In contrast all other age groups, adolescents (age 10-19 years) have experienced constant, rather than declining, HIV-related mortality in the last decade. This is due in part to poor retention in care and adherence to antiretroviral therapy (ART) in this age group. As youth living with HIV enter adulthood, they transition from pediatric to adult HIV care. The transition to adult care presents heightened challenges to retention, due to disruption of established relationships with pediatric healthcare providers and a shift to an adult care model that requires greater autonomy and offers less specialized support. Resources to support youth through the transition are lacking. Mobile technology for health promotion (mHealth) using social media is a promising approach to maintain engagement in care through the transition. mHealth reminders, education, and support from healthcare workers have been successful in improving ART adherence in the adult HIV care setting and there have been limited, though promising, results in youth care. This study will support the development of a novel mHealth strategy for the Pediatric to Adult HIV Care Transition (mPACT). The conceptual framework for this intervention is based on providing support to youth who are transitioning to adult care through a combination of virtual group peer support and 1-to-1 communication with a healthcare worker trained in youth HIV care.

The aims of this study are to (1) identify the specific barriers to successful youth transition to adult HIV care, develop the mPACT intervention messaging strategy, and create a prototype of the mHealth platform; (2) Pilot the mPACT intervention to determine its effect on the intermediate outcomes of transition preparedness, ART knowledge, stigma, depression, social and caregiver support.

Using an iterative mixed methods approach we will develop and pilot the mPACT intervention to improve transition to adult HIV care. Aim 1 focuses on identifying barriers, assessing user requirements, developing and refining communication content and strategies, and adapting existing technology platforms through in-depth individual interviews and focus group discussions and with youth living with HIV, healthcare works, caregivers, and policy makers. Aim 2 will pilot the intervention tool using a cluster randomized trial to evaluate the impact of the intervention on intermediate factors relevant to transition to adult HIV care.

AIM 1a: Youth IDIs and caregiver and HCW/peer counselor FGDs Participants in in-depth individual interviews (IDIs) and focus group discussions (FGDs) will be recruited from the study clinics. IDIs and FGDs will be performed in a private room at the clinic from which the participant was recruited. IDIs and FGDs will be conducted by a trained qualitative interviewer fluent in English and Kiswahili. Before the IDI or FGD begins, socio-demographic information will be collected in standardized questionnaires using a tablet-based system (Open Data Kit, ODK). For youth, sociodemographic data to be collected include age, HIV treatment history, HIV status disclosure, family socio-economic status, education level, and technology access. For caregivers, data to be collected include age, relationship to the youth patient, and involvement in youth's HIV care. For HCW/peer counselors, data include age, professional role, time in practice and technology access. The interviewer will then describe procedures and norms for the IDI/FGD and participants will be given a chance to ask questions. Participants will be reminded that their participation is voluntary and they may refuse to answer any question or end the interview/FGD at any time. Participants will receive unique identification numbers and will not be addressed by their names to maintain confidentiality. The interviewer will then ask questions and facilitate a discussion based on a discussion guide. Aim 1a IDIs will last approximately 1 hour and Aim 1a FGDs will last 1.5 - 2 hours. The IDI/FGD will be audio recorded and the facilitator will take written notes. Discussions will be conducted in English or Kiswahili, depending on participant preferences. Notes will be compared to audio-recordings to fill in missing information, transcribed and translated to English (if necessary). Transcribed data will be de-identified. Audio-recorded interviews will be destroyed after the transcripts have been validated. Participants will be provided refreshments and KSh 400 to compensate for their time and transportation expenses to participate in the study. We will provide this monetary compensation to each participant at the conclusion of each IDI/FGD.

AIM 1a: Policy-maker IDIs Policy-makers will be approached and recruited to participate. IDIs will be performed in a private room at the policy-maker's place of work or at the study clinic, depending on the participant's preference. IDIs will be conducted by a trained qualitative interviewer fluent in English and Kiswahili. Before the IDI begins, socio-demographic information will be collected in standardized questionnaires using a tablet-based system (Open Data Kit, ODK). Data to be collected include age, professional role, and involvement with policy-making and implementation related to youth HIV care. The interviewer will then describe procedures and norms for the IDI and participants will be given a chance to ask questions. Participants will be reminded that their participation is voluntary and they may refuse to answer any question or end the interview/FGD at any time. Participants will receive unique identification numbers and will not be addressed by their name to maintain confidentiality. The interviewer will then ask questions and facilitate a discussion based on a discussion guide. Policy-maker IDIs will last approximately 1 hour. The IDI will be audio recorded and the facilitator will take written notes. Discussions will be conducted in English or Kiswahili, depending on participant preferences. Notes will be compared to audio-recordings to fill in missing information, transcribed and translated to English (if necessary). Transcribed data will be de-identified. Audio-recorded interviews will be destroyed after the transcripts have been validated.

