User-informed tailoring of electronic and in-person peer navigation strategies to improve retention in HIV care for adolescents and young adults: a qualitative inquiry

Abstract

Poor adherence and retention in HIV care remain a major challenge among adolescents and young adults (AYA) living with HIV in sub-Saharan Africa (SSA). Strategies are urgently required to support AYA to remain in care for better health outcomes. We explored AYA preferences regarding the format and delivery of electronic and in-person peer navigation to improve HIV care outcomes. This formative qualitative study was conducted between September and November 2020 among AYA aged 14–24 years enrolled in HIV care at three clinics within Kisumu County, Kenya. We conducted two focus group discussions (FGDs) each with 8–9 participants (n=17) purposively selected based on age, gender, and clinic where they received care. FGD guides addressed preferences of format, content, timing, delivery, and frequency of electronic and in-person peer navigation. Three key themes describe the nature of AYA preferences for electronic and in-person navigation. The characteristics desired of a navigator are a person of the same age group and HIV status who has a good memory, and patience, is encouraging, friendly, and able to maintain confidentiality. AYA want the content of their interaction with the navigator to center on sharing motivational messages and also educating them on matters of HIV care, sexual and reproductive health, and mental health. The preferred navigation formats for electronic communication are platforms considered confidential such as WhatsApp, phone calls, and text messaging. AYA preferred interventions delivered through secure communication platforms by navigators with whom they have commonalities and possess characteristics that promote relationship building. The navigation interventions that prioritize confidentiality and holistic content will likely be most highly valued by AYA. Further, electronic mechanisms can help support the relationship building that is at the core of our navigation approach as well as a fundamental aspect of social work in general.

Introduction

The HIV epidemic among adolescents and young adults (AYA) is a major public health challenge in sub-Saharan Africa (SSA). The region accounts for about 1.5 million (88%) of the 1.7 million AYA estimated to be living with HIV globally (EGPAF, 2020; UNAIDS, n.d.). In Kenya, nearly a third of all new HIV infections occur among AYA (Kenya National Bureau of Statistics et al., 2015) who subsequently have lower rates of viral suppression and retention in care resulting in high mortality (Ammon et al., 2018; Enane, Davies, et al., 2018; Enane, Vreeman, et al., 2018; Ryscavage et al., 2011). Changing the negative health trajectory of AYA requires interventions that are tailored to AYA’s preferences to improve retention and viral suppression (Lanyon et al., 2020; Yehia et al., 2015).

Poor retention in care and nonadherence to antiretroviral treatment (ART) among AYA living with HIV emanate from numerous factors including stigma, non-disclosure, peer pressure, lack transport to health facilities, adolescent-to-adult care transition challenges, forgetfulness, drug-side effects, health provider attitudes and absence of adequate psychosocial support (Ammon et al., 2018; Carbone et al., 2019; Warner et al., 2018). Peer support and use of mHealth strategies including text messaging are areas of intervention that may address issues of stigma, as well as AYA retention and adherence challenges (Boyes et al., 2018a; Fairbanks et al., 2018; Hacking et al., 2019; Okeke et al., 2014). Peer support or navigation has been demonstrated to improve treatment outcomes among adults living with HIV and demonstrated potential for increased linkage, retention, and psychosocial well-being of AYA living with HIV (Boyes et al., 2018b; Casillas et al., 2019; Willis et al., 2019). This support is thought to improve treatment outcomes by providing patients with needed reminders, social and emotional support, health education, and encouragement.

Over 70% of AYA have access to a phone and about 50% to a smartphone that provide the platform to initiating mhealth (Kharono et al., 2022). While younger school-going AYAs may have challenges in accessing phones to enable mhealth, most of the older AYAs have access (Jennings Mayo-Wilson et al., 2022). mHealth strategies, including text messaging and phone outreach, have demonstrated some success in increasing adherence and clinic attendance in adult populations and in high-income countries (Goldstein et al., 2023; Lester et al., 2019). mHealth approaches that are interactive or include opportunities for bi-directional communication have, in some cases, been more effective(Chiang et al., 2018). However, these strategies’ success is contingent on confidentiality and access to electronic devices, i.e., mobile phones (Rana et al., 2015). Additionally, given global constraints on HIV funding, wide availability and use of mobile phones among AYA, and care model transitions utilizing mHealth that began in the COVID-19 pandemic, there is a need to explore the delivery of peer support in an electronic format. Furthermore, AYA are a distinct tech-savvy group with specific needs that call for tailoring of such strategies to accommodate age-specific challenges and preferences (Rana et al., 2015; Ranney et al., 2014).

