https://orcid.org/0000-0001-9703-0975
Figures
Abstract
Introduction
Adolescent girls and young women (AGYW) face barriers in accessing clinic-based HIV pre-exposure prophylaxis (PrEP) services and community-based models are a proposed alternative. Evidence from such models, however, is limited. We evaluated PrEP service coverage, uptake, and early persistence among AGYW receiving services through community and hybrid models in Namibia.
Methods
We analyzed routine data for AGYW aged 15–24 who initiated PrEP within HIV prevention programming. PrEP was delivered via three models: community-concierge (fully community-based services with individually-tailored refill locations), community-fixed (community-based initiation and refills delivered by community providers on a set schedule at fixed sites), and hybrid community-clinic (community-based initiation and referral to clinics for refills delivered by clinic providers). We examined proportions of AGYW engaged in services along a programmatic PrEP cascade, overall and by model, and assessed factors associated with PrEP uptake and early persistence (refill within 15–44 days after initiation) using multivariable generalized estimating equations.
Results
Over 10-months, 7593 AGYW participated in HIV prevention programming. Of these, 7516 (99.0%) received PrEP education, 6105 (81.2%) received HIV testing services, 6035 (98.9%) tested HIV-negative, and 2225 (36.9%) initiated PrEP. Of the 2047 AGYW expected for PrEP refill during the study period, 254 (12.4%) persisted with PrEP one-month after initiation. Structural and behavioral HIV risk factors including early school dropout, food insecurity, inconsistent condom use, and transactional sex were associated with PrEP uptake. AGYW who delayed starting PrEP were 2.89 times more likely to persist (95% confidence interval (CI): 1.52–5.46) and those receiving services via the community-concierge model were 8.7 times (95% CI: 5.44–13.9) more likely to persist (compared to the hybrid model).
Citation: Barnabee G, Billah I, Ndeikemona L, Silas L, Ensminger A, MacLachlan E, et al. (2023) PrEP uptake and early persistence among adolescent girls and young women receiving services via community and hybrid community-clinic models in Namibia. PLoS ONE 18(8): e0289353. https://doi.org/10.1371/journal.pone.0289353
Editor: Omar Sued, Pan American Health Organization, UNITED STATES
Received: October 28, 2022; Accepted: July 17, 2023; Published: August 30, 2023
Copyright: © 2023 Barnabee et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: This study was made possible through funding and support from the President’s Emergency Plan for AIDS Relief (PEPFAR) (https://www.state.gov/pepfar/) through the United States Centres for Disease Control and Prevention (CDC) (https://www.cdc.gov) Program Area B – Support for Clinical Service Delivery in Namibia, Grant No. GGH001430 and GGH002242 (Principal Investigator: Gabrielle O'Malley). Jennifer Velloza received support from the National Institute of Mental Health (NIMH K99 MH123369 and NIMH R00 MH123369). The findings and conclusions in this manuscript are those of the author(s) and do not necessarily represent the official position of the funding agencies. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
In sub-Saharan Africa (SSA), adolescent girls and young women (AGYW) are disproportionately impacted by HIV, with 25% of new HIV infections occurring among females aged 15–24 years despite representing just 10% of the population [1]. Daily oral pre-exposure prophylaxis (PrEP) is a biomedical HIV prevention method which is highly protective against HIV infection [2–4]. PrEP is an especially promising prevention strategy for AGYW as it is user-controlled, which may mitigate significant challenges AGYW face in negotiating condom use and utilizing other HIV prevention methods. Beginning with open-label trials and followed by demonstration projects in Kenya and South Africa, PrEP service delivery for AGYW has since expanded to more than 15 countries in SSA, where implementation is largely concentrated in high HIV burden areas [4–7].
Despite this progress, substantial gaps remain in access to PrEP services among AGYW, hindering their uptake of and persistence with PrEP. PrEP service delivery models broadly define the way PrEP services are delivered and to-date, PrEP has largely been delivered via clinic-based models. In these models, PrEP services are often delivered within existing HIV care and treatment clinics where considerable HIV-related stigma can prevent AGYW from seeking and using PrEP [8–10]. Judgmental treatment by healthcare providers (HCP) also acts as a barrier to accessing PrEP; providers sometimes question whether adolescents should be using PrEP or other sexual and reproductive health (SRH) services and/or may refuse to provide such services [11, 12]. Gaps in the quality of healthcare services for adolescents, including insufficient privacy in clinic consultation rooms and waiting areas, long wait times, inconvenient clinic locations, and inaccessible hours (relative to the school day), have also been cited as important barriers for AGYW to accessing and using clinic-based PrEP services [9, 10, 13].
Community-based PrEP service delivery models have been proposed to address these barriers and support AGYW to more readily obtain and use PrEP services [4, 14–17]. Community-based HIV and ART service delivery have demonstrated improved patient outcomes compared to clinic-based delivery models for multiple populations, including adolescents [18–20]. However, evidence from the delivery of PrEP services outside of clinic settings is limited [21]. Understanding the coverage of PrEP services achieved by community-based models (i.e., the proportion of AGYW in high-HIV burden settings using PrEP services such as PrEP education, HIV testing, PrEP initiation, PrEP refills) and measuring factors associated with PrEP uptake and persistence among AGYW accessing services through such models could help to identify successes and gaps in service delivery and inform scale-up in other settings. Given the high rates of early drop-off one-month after PrEP initiation among AGYW reported across PrEP research and programmatic settings, more insight into factors influencing persistence in this early period are particularly needed [15, 22–27].
We evaluated coverage of PrEP services as well as PrEP uptake and early PrEP persistence among AGYW participating in DREAMS (Determined, Resilient, Empowered, AIDS-free, Mentored, and Safe) HIV prevention programming and receiving PrEP through community and hybrid service delivery models in Namibia. We further examined factors associated with PrEP uptake and early persistence to inform future intervention opportunities to promote PrEP use among AGYW.
