Adolescent-Centered HIV Prevention: Perspectives on Acceptability of Oral Antiretroviral Pre-exposure Prophylaxis for Adolescents in a Global Priority Setting

Abstract

With oral antiretroviral pre-exposure prophylaxis (PrEP) rollout expanding to include adolescents in South Africa, research is needed to better understand perceptions of PrEP acceptability among adolescents and clinical service providers. We conducted an exploratory mixed-methods study among 57 adolescents, 16–17 years of age, living with and without HIV, and 25 clinical service providers in Cape Town, South Africa from 2015 to 2016. Cross-sectional survey and semi-structured qualitative interview data were used to explore (1) willingness to use PrEP and support partner PrEP use among adolescents living with and without HIV, (2) willingness to prescribe or support prescription of PrEP among service providers, and (3) perceptions of barriers and facilitators to PrEP implementation and interpretations of PrEP efficacy messaging for adolescent HIV prevention among all participants. Acceptability of PrEP among participants was high. Support for PrEP uptake was linked to messages that positively framed PrEP’s protection potential (i.e., success- versus failure-framed messaging) among both adolescents and providers. Adolescents living without HIV endorsed high willingness to use PrEP and adolescents living with HIV endorsed high support for partner PrEP use. However, both groups noted that potential side effects, stigma, and PrEP’s partial efficacy may hinder uptake. Clinical service providers endorsed PrEP for sexually active adolescents and shared stigma and efficacy concerns. Further, service providers expressed desire for adolescent-tailored training and integration of PrEP delivery into primary care and family planning services. Efforts to educate adolescents and service providers about PrEP should consider how message framing may influence acceptability. Community PrEP education and adolescent-friendly delivery should be prioritized to alleviate predicted PrEP stigma and facilitate uptake.

Introduction

Oral antiretroviral pre-exposure prophylaxis (PrEP) is one of the most important biomedical strategies in our toolkit for HIV prevention. While there is growing evidence of PrEP acceptability among adults (Fonner et al., 2016), research has more recently began exploring the safety and acceptability of PrEP for adolescents, who account for over one-third of all new global infections of HIV (UNAIDS, 2016). Specifically, two adolescent-focused safety studies, Adolescent Medicine Trials Network for HIV/AIDS Interventions 113 (ATN 113) and PlusPills, have demonstrated PrEP’s safety and utility among adolescents in United States and South Africa, respectively (Gill et al., 2020; Hosek et al., 2017). South African adolescents in particular represent a key population for prevention efforts with unique needs with respect to new biomedical HIV prevention technologies (Hosek et al., 2016; Shisana et al., 2014). Despite promising safety data from preliminary adolescent PrEP trials, there remains limited knowledge regarding opinions surrounding PrEP among South African adolescents and their partners, or among service providers involved with PrEP implementation.

Adolescents undergo significant biological and psychological changes and often experience unique developmental milestones, including sexual debut and increased engagement in romantic and/or sexual relationships, which are likely to influence PrEP acceptability, willingness to use PrEP, and understandings about PrEP efficacy (Grotevant & Cooper, 1985; Patton et al., 2016; Viner et al., 2012). Adolescence has also been characterized as a period of higher impulsivity, lower behavioral control, and greater risk-taking behavior (Dir et al., 2014; Steinberg, 2008), further highlighting the need to better understand how the developmental stage of adolescence influences the acceptability of HIV prevention strategies, including PrEP. Moreover, given the frequency of serodiscordant partnerships in South Africa (De Walque, 2007; Eyawo et al., 2010), research is also needed to understand acceptability of partner PrEP use among HIV-infected adolescents. The current generation of young people living with HIV in South Africa consists of many individuals who were infected at birth and survived into sexual maturity due to the availability of antiretroviral therapy (ART). These HIV-positive young people can offer relevant perspectives on how PrEP can be integrated into serodiscordant partnerships in combination with ART adherence and how to navigate status disclosure to facilitate partner PrEP initiation.

Adolescent clinical service providers can also provide useful perspectives about the acceptability of PrEP implementation with adolescents, due to their front-line roles in sexual and reproductive health counseling and medication prescription. Service providers, for example, may be able to offer perspectives about characteristics or subtypes of adolescents that should be eligible for PrEP, potential implications of PrEP prescription, and suggestions for facilitating adherence among adolescent PrEP users. Prior research has explored perceptions of PrEP among clinician samples in the United States (Koechlin et al., 2017). Clinical service providers working in HIV endemic settings such as South Africa may have different understandings of PrEP’s utility with respect to adolescents and young adults.