Despite the potential for interventions such as in-person and electronic peer navigation, evidence on how to best tailor these to support improved health outcomes for AYA is minimal. This is partly due to assumptions that what works for the other populations also works for AYA without necessarily eliciting their input and, as such, do not adequately consider AYA preferences(WHO, 2018). Even where AYA have been considered for peer support strategies adapted from adults, their voice and input have largely been unsolicited (Mackworth-Young et al., 2022). Tailored interventions that seek AYA perspectives to understand their specific navigation needs are therefore critical for future intervention designs that seek to improve HIV care outcomes for AYA. Building on the potential of peer navigation and mHealth interventions for AYA, the Adapt for Adolescents (A4A) Study [NCT04432571]) evaluated the impact of electronic and in-person delivery of peer navigation for AYA with HIV in Kenya(Abuogi et al., 2023). This formative work sought to understand how peer navigation delivered electronically and in-person can be tailored to optimally support AYA retention and improved HIV treatment outcomes.

Methods

Design

We conducted a formative qualitative study as part of a larger NIH-funded trial seeking to determine effectiveness of sequential adaptive strategies for engaging AYA in HIV care (the Adapt for Adolescents Study) in Kisumu, Kenya. The aim of this qualitative study was to understand the preferences of AYA on peer navigation strategies to inform the design of study interventions.

Settings and Study Population

AYA were recruited from three high volume health facilities in Kisumu County. Kisumu is one of the six counties in the Nyanza region of western Kenya with an estimated population of 1,155,574 as per the 2019 national census and a an HIV prevalence of 19.3%, far higher than the national average of 4.9% (NASCOP, NACC, 2014). AYA account for a 5.9% of infections with females having more than twice the burden of their male counterparts (NACC, 2015). The county has achieved an adult HIV viral suppression prevalence of 83.2% which is short of the UNAIDS revised goal of having 95% of patients retained and virally suppressed(NASCOP, 2020).

Sample size and sampling frame

As part of the broader pre-trial formative work, we conducted two FGDs, one with eight and another with nine participants (n=17) each one targeting one of the two intervention strategies of electronic and in-person navigation. The participants were selected to be consistent with the population with whom the interventions would be tested in the main A4A trial. Participants were identified by trained Research Assistants (RA) with the help of healthcare providers at the facilities. Healthcare providers then called the participants and briefly informed them about the study. Those interested in participating were asked to call the study staff or come to the facility to get more information about the study. All the participants who contacted the study staff on phone or came to the facility to obtained additional information agreed to participant and were scheduled for FGDs accordingly. Participation in the FGDs was proceeded by obtaining written consent from AYA above the age of legal consent (>18 years of age) and AYA assent as well as consent from guardians for those below 18 years. The study received ethical approval from KEMRI Scientific and Ethics Review Unit (SSC No. 3986) and Washington University in St Louis (# 202006141).

Data Collection

Two trained Research Assistants (RAs) (a moderator and notetaker) with experience working with AYA consented and conducted the FGDs. One FGD, with participants who had access to internet and access to smart phone or computer, was conducted Zoom and another one in-person(Zoom, 2016). The moderator structured the discussions using a FGD guide which was developed in advance based the study’s need to better understand planned intervention preferences. The guide topics centered on preferred qualities of a peer navigator and main features of the electronic or in-person navigation interventions (e.g., content, frequency, electronic delivery platforms). An example of the questions from the guide are: What do you think should be the content of the automated SMS health messages and phone calls or message conversations with navigators who aim to encourage adolescents and young people to keep their clinic appointments and adhere to their medication? What characteristics/qualities should a navigator/counsellor who initiates chatting through phone calls and messages possess? What is the preferred time(s) for delivering messages or phone calls to adolescents and young adults? Why do you think these are the best times? What is the preferred frequency for delivering messages or phone calls to adolescents and young adults? Why do you think this is the best frequency? For confidentiality purposes, participants referred to other participants by the number assigned to each participant in the FGD rather than using their real names. For the FGD conducted on Zoom, the moderator started by asking participants to initially switch on cameras on their gadgets for everyone to see and know each other after thereafter switch off and only switch on when contributing to the discussion to save the bandwidth. Discussions were audio-recorded using the Zoom recording functionality or audio recorders and notetaker took detailed hand-written notes as a backup. The FGDs were largely conducted in English as preferred by participants with intermittent mix with Kiswahili. Each FGD took approximately one and a half hours and participants were each reimbursed approximately $5 USD for transport or data bundles.