Methods
Program and population
This study draws from PrEP-related service delivery data for AGYW aged 15 to 24 who participated in DREAMS in Namibia’s Khomas and Zambezi regions between December 2018 and September 2019. DREAMS in Namibia is supported by the United States President’s Emergency Plan for AIDS Relief (PEPFAR) with the goal of preventing new HIV infections among high-risk AGYW through community-based provision of a comprehensive package of evidence-based biomedical, behavioral, and structural interventions, including PrEP [6]. DREAMS launched in Namibia in late-2017 and implementation focused on Khomas, Zambezi, and Oshikoto regions due to their higher estimated HIV incidence among AGYW [28]. From 2017 through 2019, DREAMS in Khomas and Zambezi regions was implemented by the International Training and Education Center for Health (I-TECH) in collaboration with the Ministry of Health and Social Services (MoHSS); the Ministry of Education, Arts, and Culture (MoEAC); the Ministry of Gender Equality, Poverty Eradication and Social Welfare; and partner non-governmental organizations Lifeline/Childline of Namibia and Star for Life [29]. In Khomas, DREAMS implementation concentrated within urban and informal settlement areas in and around Windhoek, Namibia’s capital city. In Zambezi, implementation spanned urban and rural areas, including sites along the main transportation corridor in the region.
Upon entry into DREAMS, AGYW were placed into groups based on age (i.e., 9–14, 15–19, and 20–24) and location. Within these groups, AGYW were provided HIV and gender-based violence (GBV) prevention education, sexual and reproductive health (SRH) services (including HIV services), and social services, such as GBV case management. All interventions were delivered in community-based “safe spaces”–judgment-free, safe, and private locations where groups of AGYW attended regular meetings led by a trained peer or near-peer mentor [30]. In the Namibian context, safe spaces included school classrooms; rooms or open spaces within local non-governmental organizations or community center compounds; rooms in technical colleges or universities; adolescent service delivery rooms in health centers; and church meeting rooms. All locations operated as safe spaces only at designated DREAMS meeting or service delivery days and times and were not open on a walk-in basis. Safe spaces were selected through a mapping exercise that involved interviews with AGYW and community members to gather information such as where AGYW would feel most safe to gather (and where not), how far they would be comfortable to travel, and what resources were already available in the location [31]. In December 2018, a DREAMS eligibility assessment tool was introduced at enrollment into DREAMS programming. The tool was a survey consisting of questions about the AGYW’s background (e.g., education, emotional support, financial support), alcohol and substance use, sexual behaviors (e.g., condom use, concurrent partners), and experience(s) of GBV (e.g., physical, emotional, or sexual violence). The tool was self-administered or administered via support from a DREAMS staff member, if requested. Combinations of age and certain responses were used to identify AGYW possibly more vulnerable to HIV infection and/or GBV and, thus, eligible for secondary DREAMS interventions such as economic empowerment programs.
PrEP service delivery
In Namibia, PrEP for the prevention of HIV was introduced in the 2016 National Antiretroviral (ART) Guidelines [32]. During the study period, PrEP service delivery in DREAMS in Khomas and Zambezi regions was implemented by I-TECH in collaboration with MoHSS, MoEAC, and Star for Life (in Windhoek only). To generate demand for PrEP, flyers with basic information about PrEP (e.g., what is PrEP, who is PrEP for, how do I get PrEP, how do I take PrEP) targeted to young people were designed by the DREAMS program. Flyers were disseminated at DREAMS recruitment events, HIV prevention education sessions, and health services visits. Training on the impact of HIV on AGYW, PrEP, PrEP service delivery, and adolescent-friendly service delivery was conducted with DREAMS and selected MoHSS providers to support the provision of PrEP to AGYW in DREAMS and in MoHSS facilities.
AGYW participating in DREAMS and aged 15–24 received PrEP services via three models–a “community-concierge” model, a “community-fixed” model, and a hybrid “community-clinic” model (Table 1A–1C). Definitions and frameworks for naming and describing PrEP service delivery models vary in practice and in the literature. In this paper, we conceptualize PrEP service delivery models as comprised of four building blocks–when the service is provided, where the service is provided, who is providing the service, and what services are provided [33]. We define these building blocks separately for i) initial PrEP services (education, screening, initiation) and ii) initial follow-up (0–3 months; refills, clinical monitoring, and support). Models were named according to where services were provided.
All three models were identical across initial PrEP services–PrEP education, clinical eligibility and HIV risk screening, and PrEP initiation (Table 1A). PrEP education was provided by DREAMS peer mentors during HIV/GBV prevention education sessions and by DREAMS peer mentors, health assistants (a cadre who provides HIV testing and counseling), or nurses during DREAMS health services visits at DREAMS safe spaces. Health assistants or nurses (whichever available) conducted clinical eligibility screening including HIV testing services and HIV risk assessment in the form of client-centered counseling on HIV risk and prevention methods. PrEP was offered in line with MoHSS national guidelines to AGYW aged 15 and older, who tested HIV-negative, who screened at substantial risk for HIV infection or who considered themselves at risk, and did not have clinical contraindications [32, 34]. PrEP initiation occurred the same-day as HTS and risk screening. Initial PrEP services were provided as part of a package of SRH services which included HIV testing services (HTS), syndromic screening for sexually transmitted infections (STIs), and contraceptive counseling and service provision. For those who were clinically eligible and accepted PrEP, nurses prescribed PrEP and provided additional counseling on use, adherence, and follow-up and confirmed with AGYW their next visit date. PrEP was dispensed and venous blood for required creatinine clearance (CrCl) was drawn by the prescribing nurse. Samples were dropped off at a public health facility at the end of the day for processing.
The three models, however, differed in the delivery of initial follow-up services–refills, clinical monitoring, and support. In the “community-concierge” model (Table 1A), DREAMS nurses delivered PrEP refills and clinical monitoring at community-based locations which were individually-tailored to the AGYW’s real-time needs and preferences. Locations included fixed sites (e.g., at a DREAMS safe space, after hours at their secondary school, near their home, at a market, or in a community center), a mobile van temporarily parked outside of one of these fixed community-based sites, or within the AGYW’s home or other preferred location. DREAMS providers called AGYW prior to their anticipated refill date to coordinate a convenient visit date, time, and location. Refills occurred one month after initiation and every one to three months thereafter depending on the AGYW’s needs or preferences (e.g., to align with contraceptive method provision or accommodate travel during school holidays). Refills were delivered alongside screening and management of side effects, adherence counseling, and SRH services (e.g., contraception method provision, STI screening). One month after initiation and every three months thereafter, DREAMS nurses also conducted clinical monitoring for PrEP (HIV test and repeat laboratory tests according to guidelines). DREAMS HCPs conducted SMS or phone call reminders prior to scheduled refill and follow-up visits and after missed visits. AGYW could call or message a DREAMS nurse-managed phone with questions or concerns.