PrEP implementation efforts would further benefit from a greater understanding of how adolescents and clinical service providers interpret and make decisions based on messages about PrEP’s efficacy. Research with men who have sex with men (MSM) in the United States has demonstrated a range of comprehension and operational understandings of PrEP efficacy messages, as well as preferences for percentage-based efficacy messages (as opposed to verbal paraphrases about PrEP’s efficacy, such as “highly effective”) and success-framed efficacy messages (i.e., statements describing PrEP protection potential versus its rate of failure) (Underhill et al., 2016). Improved understanding about how messaging about PrEP’s efficacy might affect acceptability among adolescents can inform the design of behavioral interventions to support adolescent PrEP users. Such an understanding is especially crucial in the context of the recent PlusPills study (Gill et al., 2020) which demonstrated initial safety and tolerability of PrEP among South African adolescents, yet also found suboptimal user adherence over time.

We examined perceptions of PrEP acceptability among three key groups: HIV-uninfected adolescents, who are potential PrEP users; adolescents living with HIV, with whom potential adolescent PrEP users might engage in sexual partnerships; and clinical service providers who work with adolescent patient populations and who might prescribe or recommend PrEP to this population. The aims of this paper were to: (1) provide insight into perceptions of PrEP acceptability among potential adolescent PrEP users and their potential partners, and explore predicted barriers and facilitators to PrEP use; (2) provide insight into service providers’ willingness to prescribe or support PrEP prescription for adolescents, and explore their perceived barriers and facilitators to PrEP implementation; and (3) describe how messages of PrEP’s efficacy based on level of adherence and message framing influence adolescent perceptions of PrEP acceptability and service providers’ willingness to prescribe PrEP to adolescents.

Method

Participants

We conducted an exploratory mixed-methods study among adolescents and clinical service providers in Cape Town, South Africa from 2015 to 2016 (Giovenco et al., 2018). We focused on an HIV endemic setting to leverage lessons from adolescents navigating serodiscordant relationships and captured perspectives from both perinatally infected adolescents, who may engage in serodiscordant partnerships, and HIV-uninfected adolescents, who could potentially benefit from taking PrEP. We then triangulated adolescent and service provider perspectives on PrEP acceptability to further inform future implementation considerations.

Adolescents living with HIV were recruited from adolescent HIV treatment clinics and were eligible if they met the following inclusion criteria: (1) 16–17 years, (2) self-reported HIV-positive status, and (3) self-reported comfort discussing HIV status in an adolescent group. HIV-negative adolescents were recruited via door-to-door community sampling and were eligible if they were (1) 16–17 years of age and (2) self-reported HIV-negative status. A trained staff member obtained informed assent and parental/caregiver consent from all participants. We excluded those who reported unknown HIV status, were unable to provide informed assent, or were unable to obtain informed consent from a parent or caregiver.

Measures and Procedure

Adolescent participants completed a brief ACASI-administered sociodemographic survey that included questions about sexual behavior and PrEP acceptability, with varying messages of PrEP’s efficacy based on levels of adherence and success- versus failure-framed messaging (see supplemental Appendix A). Each PrEP acceptability question was rated on a 5-point Likert scale ranging from “I would definitely want [my partner] to use PrEP” to “I would definitely NOT want [my partner] to use PrEP”. Then, prior to the start of qualitative procedures, facilitators verbally provided adolescents with additional information about PrEP’s estimated efficacy derived from findings from adult trials that had been completed at the time of the study, as well as information on known side effects, HIV testing requirements, and the importance of adherence. Focus group discussions were conducted among HIV-positive and HIV-negative participants separately, and discussions aimed to explore PrEP knowledge and acceptability, willingness to use PrEP or to support partner use, and how messages of PrEP’s partial efficacy might affect uptake and adherence. Focus groups were mixed with regard to gender composition. Focus group participants were then invited to participate in in-depth individual interviews to explore topics in further depth and probe for more sensitive issues, such as how stigma and disclosure might affect acceptability of PrEP, that participants may not have been comfortable sharing in a mixed-gendered group setting. Participants who expressed interest were enrolled into individual interviews until data saturation was reached. Adolescent focus groups and individual interviews were conducted in isiXhosa or English, lasted approximately 1–1.5 h, and were digitally audiotaped. Participants received 150 ZAR (approximately 10 USD) for completing each focus group and/or individual interview.

