherence, retention to HIV testing guidelines, and improve sexual health promotion through risk reduction skills building.

Another heavily debated concern related to PrEP implementation has to do with the role that community based organizations (CBOs) will play, if any. Many who work within the CBO sector believe that PrEP symbolizes a structural movement towards the medicalization of HIV prevention, and that the role of CBOs will diminish over time as a result. Observations from Project PrEPare may have implications for the potential role that CBOs could assume in PrEP implementation.

CBOs have expertise in delivering behavioral interventions of all sorts, and with linking those interventions to ancillary services. For Project PrEPare, the 3MV sessions were conducted at local community-based organizations whose staff were members of the research team and could guide participants to other services not offered through the study. In the real world, CBOs with strong relationships to medical clinics/providers, or with clinics/providers on sight, could support PrEP implementation by facilitating behavioral interventions (at all levels), maintaining relationships with clients, assisting participants with adhering to HIV/STI testing requirements, and sharing the burden of HIV/STI testing services. Theoretically, it is also more cost-effective to deliver the aforementioned services in community-based settings rather than strictly clinical settings, because service delivery does not require the same extensive degree of specialization (nurse practitioner, doctor, etc.) or the costs associated with clinical visits (i.e. registration, etc).

Despite the strong acceptability and feasibility associated with the combination intervention approach of Project PrEPare, generalization is limited due to the small sample size and the fact that it was implemented in the context of a clinical trial. Additionally, the large, urban setting within which the study was conducted allowed for access to adolescent-friendly and LGBT-friendly settings, which may not be the case in other geographic locations. Finally, measures to directly impact the influence of the behavioral intervention on retention, adherence and behavioral risk would have been useful. As the positive evidence for PrEP effectiveness grows, it is important that demonstration projects are launched to explore a variety of combination approaches toward HIV prevention, especially among adolescents and young adults.