In the “community-fixed” model (Table 1B), DREAMS nurses delivered PrEP refills and clinical follow-up services to AGYW via scheduled visits at fixed community-based locations (e.g., DREAMS safe spaces). AGYW could select their preferred visit location, date, and time from a set, scheduled rotation. At each fixed site, HCP were only available during scheduled day and time (e.g., sites were not ‘drop-in’). All other aspects of service delivery for refills, clinical monitoring, and support in the community-fixed model mirrored the community-concierge model.
Lastly, in the hybrid “community-clinic” model (Table 1C), AGYW were referred to their preferred public health facility for PrEP refills and clinical monitoring which were provided by MoHSS clinic-based HCPs. NIMART (nurse-initiated management of antiretroviral therapy) nurses, typically located in the ART clinic, prescribed refills one month after initiation and every three months thereafter. Refill visits were delivered alongside screening and management of side effects and adherence counseling. PrEP was dispensed at the ART pharmacy or clinic pharmacy where no separate ART pharmacy existed. Nurses also conducted clinical monitoring for PrEP every three months. AGYW receiving services through the hybrid community-clinic model did not routinely receive reminder calls or messages, however they could call or message a DREAMS nurse-managed phone with questions or concerns.
Model choice was geographically and temporally limited as implementation was phased over the study period. However, AGYW could choose their preferred PrEP service delivery model from all currently implemented models at their PrEP initiation.
Data collection
For this study, we used two routine programmatic data sources–electronic DREAMS participant records and paper-based MoHSS PrEP client records of AGYW aged 15–24 –in which individual-level data were documented as part of routine DREAMS program implementation and PrEP service delivery. We abstracted data on socio-demographic information (age, location, relationship type), structural and behavioral HIV risk factors at DREAMS entry (dropped out of school before grade 12, food insecurity, ever-experienced GBV, current pregnancy, alcohol and/or drug use, number of sexual partners in past 12 months, age-disparate partner(s), concurrent partners, condom use, transactional sex) and details of health service uptake and delivery (PrEP education, HTS, and PrEP initiation) from DREAMS participant records. From PrEP client records, we abstracted routinely collected information on client characteristics at PrEP initiation (previous PrEP use, any documented HIV risk factors) and service provision data from follow-up PrEP visits (scheduled and actual visit date, PrEP prescription). Individual-level data were linked across sources using the unique DREAMS participant ID.
Statistical analysis
We evaluated coverage of PrEP services using a cascade reflective of five steps of PrEP delivery in this setting [22]. Specifically, we described proportions of AGYW 15–24 at each step in the cascade with the denominator starting as the number of AGYW participating in DREAMS and each next step using the numerator of the previous step as its denominator. We excluded AGYW already on PrEP at entry into DREAMS (n = 32). Cascade steps included: 1) received PrEP education; 2) tested for HIV; 3) tested HIV-negative; 4) initiated PrEP; and 5) persisted with PrEP at month one, defined as receiving PrEP refill prescription 15–44 days after PrEP initiation (with the denominator restricted to those who initiated PrEP 45 days or more before the end of the study period). Cascade step five was further stratified by delivery model with differences between models tested using a chi-squared test.
We separately explored factors associated with PrEP uptake and early PrEP persistence among AGYW aged 15–24 as our primary outcomes of interest. ‘PrEP uptake’ was defined as the proportion of AGYW who initiated PrEP among those who received PrEP education and tested HIV-negative. ‘Early PrEP persistence’ was defined as the proportion of AGYW who persisted with PrEP at month one among those expected for a month one visit during the study period. In univariable models, we explored each potential correlate of PrEP uptake and early persistence. We then used backwards elimination to select factors for inclusion in each multivariable model by eliminating variables by highest p-value until a threshold of p < 0.2 was met. We used generalized estimating equations with log links and Poisson distributions to estimate prevalence ratios (PR) while accounting for clustering by DREAMS entry group. Analyses were conducted using RStudio (version 4.0.5) and the geepack package [35–37].
Ethical review
This project was reviewed and approved by ethics committees at the U.S. Centers for Disease Control and Prevention, the Namibian Ministry of Health and Social Services, and the University of Washington. Prior to participation in DREAMS, AGYW provided written informed consent and written parental consent was obtained from parents of AGYW aged 9–19 years. A waiver of consent was obtained for the review of DREAMS participant records and PrEP client records, which were fully anonymized before being accessed by the study team.
Results
Service coverage along a PrEP cascade
Between December 1, 2018 and September 30, 2019, 7593 AGYW aged 15–24 enrolled in DREAMS HIV prevention programming and were not already using PrEP at enrollment. The median follow-up time for participants was 142 days (min = 0, max = 301). Of enrolled AGYW, 7516 (99.0%) received PrEP education and 6105 (81.2%) received HIV testing services with 6035 (98.9%) testing HIV-negative (Fig 1). Among AGYW who tested HIV-negative, 2225 (36.9%) initiated PrEP. Of the 2047 AGYW expected for a month one visit within the study period, 254 (12.4%) persisted with PrEP. Of the 328 AGYW who did not persist with PrEP one month after initiation and had at least six months post-initiation time within the study period, 24 (7.3%) restarted PrEP. Three (0.1%) AGYW who initiated PrEP during the study period later tested positive for HIV infection; two clients received an HIV-positive test result at their month-one follow-up visit and one client received an HIV-positive test result at a follow-up visit four months post-initiation after having received 90-days PrEP prescription at their month-one visit. We found significant differences in early PrEP persistence by service delivery model, with 39.2% of AGYW persisting on PrEP at one-month post-initiation in the community-concierge, 15.4% in the community-fixed, and 4.5% in the hybrid community-clinic model (Fig 2A–2C, p<0.001).
a-c. Early PrEP persistence cascade, by service delivery model.