We also conducted in-depth individual interviews with service providers, which included HIV specialists and non-HIV specialists with experience working with adolescents in this geographic and social context. Provider participants were professionals that would potentially be involved in prescribing or supporting future adolescent PrEP use, as well as those with relevant experience handling adherence challenges and discussing daily medication use with young people. An initial seed-pool of provider participants was generated in consultation with the study team, and these initial participants referred professional peers to the study. Service providers were eligible if they were (1) 18 years or older and (2) had three or more years of experience providing health-related services to adolescents. Following written consent, service providers completed a brief quantitative survey administered via ACASI software. The survey included sociodemographic questions and questions about the service provider’s willingness to prescribe or support prescription of PrEP, with varying messages of PrEP’s efficacy based on levels of adherence and success- versus failure-framed messaging, scored on a 5-point Likert scale (see supplemental Appendix B). Following the survey, service providers were given a brief description of oral PrEP and individual interviews guided by semi-structured protocols were conducted. Interviews were aimed at exploring provider participants’ opinions on willingness to prescribe PrEP and which type of adolescents should be eligible for PrEP, predicted barriers and facilitators to PrEP implementation, and suggestions for supporting uptake, adherence, and protective behaviors during future PrEP rollout among adolescents. Interviews were conducted in isiXhosa or English, lasted approximately 1–1.5 h, and were digitally audiotaped. Service providers received 300 ZAR (approximately 20 USD) for completing an individual interview.

Analytic Plan

We conducted descriptive analysis of quantitative data using SPSS 24.0 and thematic analysis of qualitative data using NVivo 10. Qualitative data were transcribed verbatim and then translated into English. Analysis was guided by Braun and Clarke’s (2006) thematic analysis framework, which is a systematic process for identifying themes and patterns related to our overarching research questions. First, common words, phrases, sentences, and ideas from the transcripts were clustered to develop a codebook. All transcripts were then independently coded by two coders using open-coding, axial coding, and coding of marginal remarks and comparisons. To assess for intercoder reliability, the coders independently reviewed a subset of transcripts and examined the degree to which the coding scheme was consistently applied. A third member of the research team helped resolve inconsistencies across coders. Excerpts under specific codes and sub-codes were compiled across all interviews and meaning from these codes was formulated to produce clusters of themes presented below.

Results

Sample Characteristics

Participants were 57 adolescents and 25 service providers (see Table 1). We conducted five focus groups with adolescents living with HIV (involving 24 participants total) and five focus groups with HIV-negative adolescents (involving 31 participants total). Individual interviews were conducted among 10 HIV-positive and 25 HIV-negative adolescents, which included a subset of 33 adolescents who participated in the focus groups and two additional adolescents who missed their scheduled focus group session. Lastly, we conducted 25 interviews with clinical service providers (adolescent HIV researchers, doctors, nurses, counselors, etc.).

Table 1.

Participant socio-demographic characteristics

Adolescents (n=57)		N (%)	Service Providers (n=25)		N (%)
Age	16 years	32 (56%)	Age		41 (30–63)*
	17 years	25 (44%)
			Gender	Female	24 (96%)
Gender	Female	32 (56%)		Male	1 (4%)
	Male	24 (42%)
	Other	1 (2%)	Race	Black African	12 (48%)
				White	10 (40%)
HIV status	HIV negative	33 (58%)		Other	3 (12%)
	HIV positive	24 (42%)
			Primary language	IsiXhosa	9 (36%)
Race	Black African	55 (96.5%)		English	13 (52%)
	White	2 (3.5%)		Afrikaans	1 (4%)
				Other	2 (8%)
Primary language	IsiXhosa	53 (93%)
	English	3 (5%)	Education	< Grade 12	2 (8%)
	Afrikaans	1 (2%)		Grade 12/Certificate	10 (40%)
				Bachelors/Honours	3 (12%)
Education	≤ Grade 6	2 (4%)		Master’s degree	5 (20%)
	Grade 7–8	18 (31.5%)		Doctoral degree	3 (12%)
	Grade 9–10	34 (59%)		Other	2 (8%)
	Grade 11–12	3 (5.5%)
			Profession	Doctor	10 (40%)
Sexual Orientation	Heterosexual	50 (88%)		Counsellor	7 (28%)
	Bisexual	3 (5%)		Nurse	4 (16%)
	Gay or Lesbian	2 (3.5%)		Psychologist	2 (8%)
	Other	2 (3.5%)		Other**	2 (8%)

^*Mean (Range)

^**Other included a patient support manager and a researcher

Adolescents were 16–17 years of age, 56% female, and 97% Black African. Most (93%) spoke primarily isiXhosa, and 96% had completed grade 7 or higher. Twenty-four of the adolescent participants (42%) identified as living with HIV. The majority (88%) of adolescents identified as heterosexual. More than two-thirds (70%) of adolescents had ever engaged in vaginal sex and 16% had ever engaged in anal sex. Of those who reported any vaginal sex, few (16%) reported having one or more vaginal sex partners of HIV-positive status, and 34% reported one or more vaginal sex partners of HIV-unknown status. Less than half (44%) of adolescents who had ever engaged in anal sex reported one or more anal sex partners of HIV-positive status, and 56% reported one or more anal sex partners of unknown HIV status. Among sexually active adolescents, 30% of adolescents reported inconsistent condom use with a casual partner and 53% reported inconsistent condom use with a main partner.