Data Analysis

RAs transcribed by hand and translated (as needed) the audio files and used the finalized FGD notes to supplement data from transcripts including documentation of non-verbal cues and body language. Although Zoom has automatic transcription capabilities, we did not use this functionality because some of FGD participants expressed themselves in local language that could not be transcribed by the software. We applied a combination of deductive and inductive approaches to develop a codebook based on deductive themes in the guide and those emerging from the transcripts. The broad deductive codes included preferred qualities of a peer navigator, content, and timing of electronic and in-person navigation support and preferred electronic platforms. A Qualitative Researcher (BAO) trained in qualitative analysis manually coded the two transcripts that were then reviewed by the lead author (ZAK) before the two discussed any discrepancies that emerged such as unclear codes or those that were duplicates that needed to be redefined or collapsed. The discrepancies were resolved through discussions and consensus. In some instances, resolution involved the identification of new codes or the refinement of existing codes. Overall, there was agreement on most applications of the codes. We used content analysis with categorization to derive dominant themes through careful examination and constant comparison between paragraphs and pages within transcripts, and also between transcripts (Elo & Kyngäs, 2008; Hennink et al., 2020; Maxwell & Miller, 2008). . This careful analysis helped us gain a clear understanding of the AYA preferences for intervention strategies.

Results

Socio-demographic attributes

Out of 17 participants taking in the FGDs 53% were female and 65% were 18 years of age or older. All the participants reported to be single (never married) and only one (6%) had a child. Three quarters (71%) were still in school with 65% having been on ART for more than 10 years (Table 1).

Table 1:

Socio-demographic characteristics of AYA (n=17)

Attribute	Frequency	Percent
Gender
Male	8	47
Female	9	53
Age
<18	6	35
≥ 18	11	65
Education level
Out of school	5	29
In school	12	71
Marital status
Married	0	0
Single	17	100
Children
Has child	1	6
Has no child	16	94
Time on ART
<10 years	6	35
≥ 10 years	11	65

From our analysis, important information on three main topics increased our understanding around AYAs preferences for electronic and in-person peer navigation. The first topic was qualities of peer navigators that touched on navigator demeanor, gender, their knowledge of AYA issues and right attitude and mood. The second was digital platforms and language of communication, and the third was frequency and content of peer navigation.

Qualities of peer navigators

Participants shared their perception of the preferred qualities of a peer navigator that included patience, encouragement, friendliness, ability to maintain confidentiality and having a good memory. There seemed to be a consensus that AYA were more comfortable with navigators within the same age range and generally of the same HIV status. However, a few AYA preferred slightly older navigators who could provide additional counseling.

...you can’t expect an old person to pass the information to us. We will not understand, yes, but if you take somebody who understands, the adolescents not all that young but he can be middle age there. let’s say for example... (name omitted) she is not too old not too young but she usually understands us and knows how to talk to us. So when passing the information, it depends with the age bracket of the informer and the informed (Electronic navigation FGD).

AYA had mixed preferences for the navigator’s gender. While some AYA preferred a gender mix where a male patient is paired with a female navigator, others did not have a strong gender preference as long as the navigator had a positive attitude. Some who preferred the same gender peer navigator argued that that would encourage open and candid discussions that may not be achieved if it were with opposite gender.

I prefer same genders... I am always kind of shy... so that is why I chose a male gender (In-person navigation FGD).

...We can receive the information well on any gender only if you are qualified in giving the right information as needed (Electronic navigation FGD).

For effective navigation, participants suggested that the navigators need to know the adolescents well, including their lives outside of the clinic or hospital, such as having knowledge about their home and school life, to offer maximum support. Most participants were comfortable with their parents and siblings knowing about the navigator since their HIV status was known to them and reported being at ease with their nuclear family members.

...a peer navigator should know all the adolescents and where they are coming from or for example let us say I have a problem and then I call the peer navigator... the peer navigator should know where I am coming from, and learning so if I have a problem she can approach me at any time, she can come to me, not only in the hospital when I go for medication (In-person navigation FGD).

Participants pointed out that mood swings, lack of commitment and irritating conversations to be some of the likely challenges anticipated with navigation. They gave an example where a navigator might ask demeaning or irritating questions to the AYA who might opt not to respond thereby compromising the goal of peer navigation.

There can be disagreement, when the peer navigator is talking to you one-on-one but you refuse to talk... what I am trying to say is that sometimes the questions might be irritative... so you may not be willing to answer some of them (In-person navigation FGD).