Factors associated with PrEP uptake
Overall, 5742 AGYW were eligible for PrEP, defined as those who received PrEP education and tested HIV-negative, and were included in our analysis of factors associated with PrEP uptake. We excluded AGYW with missing data for all structural, biological, and behavioral HIV risk factors assessed at DREAMS entry (n = 293); missing all data were not associated with PrEP uptake (p = 0.81). As shown in Table 2, most AGYW were aged 15–19 years (n = 3374; 58.8%), located in Khomas Region (n = 4305; 75.0%), and classified at higher risk for HIV infection and eligible for secondary DREAMS interventions (n = 4653; 81.0%). Among AGYW, 24.7% (n = 1396) reported having had dropped out of school before grade 12, 32.6% (n = 1850) reported food insecurity, and 26.2% (n = 1499) reported ever experiencing any type of GBV. Intermittent or no condom use was the most commonly reported behavioral risk factor (n = 2316; 58.9%). PrEP uptake among AGYW who were not missing all data on structural, biological, and behavioral HIV risk factors assessed at DREAMS entry and who tested HIV-negative (n = 5742) was 36.8% (n = 2115).
Multivariable analysis revealed several socio-demographic, structural, and behavioral risk factors associated with PrEP uptake (Table 2). PrEP uptake was more likely among AGYW in Zambezi (aPR: 1.31; 95% CI: 1.10–1.56) as well as among those who reported dropping out of school before grade 12 (aPR: 1.15; 95% CI: 1.04–1.28), food insecurity (aPR: 1.15; 95% CI: 1.06–1.24), intermittent or no condom use (aPR: 1.22; 95% CI: 1.13–1.33), having two or more partners in the past 12 months (aPR: 1.14; 95% CI: 1.04–1.24), ever having concurrent partners (aPR: 1.13; 95% CI: 1.02–1.24), or ever having transactional sex (aPR: 1.22; 95% CI: 1.07–1.39). PrEP uptake was less likely among AGYW who reported no current partner (aPR: 0.81; 95% CI: 0.71–0.91).
Factors associated with early PrEP persistence
Overall, 1953 AGYW were expected for their month one visit by the end of the study period and included in our analysis of factors associated with early PrEP persistence. We excluded AGYW who were missing data for all structural, biological, and behavioral HIV risk factors assessed at DREAMS entry (n = 94); missing all data were not associated with early PrEP persistence (p = 0.45). As shown in Table 3, 47% of AGYW (n = 917) were aged 15–19 years. Very few AGYW (n = 53; 2.7%) delayed starting PrEP, defined as starting PrEP at the second or later health service visit. Among AGYW, 53.7% (n = 1049) received PrEP services via the hybrid community-clinic model, 34.3% (n = 669) via the community-fixed model, and 12.0% (n = 235) via the community-concierge model. At PrEP initiation, 4.4% of AGYW (n = 86) reported having an HIV-positive partner, whereas 44.2% (n = 864) reported their partner’s HIV status as unknown. Overall, 33.2% of AGYW (n = 648) considered themselves at risk, but did not disclose any specific risk factors during risk screening. Early PrEP persistence among AGYW who were not missing all data on structural, biological, and behavioral HIV risk factors assessed at DREAMS entry and who were expected for their month one visit (n = 1953) was 12.3% (n = 240).
Multivariable analysis found early PrEP persistence to be 8.7 times more likely among AGYW receiving services in the community-concierge model (aPR: 8.70; 95% CI: 5.44–13.9) and 2.97 times more likely in the community-fixed model (aPR: 2.97; 95% CI: 1.80–4.90) compared to the hybrid community-clinic model (Table 3). In addition, AGYW who delayed PrEP start were 2.89 times more likely to persist one-month after initiation (aPR: 2.89; 95% CI: 1.52–5.46) than those who started on their first health services visit. Early persistence was also more likely among AGYW aged 20–24 or who reported only having casual partners (compared to a main partner). Having a partner with unknown HIV status was the only factor negatively associated with early persistence.
Discussion
We examined PrEP-related service coverage and evaluated PrEP uptake and persistence among AGYW participating in DREAMS and receiving PrEP via three different service delivery models in two regions in Namibia. PrEP services delivered within DREAMS community-based HIV prevention programming achieved moderate PrEP uptake. PrEP persistence one month after initiation was low overall, however differences by service delivery model were significant, with rates of early persistence substantially and significantly higher among AGYW receiving services in the community-concierge model and the community-fixed model compared with the hybrid community-clinic model.
In all three models, initial PrEP services (e.g., education, screening, initiation) were delivered identically, integrated within community-based HIV prevention programming for AGYW. Our results suggest this approach may support high coverage of initial PrEP services and elicit higher rates of PrEP uptake among AGYW than have been observed through other delivery approaches in SSA. In Kenya, PrEP uptake was only 4% among AGYW receiving services in family planning clinics and 21.7% among women and girls aged 15 and older receiving services in maternal and child health clinics [25, 38]. In the SEARCH trial, 19% of AGYW who tested HIV-negative and were assessed to be at elevated risk of HIV initiated PrEP at community-based health fairs and facilities in rural communities in Kenya and Uganda [39]. Factors that may have contributed to the higher uptake observed in our study include the targeting of AGYW at higher risk for HIV for participation in DREAMS, the DREAMS peer/near-peer mentoring and ‘safe space’ approach where AGYW can build supportive social networks, the use of adolescent-friendly service approaches and service providers, and the concurrent participation in HIV/GBV prevention education, which may build awareness and knowledge of one’s HIV risk and prevention options [40].
The early drop-off rates in our study are similar to those found elsewhere in SSA. Several studies have shown low rates of early PrEP persistence (31–38%) among AGYW receiving services through clinic, mobile clinic, and community-based models in Kenya and South Africa [22, 25–27]. In Namibia, the nascency of the PrEP program and its data systems may have contributed to the low persistence seen across all three models. No clinical trials, open label studies, demonstration projects, or awareness or demand creation campaigns were conducted prior to or concurrent with this study in Namibia. Thus, PrEP awareness was very low among AGYW as well as within their communities, contributing to social and internalized stigma surrounding PrEP use that may have negatively impacted persistence [41]. Data from the United States has shown annual increases in the proportions of clients refilling PrEP prescriptions from PrEP introduction in 2012 through 2017, suggesting a link between greater widespread community awareness and acceptance of PrEP and increases in persistence over time which may be generalizable to other contexts and populations [42, 43]. In addition, the lack of a data system that would enable linkage of client PrEP records across service delivery locations meant that AGYW who returned for PrEP refill to any public or community-based clinic outside of this study would not have been captured.