Service providers ranged in age from 30 to 63 years (mean age = 41 years) and were primarily female (96%) and Black African (48%). Approximately half (52%) spoke English at home, 48% had a high school or equivalent degree, 12% had a bachelor’s degree, and 32% had a master’s or doctoral degree. The sample consisted primarily of doctors (40%), counselors (28%), nurses (16%), and psychologists (8%).

Quantitative Findings

Among HIV-negative adolescents, 82% responded they would “definitely” or “probably” want to use PrEP in the context of sex with an HIV-positive partner or a partner of unknown HIV status. Further, among HIV-positive adolescents, 100% responded they would “definitely” or “probably” support the use of PrEP by an HIV-negative partner (see Table 2).

Table 2.

Adolescent PrEP acceptability and service provider willingness to prescribe or support prescription of PrEP

HIV-negative adolescent questions (n=33)	I would definitely or probably want to use PrEP	I don’t know if I would want to use PrEP	I would definitely or probably not want to use PrEP
Imagine that you are HIV-negative, and you have a partner who is HIV-positive. How would you feel about using PrEP?	27 (82%)	6 (18%)	0 (0%)
Imagine that you are HIV-negative, and you have a partner whose HIV status you do not know. How would you feel about using PrEP?	27 (82%)	4 (12%)	2 (6%)
HIV-positive adolescent questions (n=24)	I would definitely or probably want my partner to use PrEP	I don’t know if I would want my partner to use PrEP	I would definitely or probably not want my partner to use PrEP
Imagine that you are HIV-positive, and you have a partner who is HIV-negative. How would you feel about your partner using PrEP?	24 (100%)	0 (0%)	0 (0%)
Service provider questions (n=25)	Definitely or probably yes	Not sure	Definitely or probably no
Suppose that PrEP is proven to be safe and effective. How likely would you be to prescribe or support prescription of PrEP to sexually active young people who are under 18 years of age?	25 (100%)	0 (0%)	0 (0%)
How likely would you be to prescribe or support prescription of PrEP to sexually inactive young people who are under 18 years of age and considering sexual debut?	17 (68%)	7 (28%)	1 (4%)
How likely would you be to prescribe or support prescription of PrEP to sexually inactive young people who are under 18 years of age and using intravenous drugs?	19 (76%)	4 (16%)	2 (8%)

Willingness to use PrEP among HIV-negative adolescents varied by adherence-based efficacy messages and message framing (see Table 3). For example, when asked about interest in using PrEP daily with 90% efficacy, 82% of adolescents reported they would “probably” or “definitely” want to use PrEP. When asked about interest in using PrEP most days with 60–70% efficacy, 76% reported they would “probably” or “definitely” want to use PrEP. Success versus failure message framing also influenced PrEP acceptability. Adolescents were more likely to report that they would want to use PrEP if messages were framed based on PrEP’s protection potential as opposed to their rate of failure. For example, “PrEP reduces risk for HIV infection by 90%” (79%) was more acceptable than “PrEP has a failure rate of 10%” (64%).

Table 3.

PrEP efficacy messages based on levels of adherence and success- versus failure-framed messaging

HIV-negative adolescent questions (n=33)	I would definitely or probably want to use PrEP	I don’t know if I would want to use PrEP	I would definitely or probably not want to use PrEP
Imagine you had to always use PrEP as prescribed, taking pills on schedule so that PrEP is 90% effective. How would you feel about using PrEP?	27 (82%)	4 (12%)	2 (6%)
Imagine you had to typically use PrEP as prescribed, taking pills most days but missing some pills so that PrEP is 60–70% effective. How would you feel about using PrEP?	25 (76%)	5 (15%)	3 (9%)
Suppose PrEP reduces risk for HIV infection by 90%. How would you feel about using PrEP?	26 (79%)	6 (18%)	1 (3%)
Suppose PrEP has a failure rate of 10%. How would you feel about using PrEP?	21 (64%)	6 (18%)	6 (18%)
Service provider questions (n=25)	Definitely or probably yes	Not sure	Definitely or probably no
Suppose that a young person (<18 years) will always use PrEP as prescribed, taking pills on schedule so that PrEP is 90% effective. How likely would you be to prescribe or support prescription of PrEP to this young person?	25 (100%)	0 (0%)	0 (0%)
Suppose that a young person (<18 years) will typically use PrEP as prescribed, taking pills most days but missing some pills so that PrEP is 60–70% effective. How likely would you be to prescribe or support prescription of PrEP to this young person?	19 (76%)	4 (16%)	2 (8%)
Suppose PrEP reduces risk for HIV infection by 90%. How likely would you be to prescribe or support prescription of PrEP to sexually active young people who are under 18 years of age?	24 (96%)	1 (4%)	0 (0%)
Suppose PrEP has a failure rate of 10%. How likely would you be to prescribe or support prescription of PrEP to sexually active young people who are under 18 years of age?	19 (76%)	4 (16%)	2 (8%)