Digital platforms and language of communication

Participants acknowledged that indeed AYA may miss clinic appointments or medications and expressed appreciation for reminders through calls and messages. They felt that the reminders are effective in promoting adherence to medications and clinic visits which may be delivered as part of electronic navigation. Digital platforms such as WhatsApp, text messages, and phone calls were widely preferred by participants. These platforms were generally considered private and confidential with little public interference or chances of inadvertent disclosure. Other platforms considered confidential were Zoom, Microsoft teams, and WhatsApp video calls. Facebook and Instagram were considered public platforms and were not preferred due to privacy and confidentiality concerns, even through direct messaging options. However, participants mentioned that these public platforms can be effective in creating general awareness on the importance of HIV care engagement for behavior change of this population.

Okay, to me I think WhatsApp is good but on Facebook, there is somebody who is going to see you and start disclosing about you but WhatsApp is (Electronic navigation FGD).

To facilitate e-navigation for adolescents in school where phones are prohibited, participants suggested guardians/parents to receive the reminders and relay them to the adolescents or pick up medication on their behalf. This channeling of e-navigation messages and other communications to AYA while in school was thought to be more effective than waiting until school break for AYA to receive the messages. The participants observed that guardians/parents know their children well and they have ability to decipher the information that needs to be relayed immediately to encourage AYA, such as medication adherence, and those messages that can be delayed. AYA in school preferred parents picking medication on their behalf.

Two-way short and informal messages with a feedback loop were most preferred to keep track of participants’ engagement during navigation. They preferred messages that are in a language that is a mix of English and Kiswahili to cater for the needs of AYA who are not fluent in either language. They also preferred messages that serve three goals of being motivational, informational (reminders), and educational.

I think it should come as an encouragement message. And also as a reminder. And also it will be effective if you get the view of the person who has received the message, maybe a kind of feedback... (Electronic navigation FGD).

Further, the AYA wanted the language used in the messages to be HIV status neutral. They preferred the use of words that could not be linked to HIV even upon Google search. They argued that general terms such as ‘medication’ which can be for any disease such as TB, hypertension, cancer, or other agreed upon symbols or words could be used. Overall, the participants agreed that use of coded language was necessary when sending messages about their HIV care to protect their confidentiality in case their phone was with someone else they had not disclosed to. They observed that use of coded language was also helpful when they are reminding and encouraging each other while with other peers.

... So like he asked me that (name omitted) did you take your xxxx today in the morning then I asked him what? Then he asked sweets. Then my brain knocked that my sweets... my sweets then I had to agree that my ARVs are my sweets...So we had to brain wash that guy [in their company] but whatever we were talking about we knew ourselves (Electronic navigation FGD).

Frequency and content of peer navigation

For in-person navigation, school going participants preferred quarterly (after every 3 months) navigation meet-ups when on school break and after clinic visits, while participants not in school preferred monthly navigation meetings. The general preference for the meeting venue was in a private room within the hospital as it was considered more private and the content of discussions suited the health facility setting.

...it depends with whatever discussions I have for the day... that is why I was saying after every time I go for my medication [most adolescents are given a 3-month return date] so that he or she knows how I am progressing... the discussion will depend on that and what I have faced when I was at home... so I prefer the hospital because at least there is room for confidentiality (In-person navigation FGD).

AYA also wanted their interaction with navigators to include providing education on important topics to them. The topics suggested covered HIV care, sexual and reproductive health, substance abuse and general updates on developments in facilities and communities.

...they should put the major topics... yeah like let’s say depression, alcohol use... drugs yeah... bullying, and sexual activities those are the things that they should be counselled on (In-person navigation FGD).

Discussion

The goal of this study was to understand how peer navigation delivered electronically, and in-person can be tailored to offer optimal support to and improve HIV treatment outcomes based on AYA preferences. We found that AYA preferred navigators in the same age group and not necessarily of the same gender but people with ability to promptly develop personalized relationships with encouragement and support. AYA valued navigators knowing about the home environment of the AYA and what is important in their lives beyond HIV. AYA desired assurance of confidentiality because of the sensitive nature of their interactions and only sending messages that are status neutral with no HIV connotations. We also found that the need for confidentiality informed the preferred choice of the channel of communication with the most preferred approaches being WhatsApp, text messages and phone calls which they considered more secure.