More than 80% of AGYW in the study were classified as higher risk for HIV and eligible for secondary DREAMS interventions, yet only 37% initiated PrEP. After controlling for behavioral risk factors, AGYW who had dropped out of school before grade 12, were food insecure, or living in a region with high HIV prevalence and incidence among girls and women (Zambezi) were more likely to initiate PrEP, which suggests AGYW may perceive their HIV risk or make PrEP use decisions based not only on their own and their partners’ sexual behaviors but also based on their individual socio-economic circumstances and their awareness of local area-level prevalence or incidence [28, 44, 45]. The inclusion of information about structural HIV risk and local HIV prevalence or incidence in PrEP decision support tools may promote PrEP uptake by helping AGYW recognize broader factors that contribute to their HIV risk as well as supporting more informed prevention choices [46, 47].
Research has shown that in the first months after initiation AGYW may be in a period of adjustment, considering if PrEP is really right for them [10]. Examining PrEP persistence in this early period and in a real-world PrEP delivery context is particularly important to gain an understanding of who is truly interested in and able to continue PrEP after a period of initial experimentation. In our study, we found early PrEP persistence was highly correlated with service delivery model with AGYW receiving services through the community-concierge model 8.70 times more likely to persist and the community-fixed model 2.97 times more likely to persist compared to the hybrid community-clinic model. This finding indicates that PrEP service delivery responsive to AGYWs’ real-time needs and constraints can support higher rates of persistence. Both models featured provider continuity, where the same DREAMS adolescent-friendly providers delivered all PrEP services. Such continuity, particularly where trusting relationships are formed with adolescent-friendly providers, may facilitate PrEP persistence among AGYW [10, 48, 49]. Both models also featured flexibility in service location, where an individual initiates PrEP in one location and receives refills in a different location. This flexibility in service delivery, when client-led and client-centered, supports higher rates of persistence [49, 50]. Where differences in service location are provider- or model-led, as to some extent in the community-fixed model and completely in the hybrid community-clinic model, it may introduce barriers to persistence, as was also experienced by AGYW receiving PrEP services through a mobile clinic in South Africa [48]. Interventions such as tailored counseling or peer navigation may be needed to support AGYW as they transition between community- and clinic-based PrEP service locations over time, especially given their high mobility and in order to bring PrEP for AGYW to scale [10, 40, 48, 51].
Early PrEP persistence was also more likely among AGYW who delayed starting PrEP until their second or later health services visit. Despite being a user-controlled method, qualitative research has highlighted that AGYW may delay starting PrEP in order to discuss and obtain support for PrEP use from their peers, parents or partners and that positive disclosure experiences and outcomes can be leveraged to support continued PrEP use whereas negative experiences and outcomes can lead to PrEP discontinuation [10, 52, 53]. Integration of disclosure support within PrEP counseling for those who would like to discuss their PrEP use or intended use with a peer, family member, caregiver, or partner could help AGYW obtain the emotional and financial support needed to persist with PrEP [52, 54, 55].
The strength of our study includes the use of routine programmatic data to evaluate PrEP-related service coverage, measure key PrEP outcomes, and identify factors associated with these outcomes among AGYW receiving services through community and hybrid community-clinic delivery models in real-world programmatic settings. In addition, this study is one of few to examine associations between structural HIV risk and service delivery related factors and PrEP outcomes [21]. Our study, however, has some limitations. Broadly, programmatic data are often underused for research due to concerns of incompleteness and inaccuracy, though we applied several strategies to improve PrEP data quality during program implementation [56]. Model selection was likely biased by geographic location, age, and school status as the service delivery models available differed by region and delivery preferences among AGYW varied by age and school status. However, as a real-world implementation project these results may simulate how these PrEP delivery options might be offered and selected in resource-limited settings. Regarding early persistence, we relied on PrEP prescription data, as PrEP dispensing data were not available, and assumed that all AGYW prescribed PrEP were dispensed PrEP in-room or picked up their refill from the pharmacy. Our ability to make comparisons and conclusions regarding the performance of service delivery models across studies is limited given variability in the reporting of PrEP cascade components, lack of model specificity, differences in service model maturity, and inconsistent definitions and measures of uptake and persistence across studies [7, 42, 57]. Structural and behavioral risk factors included in this study were self-reported at the time of DREAMS entry, thus, underreporting may have occurred if AGYW were reluctant to share this sensitive and private information or factors may have been misclassified if they changed over time.
Conclusion
PrEP is an important HIV prevention strategy for AGYW and there is an urgent need to identify and scale more person-centred and responsive PrEP service delivery models and strategies that will meet the needs and preferences of AGYW more effectively and comprehensively. Clinic-based delivery models, however, often present barriers for AGYW in accessing and using PrEP. Our study identifies several opportunities to promote PrEP uptake and persistence through community-based and hybrid PrEP service delivery models. First, programs could consider integrating PrEP services within community-based HIV prevention programming for AGYW to increase their awareness of and access to PrEP services. Second, given the mobility and unpredictability AGYW face in their daily lives, programs could consider community-based service delivery models that employ individualized approaches to the delivery of PrEP refills and follow-up services responsive to client’s real-time needs and constraints as was done in the community-concierge model here. Third, inclusion of information on structural HIV risk as well as area-level HIV prevalence or incidence, in addition to behavioral HIV risk, in PrEP decision support tools could be used to generate a broader understanding of clients’ vulnerability to HIV infection and promote uptake of PrEP or other prevention options among AGYW. Lastly, recognizing that models combining community-based, clinic-based, and other innovative delivery modalities (e.g., pharmacy, telehealth) may be needed to provide PrEP services at scale, programs could consider strategies such as tailored counseling or navigation support to prepare and support clients to receive services across different providers and service locations promoting PrEP persistence as users access PrEP via different service delivery modalities over time.