Adolescents and service providers answered all variations of efficacy messages

All service providers reported that they would “definitely” or “probably” prescribe PrEP or support the prescription of PrEP to sexually active adolescents under 18 years of age and 68% reported that they would prescribe PrEP to sexually inactive adolescents who were considering sexual debut. Over three-quarters (76%) said they would prescribe or support the prescription of PrEP to sexually inactive adolescents who use injection drugs (see Table 2).

Willingness to prescribe or support the prescription of PrEP varied by adherence-based efficacy messages and message framing (see Table 3). When asked about likelihood of prescribing or supporting PrEP for an adolescent under 18 years who will take PrEP daily with 90% efficacy, 100% reported they would “definitely” or “probably” prescribe PrEP to this adolescent. Further, when asked about likelihood of prescribing PrEP to an adolescent who would take PrEP most days with 60–70% efficacy, 76% reported they would “definitely” or “probably” prescribe PrEP to this adolescent. Clinicians were more likely to support the prescription of PrEP to sexually active adolescents if told “PrEP reduces risk for HIV infection by 90%” (96%) as opposed to “PrEP has a failure rate of 10%” (76%).

Qualitative Findings

We explored both adolescents’ and service providers’ qualitative narratives about PrEP acceptability. First, adolescent perspectives on willingness to use PrEP, willingness to support a partner’s PrEP use, and concerns surrounding potential future PrEP use were explored. Second, service provider perspectives on willingness to prescribe PrEP and which adolescents should be eligible for PrEP, concerns surrounding PrEP implementation, and suggestions for facilitating the future rollout of adolescent PrEP were explored. Key themes that emerged from the data are described.

High Adolescent Acceptability but Concerns that Side Effects, Stigma, and PrEP’s Partial Efficacy May Hinder Uptake

Adolescents predicted that their peers would evince high willingness to use PrEP, primarily driven by a lack of awareness around partner HIV status: “I think that we can use it because of what?–We don’t know our partners’ HIV status, most of us that is.” (HIV-negative adolescent; focus group). Adolescents also predicted high support for PrEP among HIV-positive adolescents in serodiscordant partnerships. For example, one adolescent explained: “If you, the person who is HIV positive, really loves your partner and you see a future with them—you would encourage them, ‘Let’s go. Go and ask for this pill’” (HIV-positive adolescent; focus group).

While most adolescents living with HIV agreed they would support an HIV-negative partner taking PrEP, many expressed concerns about discussing PrEP with their partners if they had not previously disclosed their HIV-positive status. One adolescent explained: “The condom prevents almost everything. STIs—all the things in existence. PrEP is meant directly [for HIV]—So, the person needs to first be honest with [their partner] that, ‘I am this way—So, that’s why I want you to use PrEP. So that I don’t infect you’” (HIV-positive adolescent; focus group).

Another adolescent explained that it may be even more difficult to advise casual partners to use PrEP: “Yes, I will advise my partner to take PrEP, but then it’s not easy to be open… Sometimes you have to hide [your status] for a few years—until you see that your relationship is serious. Then you can become open” (HIV-positive male adolescent; individual interview).

Adolescents predicted that both girls and boys would be interested in using PrEP. Girls would be interested in protecting themselves from their male partners and may also be more likely to uptake PrEP since they more regularly visit clinics: “It’s the girls [who may like PrEP]. Because the guys are too lazy to attend a clinic, even if they have a problem that needs them [to go] to the clinic” (HIV-negative female adolescent; individual interview). Adolescents explained that boys would also be interested in PrEP because they are often the ones who engage with multiple partners: “It’s the boys [that would love PrEP]. Because boys are random. They are players—they cheat on girls” (HIV-negative male adolescent; individual interview).