The AYA population has varying peer navigation preferences that suit their individual circumstances(Toth et al., 2018). While some require minimum encouragement to continue with care engagement, others need more intensive and innovative navigation to remain in care. Several researchers have discussed the need for differentiated and targeted interventions for AYAs. For example, Desai and co-authors describe the heterogeneity of AYA and advocate for differentiated and integrated services that take into account their demographic, sexual, and socio-cultural differences(Desai et al., 2023). Further, Shahmanesh and colleagues found that AYA want accessible youth-friendly clinical services that overall help to improve HIV engagement experience(Shahmanesh et al., 2021).

Our findings that electronic and in-person navigation were acceptable to AYA for follow-up and support during HIV care and treatment mirror others elsewhere (Pannier et al., 2019; Ranney et al., 2014; Warner et al., 2018). These studies independently revealed that in-person navigation was most acceptable but required reinforcement of multiple communication platforms including phone calls, text messages and email for follow-ups. Separately, Carbone et al showed that adolescent girls and young women enrolling in perinatal HIV services in Malawi preferred mentor-mother support from their peers who had successfully navigated perinatal HIV prevention care (Carbone et al., 2019). The preference could have been due to the belief that sharing their experience with someone in the support system (peer navigator) with similar previous experience could possibly reduce judgmental attitudes during navigation and therefore reduce stigma and stigma related challenges.

Mobile health (mhealth) is emerging as a key strategy to address the healthcare needs of AYA who report barriers to in-person care. AYA have access to phones and other smart gadgets(Champion et al., 2019; Kharono et al., 2022) which is an acceptable and preferred mode of receiving some interventions(Rankine et al., 2023). A systematic review of effectiveness of school-based mhealth interventions to prevent multiple lifestyle risk behaviors among adolescents demonstrated the effectiveness of the interventions to improve physical activities and fruit and vegetable intake and reduce screen time(Champion et al., 2019).

Due to stigma and other concerns, privacy and confidentiality has been shown to be important for AYA’s status disclosure, enrolment and retention HIV care as well as uptake of other support services(Khan et al., 2023; Miyingo et al., 2023). It is for this reason that among the qualities of a navigator that AYA desired trustworthiness to keep confidential any information they access in their peer navigation relationship. Indeed, suspicion of breach of privacy and confidentiality has been responsible for AYA avoiding some health facilities and even programs. For instance, HIV delivery models that were perceived to lack privacy and confidentiality received minimal interest among AYA(Miyingo et al., 2023). As such, HIV care interventions, including electronic and in-person navigation strategies, that intentionally put issues of privacy and confidentiality at the forefront are likely to be successful.

This study had a few limitations. First, we had a small sample size of two FGD (n=17 participants) each focused on one of the two strategies being assessed. As such these two may not have captured all the diverse views of AYA but were likely sufficient to identify the majority of possible themes. (Guest et al., 2017). These data were complemented in the overall trial by a discrete choice survey and a human-centered design workshop(Akama et al., 2023; Eshun-Wilson et al., 2022). Findings were consistent across these formative research components, supporting our FGD findings even with a small sample. However, we maximized their potential by sampling participants from different facilities thus enabling us to achieve heterogeneity in key characteristics in our purposive sampling. Second, given that the FGDs had adolescents of mixed gender, educational levels, and marital status, it is possible some participants may have been less comfortable expressing their opinions and views. However, our experienced facilitators provided an opportunity for all participants to speak by giving protected time to those who seemed shy or uncomfortable speaking. Third, our study enrolled AYA aged 14–24, the age range that accommodates both young and older adolescents and therefore the variation in navigation preferences could have been influenced by developmental age. Despite these limitations, our study gives important information to guide in designing interventions and approaches that are adolescent friendly to keep them engaged in HIV care.

Conclusions

In conclusion, we established that AYA found both electronic and in-person navigation acceptable especially with navigators with ability to promptly develop a personalized relationship that is supportive of medication course. Further, AYA desired assurance of confidentiality because of the sensitive nature of their interactions and only sending messages that are status neutral with no HIV connotations. Specifically, AYA’s preferred interventions would be those delivered through secure communication platforms by navigators with whom they have common bonds and possess characteristics that support development of strong relationships. Additionally, electronic mechanisms can help support the relationship building that is at the core of our navigation approach as well as a fundamental aspect of social work in general. These findings are crucial in informing the design of adolescent interventions, especially those that relate to in-person and electronic peer navigation to retain AYA in care.

Data availability statement

All anonymized data relevant to this study are readily available upon request from the corresponding authors and approval by the Kenya Medical Research Institute’s Scientific and Ethics Review Unit in line with the requirements of the Kenya Data Protection Act 2019.