Acknowledgments
We would like to thank the Namibia Ministry of Health and Social Services, the Ministry of Education, Arts, and Culture, the Khomas and Zambezi Regional Health and Education Directorates, respectively, and Star for Life in Namibia for their support to the implementation of community-based PrEP service delivery within the I-TECH DREAMS program. Community-based PrEP service delivery would not have been possible without the tremendous dedication and skills of the I-TECH and Star for Life DREAMS healthcare providers and program staff, with special thanks to Irine Birir, Laura Brandt, Rachel Halweendo, Rosanne Kahuure, Petrina Kamati, Hilma Kambundu, Theopolina Kueyo, Lingiwe Mlanga, Thelma Mukwaso, Charity Muwana, Josua Nghipangelwa, Eucabeth Onyikwa, Martha Shilomboleni, Kristine Shipena, Pelyn Shoshore, Rochester Sibuku, Fortune Sisamu, and Sharon Zambwe.
References
- 1.
Confronting Inequalities. Lessons for pandemic responses from 40 years of AIDS. UNAIDS Global AIDS Update 2021. Geneva, Switzerland: Joint United Nations Programme on HIV/AIDS, 2021.
- 2. Baeten JM, Donnell D, Ndase P, Mugo NR, Campbell JD, Wangisi J, et al. Antiretroviral prophylaxis for HIV prevention in heterosexual men and women. N Engl J Med. 2012;367(5):399–410. Epub 2012/07/13. pmid:22784037; PubMed Central PMCID: PMC3770474.
- 3. Fonner VA, Dalglish SL, Kennedy CE, Baggaley R, O’Reilly KR, Koechlin FM, et al. Effectiveness and safety of oral HIV preexposure prophylaxis for all populations. AIDS. 2016;30(12):1973–83. Epub 2016/05/06. pmid:27149090; PubMed Central PMCID: PMC4949005.
- 4. Celum C, Baeten J. PrEP for HIV Prevention: Evidence, Global Scale-up, and Emerging Options. Cell Host Microbe. 2020;27(4):502–6. Epub 2020/04/10. pmid:32272075.
- 5. Global PrEP Tracker [Internet]. AVAC. 2021 [cited August 11, 2021]. Available from: https://www.prepwatch.org/resource/global-prep-tracker/.
- 6. Saul J, Bachman G, Allen S, Toiv NF, Cooney C, Beamon T. The DREAMS core package of interventions: A comprehensive approach to preventing HIV among adolescent girls and young women. PLoS One. 2018;13(12):e0208167. Epub 2018/12/12. pmid:30532210; PubMed Central PMCID: PMC6285267.
- 7. Dunbar MS, Kripke K, Haberer J, Castor D, Dalal S, Mukoma W, et al. Understanding and measuring uptake and coverage of oral pre-exposure prophylaxis delivery among adolescent girls and young women in sub-Saharan Africa. Sex Health. 2018;15(6):513–21. Epub 2018/11/09. pmid:30408431; PubMed Central PMCID: PMC6429961.
- 8. Camlin CS, Koss CA, Getahun M, Owino L, Itiakorit H, Akatukwasa C, et al. Understanding Demand for PrEP and Early Experiences of PrEP Use Among Young Adults in Rural Kenya and Uganda: A Qualitative Study. AIDS Behav. 2020;24(7):2149–62. pmid:31955361; PubMed Central PMCID: PMC7909847.
- 9. Ongolly FK, Dolla A, Ngure K, Irungu EM, Odoyo J, Wamoni E, et al. "I Just Decided to Stop:" Understanding PrEP Discontinuation Among Individuals Initiating PrEP in HIV Care Centers in Kenya. J Acquir Immune Defic Syndr. 2021;87(1):e150–e8. pmid:33492024; PubMed Central PMCID: PMC8026512.
- 10. Rousseau E, Katz AWK, O’Rourke S, Bekker LG, Delany-Moretlwe S, Bukusi E, et al. Adolescent girls and young women’s PrEP-user journey during an implementation science study in South Africa and Kenya. PLoS One. 2021;16(10):e0258542. Epub 20211014. pmid:34648589; PubMed Central PMCID: PMC8516266.
- 11.
Pilgrim N, Mathur S, Gottert A, Rutenberg N, Pulerwitz J. Building evidence to guide PrEP introduction for adolescent girls and young women. Washington, DC: Population Council; 2016.
- 12. O’Malley G, Beima-Sofie KM, Roche SD, Rousseau E, Travill D, Omollo V, et al. Health Care Providers as Agents of Change: Integrating PrEP With Other Sexual and Reproductive Health Services for Adolescent Girls and Young Women. Frontiers in Reproductive Health. 2021;3. pmid:36303982
- 13.
Adolescent friendly health services for adolescents living with HIV: from theory to practice. Technical Brief. Geneva, Switzerland: World Health Organization; 2019 (WHO/CDS/HIV/19.39). Licence: CC BY-NC-SA 3.0 IGO.
- 14. Celum CL, Delany-Moretlwe S, McConnell M, van Rooyen H, Bekker LG, Kurth A, et al. Rethinking HIV prevention to prepare for oral PrEP implementation for young African women. J Int AIDS Soc. 2015;18(4 Suppl 3):20227. Epub 2015/07/23. pmid:26198350; PubMed Central PMCID: PMC4509892.
- 15. Irungu EM, Baeten JM. PrEP rollout in Africa: status and opportunity. Nat Med. 2020;26(5):655–64. Epub 2020/05/15. pmid:32405065.
- 16. Pettifor A, Stoner M, Pike C, Bekker LG. Adolescent lives matter: preventing HIV in adolescents. Curr Opin HIV AIDS. 2018;13(3):265–73. pmid:29528850; PubMed Central PMCID: PMC5902132.
- 17. Ramraj T, Chirinda W, Jonas K, Govindasamy D, Jama N, McClinton Appollis T, et al. Service delivery models that promote linkages to PrEP for adolescent girls and young women and men in sub-Saharan Africa: a scoping review. BMJ Open. 2023;13(3):e061503. Epub 20230327. pmid:36972966; PubMed Central PMCID: PMC10069497.
- 18. Labhardt ND, Ringera I, Lejone TI, Klimkait T, Muhairwe J, Amstutz A, et al. Effect of Offering Same-Day ART vs Usual Health Facility Referral During Home-Based HIV Testing on Linkage to Care and Viral Suppression Among Adults With HIV in Lesotho: The CASCADE Randomized Clinical Trial. JAMA. 2018;319(11):1103–12. pmid:29509839; PubMed Central PMCID: PMC5885884.