Adolescents described three major concerns with PrEP acceptability. First, adolescents described a fear of side effects, but many noted that protective benefits of PrEP would outweigh potential side effects. One adolescent explained, “No I can use it, but on the other side it terrifies me if you say there is maybe a problem with kidneys or stomach cramps, but I can use it, because I want to be protected against HIV” (HIV-negative male adolescent; individual interview).

Second, adolescents were concerned with the potential stigma that PrEP would elicit. HIV-positive adolescents suggested that their partners’ PrEP use would reveal their HIV-positive status: “Once someone starts PrEP their partner may be [believed to be] HIV positive…You wouldn’t be using it if both of you are HIV negative.” (HIV-positive adolescent; focus group). HIV-negative adolescents expressed fears that PrEP use would be interpreted by partners, peers, and clinicians as an indication of partner infidelity or sexual promiscuity. Further, adolescents, many of whom still live at home, were concerned about PrEP’s resemblance to HIV treatment and feared family members might misinterpret their PrEP use and potential side effects:

I am scared of using PrEP because it will be said that ‘this child is taking pills every day, what kind of a child that takes pills every day? They are sick, there is a sickness that they have.’ Maybe, I am developing pimples on the body and it will be said I have AIDS (HIV-negative female adolescent; individual interview).

To facilitate PrEP implementation and alleviate potential stigma, adolescents highlighted the need for community education. For example, one adolescent suggested: “In order for us to be understood by people in our communities, they can distribute papers and give details about what PrEP is and how PrEP is used” (HIV-negative female adolescent; individual interview).

Third, adolescents expressed concerns regarding PrEP’s partial efficacy. Some adolescents explained that, even when daily PrEP is taken on schedule and efficacy was estimated to be approximately 90%, PrEP’s partial efficacy could deter adolescents:

PrEP is also not 100 percent. Just as the condom is also not 100 % [protection]. There exists that 10 [percent]. So it’s the same. But for me, PrEP is even worse. Because PrEP, in others it has been found to, they become sick or something. Yeah, the condom, I have never heard it being that a person got sick because of the condom (HIV-negative adolescent; focus group).

Other adolescents were not concerned with the partial efficacy of PrEP when taken regularly: “There are people who are not anxious – they won’t ask themselves what happens to the 10%” (HIV-negative male adolescent; individual interview). Another adolescent explained how messages of partial efficacy may reinforce the need for combination prevention: “It’s a 90%. So, you can assist PrEP with a condom” (HIV-negative adolescent; focus group).

Lastly, adolescents expressed concerns that, while PrEP would be helpful to adolescents, uptake may be hindered by perceptions of HIV risk among adolescents: “Today’s kids don’t care about this thing called, like, HIV/AIDS. What they care about the most is pregnancy. They more concerned about it. It’s as if, even if they, like, no matter who they sleep with, it’s as if, like, they won’t get infected” (HIV-negative adolescent; focus group).

Cautious Acceptability Among Clinical Service Providers and a Desire for Adolescent-Tailored Training and Integration of PrEP Delivery into Primary Care and Family Planning Services

Service providers endorsed PrEP for sexually active adolescents: “I think the messages should be whilst an adolescent is sexually active, this should be available for them. It should be and I will prescribe it. I’d rather prescribe that than someone coming back with HIV infection.” Further, providers highlighted the utility of PrEP for adolescents in serodiscordant partnerships and for adolescents who don’t use condoms:

I think [PrEP should be prescribed] in a serodiscordant couple, and I think in older adolescents who are engaging in high-risk sex—not high-risk sexual behaviors as in prostitutes and things like that—I’m just thinking teens who are sexually active who maybe don’t like or don’t want to use condoms.

Some service providers were concerned about prescribing PrEP to adolescents engaged in substance use for fear they might divert PrEP for illicit purposes or have trouble with adherence. However, many service providers agreed that PrEP should be offered to these adolescents since they would likely be at higher risk for acquiring HIV: “…if they are using drugs and alcohol, they’re more likely to get involved in riskier sexual practices, so I think I’d be more likely to prescribe it in that sense, even they’d probably be a bit less adherent.” Further, another service provider explained:

If they were using alcohol and drugs, I don’t believe in discriminating. All people are precious and they might be on alcohol and drugs now but—to play around with their actual lives and not give them the opportunity to grow out of it—I wouldn’t take that as a consideration.

Service providers also expressed concerns surrounding adherence among adolescents: “That’s what worries me about PrEP in adolescence —I’m very happy to prescribe and encourage, but it’s that next step we always struggle with. It’s how do you actually get people to change their behavior because it really is a behavioral concept.” Another service provider explained: “I’m very nervous with this age group of adolescence… adherence is bad, so we are going to make the pill to be not efficacious. It won’t be efficient enough, so we going to have a lot of failures…What will be the next plan for them?”

Similar to adolescents, service providers were concerned that stigma may be a barrier to the rollout of adolescent PrEP given its resemblance to HIV treatment: “These are antiretroviral drugs and currently if someone walks in and sees your Truvada sitting on the shelf, they’re going to assume that you are actually positive…so they may not want to have the drugs in their possession.” Further, another service provider predicted that stigma in clinics providing sexual and reproductive health services may deter adolescents from asking for PrEP:

My experience in some clinics has been where teenagers asked for condoms, that they were scolded because you’re having sex and you’re not supposed to have sex. I think something similar might happen in that way, but I do think it would get support because people want to prevent illnesses.

To facilitate future PrEP implementation, service providers advised that adolescents leverage their support networks who can also help to support adherence:

Adolescents using PrEP will need support. It has to be something that, at one point, that gets to be shared between a partner or even a caregiver, a parent. It doesn’t have to be a secret, because at one stage, I mean, there has to be support so that in case one has forgotten to take the medicine, your partner or even a parent or caregiver all say, ‘Have you forgotten to take your meds?’ It can’t be totally a secret.

Service providers also expressed a desire for adolescent-tailored education and training: “First of all, you yourself needs to understand the product. This means education. The health staff needs to be trained… What are the key messages I must communicate with the person that is going to use it? This is very crucial, otherwise we are missing the boat.”

Similar to adolescents, service providers also expressed concerns regarding PrEP’s partial efficacy. However, while willingness to prescribe PrEP was dependent on the drug’s efficacy, most providers agreed that some degree of incomplete efficacy was acceptable given the level of risk: “…the benefits outweigh the risks at 60–90 [percent PrEP efficacy]. Below 50 [percent PrEP efficacy], you start to think, ‘This kid is not even gonna take it. Why put this drug out there in the community when it could be diverted or smoked?’”.

Lastly, service providers identified integration of PrEP with both sexual reproductive health services and primary health care. One service provider explained: “This should come as a drug therapy with family planning, because they are just enjoying sex the way they want. Then you forget that there is teenage pregnancy… I will go with that, if possible, to make it work, tacked together with the family planning.” Another provider explained:

This should happen at the primary health care level… I think this is everyday life… Why must I take a big gun to kill a fly? This is a fly, this is what I deal with. It’s the fly. When they come to this clinic, the basic needs should be fulfilled for them.

Discussion

Our study found that adolescent PrEP was consistently endorsed among both adolescents (living with and without HIV) and providers in South Africa, but both groups expressed cautiousness and concerns that should be considered in future PrEP implementation efforts. These findings align with the high need for effective HIV prevention strategies for South African adolescents, and also reflect the complex realities for implementing biomedical HIV prevention programs in the South African context. Both adolescents and providers shared strong concerns about PrEP-related stigma as a barrier to uptake and adherence, as well as concerns about partial efficacy of this medication and the need for continued condom promotion for South African adolescents to address problems related to PrEP adherence and ongoing risks for other STIs. Adolescents, but not providers, expressed concern about PrEP’s potential side effects as a barrier to PrEP uptake. Some providers uniquely expressed concern about prescribing PrEP to adolescents who use illicit drugs due to additional adherence challenges in this group and the possibility for medication diversion. However, most providers ultimately endorsed PrEP for adolescents who use illicit drugs given their increased risk of acquiring HIV.

Findings from the first open-label PrEP safety and acceptability study conducted exclusively among adolescents in South Africa were only recently published (Gill et al., 2020), and while PrEP was found to be safe and well-tolerated, adherence was found to be challenging. The current mixed-methods data from 2015 to 2016 remain relevant in light of these recent trial results. While we that expect concerns surrounding the safety and tolerability of PrEP to be less pronounced today, concerns surrounding PrEP delivery, uptake, and adherence are still highly relevant. Our findings underscore the need to develop a tailored implementation approach for adolescent PrEP rollout in South Africa that incorporates considerations for conveying messages of PrEP efficacy to both potential adolescent PrEP users and service providers, and that addresses user and provider concerns surrounding PrEP acceptability and adherence.

Acceptability of PrEP among both adolescents and service providers varied depending on adherence and efficacy messages and message framing. Greater support for PrEP was reported for adolescents who would use daily PrEP on schedule with 90% efficacy. Further, higher support for PrEP uptake was linked to success-framed messaging, which is similar to message framing preferences observed among a sample of MSM in the United States (Underhill et al., 2016). Efforts to educate adolescents about PrEP should be careful to explain adherence and efficacy messages and should consider how message framing may influence PrEP acceptability. Additional research on most effective messaging for adolescents is needed. Further, research also should explore how messages about PrEP may precipitate stigma.