- 19. Fatti G, Jackson D, Goga AE, Shaikh N, Eley B, Nachega JB, et al. The effectiveness and cost-effectiveness of community-based support for adolescents receiving antiretroviral treatment: an operational research study in South Africa. J Int AIDS Soc. 2018;21 Suppl 1. pmid:29485714; PubMed Central PMCID: PMC5978711.
- 20. Barnabas RV, Szpiro AA, van Rooyen H, Asiimwe S, Pillay D, Ware NC, et al. Community-based antiretroviral therapy versus standard clinic-based services for HIV in South Africa and Uganda (DO ART): a randomised trial. The Lancet Global Health. 2020;8(10):e1305–e15. pmid:32971053
- 21. de Dieu Tapsoba J, Zangeneh SZ, Appelmans E, Pasalar S, Mori K, Peng L, et al. Persistence of oral pre-exposure prophylaxis (PrEP) among adolescent girls and young women initiating PrEP for HIV prevention in Kenya. AIDS Care. 2021;33(6):712–20. Epub 2020/09/22. pmid:32951437; PubMed Central PMCID: PMC7981281.
- 22. Were D, Musau A, Mutegi J, Ongwen P, Manguro G, Kamau M, et al. Using a HIV prevention cascade for identifying missed opportunities in PrEP delivery in Kenya: results from a programmatic surveillance study. J Int AIDS Soc. 2020;23 Suppl 3:e25537. Epub 2020/07/01. pmid:32602658; PubMed Central PMCID: PMC7325512.
- 23. Claassen CW, Mumba D, Njelesani M, Nyimbili D, Mwango LK, Mwitumwa M, et al. Initial implementation of PrEP in Zambia: health policy development and service delivery scale-up. BMJ Open. 2021;11(7):e047017. Epub 2021/07/11. pmid:34244265; PubMed Central PMCID: PMC8273462.
- 24. Celum CL, Delany-Moretlwe S, Baeten JM, van der Straten A, Hosek S, Bukusi EA, et al. HIV pre-exposure prophylaxis for adolescent girls and young women in Africa: from efficacy trials to delivery. J Int AIDS Soc. 2019;22 Suppl 4:e25298. Epub 2019/07/23. pmid:31328444; PubMed Central PMCID: PMC6643076.
- 25. Kinuthia J, Pintye J, Abuna F, Mugwanya KK, Lagat H, Onyango D, et al. Pre-exposure prophylaxis uptake and early continuation among pregnant and post-partum women within maternal and child health clinics in Kenya: results from an implementation programme. The Lancet HIV. 2020;7(1):e38–e48. pmid:31813837
- 26. Mugwanya KK, Pintye J, Kinuthia J, Lagat H, Abuna F, Begnel ER, et al. Persistence with PrEP use in African adolescents and young women initiating PrEP. CROI, Conference on Retroviruses and Opportunistic Infections; Seattle, Washington, USA 2019.
- 27. Celum CL, Bukusi EA, Bekker LG, Delany-Moretlwe S, Kidoguchi L, Omollo V, et al. PrEP use and HIV seroconversion rates in adolescent girls and young women from Kenya and South Africa: the POWER demonstration project. J Int AIDS Soc. 2022;25(7):e25962. pmid:35822945; PubMed Central PMCID: PMC9278271.
- 28.
IN DANGER: UNAIDS Global AIDS Update 2022. Geneva: Joint United Nations Programme on HIV/AIDS; 2022. License: CC BY-NC-SA 3.0 IGO.
- 29. MacLachlan EW, Korn AK, Ensminger AL, Zambwe S, Kueyo T, Kahuure R, et al. Bottlenecks and Solutions During Implementation of the DREAMS Program for Adolescent Girls and Young Women in Namibia. Glob Health Sci Pract. 2022;10(5). Epub 20221031. pmid:36316146; PubMed Central PMCID: PMC9622280.
- 30.
Baldwin W. "Creating ’safe spaces’ for adolescent girls," Promoting Healthy, Safe, and Productive Transitions to Adulthood Brief no. 39. New York: Population Council, 2011.
- 31. Austrian K, Ghati D. Girl Centered Program Design: A Toolkit to Develop, Strengthen and Expand Adolescent Girl Programs. Population Council; 2010.
- 32.
National Guidelines for Antiretroviral Therapy, Namibia. Fifth ed. Windhoek, Namibia: Directorate of Special Programmes, Ministry of Health and Social Services; 2016.
- 33. Duncombe C, Rosenblum S, Hellmann N, Holmes C, Wilkinson L, Biot M, et al. Reframing HIV care: putting people at the centre of antiretroviral delivery. Trop Med Int Health. 2015;20(4):430–47. Epub 20150216. pmid:25583302; PubMed Central PMCID: PMC4670701.
- 34.
National Guidelines for Antiretroviral Therapy, Namibia. Sixth ed. Windhoek, Namibia: Directorate of Special Programmes, Ministry of Health and Social Services; 2019.
- 35. Halekoh U, Højsgaard S, Yan J. The R Package geepack for Generalized Estimating Equations. Journal of Statistical Software. 2006;15/2:1–11.
- 36. Yan J. Yet Another Package for Generalized Estimating Equations. R-News. 2002;2/3:12–4.
- 37. Yan J, Fine JP. Estimating Equations for Association Structures. Statistics in Medicine. 2004;23:859–80. pmid:15027075
- 38. Sila J, Larsen AM, Kinuthia J, Owiti G, Abuna F, Kohler PK, et al. High Awareness, Yet Low Uptake, of Pre-Exposure Prophylaxis Among Adolescent Girls and Young Women Within Family Planning Clinics in Kenya. AIDS Patient Care STDS. 2020;34(8):336–43. Epub 2020/08/08. pmid:32757980; PubMed Central PMCID: PMC7415219.
- 39. Koss CA, Charlebois ED, Ayieko J, Kwarisiima D, Kabami J, Balzer LB, et al. Uptake, engagement, and adherence to pre-exposure prophylaxis offered after population HIV testing in rural Kenya and Uganda: 72-week interim analysis of observational data from the SEARCH study. The Lancet HIV. 2020;7(4):e249–e61. pmid:32087152
- 40. Jackson-Gibson M, Ezema AU, Orero W, Were I, Ohiomoba RO, Mbullo PO, et al. Facilitators and barriers to HIV pre-exposure prophylaxis (PrEP) uptake through a community-based intervention strategy among adolescent girls and young women in Seme Sub-County, Kisumu, Kenya. BMC Public Health. 2021;21(1):1284. Epub 20210701. pmid:34210288; PubMed Central PMCID: PMC8252310.