Stigma was predicted as a serious barrier for PrEP uptake among both adolescents and service providers. Adolescents expressed fears that PrEP use would be interpreted by partners, peers, service providers, and family members as an indication of partner infidelity, sexual promiscuity, or HIV-positive status. Service providers elaborated on how stigma expressed or experienced in clinics may deter adolescents from seeking PrEP. This finding is consistent with other PrEP research, where stigma has emerged as a significant social harm that can arise from PrEP use across contexts and can directly impact PrEP uptake and adherence (Eaton et al., 2017; Golub, 2018; Haire, 2015; Peng et al., 2018; van Der Straten et al., 2014). Further, despite advancements in adolescent PrEP research in South Africa, recent qualitative among South African adolescent girls and young women (AGYW) taking open-label daily oral PrEP found that stigma was a significant barrier to obtaining social support for PrEP use (Giovenco et al., 2021; Velloza et al., 2020).

To facilitate PrEP implementation and alleviate stigma, service providers expressed a desire for adolescent-tailored PrEP education and training and advised that PrEP delivery be integrated into both primary health care and sexual and reproductive health services. Further, adolescents highlighted the need for community education to help legitimize their explanations of PrEP to the important people in their lives and service providers recommended adolescents interested in using PrEP leverage their social support networks to help facilitate adherence. These recommendations for PrEP delivery are still relevant for South Africa today. Research has previously highlighted the need to integrate PrEP delivery with other adolescent services, and, more recently, the importance of differentiated service delivery models for PrEP users (Celum et al., 2015; O’Malley et al., 2019). Further, qualitative research among South African AGYW has shown that many groups are still unfamiliar with PrEP, making it difficult for young people to disclose their PrEP use and obtain adherence support (Giovenco et al., 2021). Community-wide PrEP education with careful messaging that does not reinforce stigma and adolescent-friendly PrEP delivery services are necessary to meet the needs of South African youth.

We note that no adolescent participants were currently using PrEP because it had not yet been approved for use by the South African government at the time of the study. Thus, their comments were based on perceptions and speculations about the normative influences of PrEP in their peer groups. Further, due to the qualitative nature of the study, our sample size was small, and our findings might not generalize to other populations. Service providers who were included in this study may be those most open to prescribing PrEP. Additionally, our sample of adolescents was restricted to those who were willing and able to obtain parental consent, which might have selectively included individuals and with greater comfort discussing HIV prevention and greater family support for engaging in sexual health services. Further, because this study took place prior to the release of any demonstration or efficacy data on PrEP with adolescents in South Africa, findings might be temporally bound and are likely to adapt once PrEP is more widely recognized among South African adolescents. Next, we were not able to match speaker identities from adolescent focus groups and service provider interviews with quantitative survey data to add adolescent gender and service provider profession descriptions to relevant quotes. Finally, given the semi-structured nature of the interview guides and our analytic approach, we were not able to quantify the number of participants who endorsed each of the themes presented.

This investigation informs recommendations for future research. First, quantitative research should examine the effects of predicted barriers on PrEP uptake and use. Larger, open-label adolescent PrEP studies are needed for this work. Next, the acceptability of alternative PrEP modalities (topical, injectable, insertable, etc.) and dosing regimens (intermittent, on-demand, etc.) should be examined to determine the relevance of predicted barriers for more discreet or longer-acting PrEP formulations, as well as how the efficacy of various PrEP formulations impacts acceptability. Further, more research is needed to identify feasible, scalable, and cost-effective PrEP delivery strategies for adolescents in South Africa. Given ongoing advances in PrEP research with adolescents (Gill et al., 2020) and alternative PrEP formulations (e.g., long-acting injectables; Clement et al., 2020), continued research on the implementation considerations, developmental readiness, and contextual factors to support safe and appropriate PrEP use among adolescents is warranted.

Conclusions

A strategic, developmentally tailored, and contextually relevant approach to PrEP implementation among adolescents is needed. Efforts to educate adolescents and service providers about PrEP should consider how efficacy messaging and message framing decisions may influence PrEP acceptability. Counseling for adolescents considering PrEP for themselves or their partners and training for service providers prescribing or supporting adolescent PrEP should address perceived barriers, including PrEP-related stigma, and community-wide PrEP education and adolescent-friendly PrEP delivery services should be prioritized. Research is needed to inform the tailoring of PrEP messaging strategies for adolescents, to better understand PrEP implementation challenges in South Africa, and to compare how alternative PrEP formulations may alter perceived barriers to PrEP use.