- 41. Velloza J, Khoza N, Scorgie F, Chitukuta M, Mutero P, Mutiti K, et al. The influence of HIV-related stigma on PrEP disclosure and adherence among adolescent girls and young women in HPTN 082: a qualitative study. J Int AIDS Soc. 2020;23(3):e25463. pmid:32144874; PubMed Central PMCID: PMC7060297.
- 42. Pyra MN, Haberer JE, Hasen N, Reed J, Mugo NR, Baeten JM. Global implementation of PrEP for HIV prevention: setting expectations for impact. J Int AIDS Soc. 2019;22(8):e25370. Epub 2019/08/29. pmid:31456348; PubMed Central PMCID: PMC6712462.
- 43. Delaney KP, Sanchez T, Bowles K, Oraka E, DiNenno E, Sullivan P. Awareness and Use of PrEP Appear to Be Increasing Among Internet Samples of US MSM. CROI, Conference on Retroviruses and Opportunistic Infections; Boston, Massachusetts, USA 2016.
- 44. Ransome Y, Bogart LM, Kawachi I, Kaplan A, Mayer KH, Ojikutu B. Area-level HIV risk and socioeconomic factors associated with willingness to use PrEP among Black people in the U.S. South. Ann Epidemiol. 2020;42:33–41. Epub 2020/01/04. pmid:31899083; PubMed Central PMCID: PMC7056502.
- 45.
Ministry of Health and Social Services (MoHSS) Namibia. Namibia Population-based HIV Impact Assessment (NAMPHIA) 2017: Final Report. Windhoek, Namibia: MoHSS; 2019.
- 46. Sewell WC, Solleveld P, Seidman D, Dehlendorf C, Marcus JL, Krakower DS. Patient-Led Decision-Making for HIV Preexposure Prophylaxis. Curr HIV/AIDS Rep. 2021;18(1):48–56. Epub 20210108. pmid:33417201; PubMed Central PMCID: PMC8086908.
- 47. Seidman D, Travill D, Celum C, Dehlendorf C, Wilson W, Ndlovu M, et al. MyPrEP Decision Support Tool Increases PrEP Persistence in Adolescent Girls and Young Women Attending an Urban Primary Health Care Clinic in South Africa. HIV Research for Prevention (HIV R4P); Virtual 2021.
- 48. Rousseau E, Bekker LG, Julies RF, Celum C, Morton J, Johnson R, et al. A community-based mobile clinic model delivering PrEP for HIV prevention to adolescent girls and young women in Cape Town, South Africa. BMC Health Serv Res. 2021;21(1):888. Epub 20210828. pmid:34454505; PubMed Central PMCID: PMC8403426.
- 49. Barnabee G O’Bryan G, Ndeikemona L, Billah I, Silas L, Morgan KL, et al. Improving HIV pre-exposure prophylaxis persistence among adolescent girls and young women: Insights from a mixed-methods evaluation of community, hybrid, and facility service delivery models in Namibia. Front Reprod Health. 2022;4:1048702. Epub 20221205. pmid:36545490; PubMed Central PMCID: PMC9760915.
- 50. Gilbert HN, Wyatt MA, Pisarski EE, Asiimwe S, van Rooyen H, Seeley J, et al. How community ART delivery may improve HIV treatment outcomes: Qualitative inquiry into mechanisms of effect in a randomized trial of community-based ART initiation, monitoring and re-supply (DO ART) in South Africa and Uganda. J Int AIDS Soc. 2021;24(10):e25821. pmid:34624173; PubMed Central PMCID: PMC8500674.
- 51. Roche SD, Odoyo J, Irungu E, Kwach B, Dollah A, Nyerere B, et al. A one-stop shop model for improved efficiency of pre-exposure prophylaxis delivery in public clinics in western Kenya: a mixed methods implementation science study. J Int AIDS Soc. 2021;24(12):e25845. pmid:34898032; PubMed Central PMCID: PMC8666585.
- 52. Giovenco D, Gill K, Fynn L, Duyver M, O’Rourke S, van der Straten A, et al. Experiences of oral pre-exposure prophylaxis (PrEP) use disclosure among South African adolescent girls and young women and its perceived impact on adherence. PLoS One. 2021;16(3):e0248307. Epub 20210305. pmid:33667264; PubMed Central PMCID: PMC7935254.
- 53. Pintye J O’Malley G, Kinuthia J, Abuna F, Escudero JN, Mugambi M, et al. Influences on Early Discontinuation and Persistence of Daily Oral PrEP Use Among Kenyan Adolescent Girls and Young Women: A Qualitative Evaluation From a PrEP Implementation Program. J Acquir Immune Defic Syndr. 2021;86(4):e83–e9. Epub 2020/12/05. pmid:33273211.
- 54. Amico KR, Wallace M, Bekker LG, Roux S, Atujuna M, Sebastian E, et al. Experiences with HPTN 067/ADAPT Study-Provided Open-Label PrEP Among Women in Cape Town: Facilitators and Barriers Within a Mutuality Framework. AIDS Behav. 2017;21(5):1361–75. Epub 2016/06/19. pmid:27317411; PubMed Central PMCID: PMC5378745.
- 55. Golub SA. PrEP Stigma: Implicit and Explicit Drivers of Disparity. Curr HIV/AIDS Rep. 2018;15(2):190–7. pmid:29460223; PubMed Central PMCID: PMC5884731.
- 56. Hung YW, Hoxha K, Irwin BR, Law MR, Grepin KA. Using routine health information data for research in low- and middle-income countries: a systematic review. BMC Health Serv Res. 2020;20(1):790. Epub 20200825. pmid:32843033; PubMed Central PMCID: PMC7446185.
- 57. Stankevitz K, Grant H, Lloyd J, Gomez GB, Kripke K, Torjesen K, et al. Oral preexposure prophylaxis continuation, measurement and reporting. AIDS. 2020;34(12):1801–11. Epub 2020/06/20. pmid:32558660.