AIM 1b: User-centered design FGDs with youth and HCW/peer counselors Based on input from interviews and FGDs conducted in Aim 1a, we will develop a prototype mHealth intervention that will use WhatsApp to provide peer support groups and 1-to-1 text messaging between youth and HCWs. The prototype will include messaging content, facilitation guides, and an interface to manage communication that will be used by facilitators and HSWs. To evaluate the acceptability of the prototype and to get feedback on how it can be improved, we will conduct 2 rounds of user-centered designed FGDs with youth and HCW.

We will evaluate the prototype intervention using user-centered design FGDs with youth living with HIV who are in the process of transitioning to adult care or have completed the transition and with HCWs/peer counselors who care for youth living with HIV. Participants for each set of FGDs will be recruited as described above. Following informed consent, participants will complete an enrollment visit in which questionnaire data will be collected using tablet-based ODK forms. For youth, data to be collected include sociodemographic characteristics such as age, family socioeconomic status, education level, technology access, HIV treatment history, social support, depression, stigma, ART knowledge and motivation, ART adherence. For HCWs/peer counselors, data include age, professional role, time in practice and technology access. FGDs will be conducted in 2 rounds. In the first round, FGDs with youth will explore their comprehension, interest and reactions to prototypical messages and discussion topics. FGDs with HCW will explore their reactions and comfort with prototypical messages and discussion topics. HCW will also be presented with the prototype message management interface, training materials, and their input sought on improvements. The second round of FGDs will recruit a new group of participants to evaluate the modified content and interface. Participants will be provided refreshments and KSh 400 (approximately $4) to compensate for their time and transportation expenses to each study visit, provided to participants at the conclusion of each visit.

Aim 2 Pilot intervention trial

Consent forms and data collection materials will be generated based on findings from the formative work conducted in Aim1. These materials will be submitted for approval as a modification prior to initiation of the pilot trial.

Intervention package: The intervention package will be piloted at the 4 intervention facilities and compared to the 4 control facilities. The intervention package will consist of a WhatsApp group that is setup and organized by the study. Participants will be divided into 1-3 WhatsApp groups, each with 10-55 youth. Each group will be facilitated by a study staff member trained as a counselor (the same facilitator will facilitate all groups). The facilitator will send weekly messages to the group, containing content related to the topic areas outlined in Table 4. Messages from the facilitator will not contain overtly HIV-related language. These messages are intended to spark unstructured discussion among study participants, or with the facilitator, at any time through the WhatsApp group or individually. Subsequent messages from the study will answer participant questions or correct misinformation. Standard operating procedures (SOPs) will be developed for standardizing responses to the questions received so that study staff follow national guidelines and local practice standards to respond. The process will be adaptive as new scenarios unfold - a new question not addressed by the SOPs will be discussed by study team and SOP developed if it is a question likely to be encountered repeatedly. Weekly review of the log of WhatsApp chats by the study PIs will enable standardization and quality assurance of WhatsApp interactions.

At the start of the intervention, participants will be invited to attend an optional in-person meeting at the study clinic with the other members of their WhatsApp group to establish connection and discuss the name and agreements of the group. This meeting will be a completely optional component of the intervention. Participants who attend this meeting will be provided refreshments and KSh400 (approximately $4) to compensate for their time and transportation expenses, provided at the end of the meeting.

The study team will monitor the discussion to learn about the content of the discussions and how it relates to the intervention aims. We will also monitor for abusive or problematic content or behavior in the discussion and will intervene if necessary. Guidelines and ground rules for the discussion group will be presented to all members when they begin participation and all participants will be informed that the content of the discussions will be monitored by the study team. They will also be informed that the content will be kept confidential, but that it is possible that members of the discussion group could potentially disclose confidential information outside of the group, though this will be explicitly discouraged by the study team. The technology development for the intervention package will be conducted by a contractor that has not yet been identified. The identity of the contractor will be submitted to the ERC prior to the start of the contract period.

Study population: The pilot trial will be conducted at 8 facilities. To maximize internal validity of the pilot, facilities will be selected based on adequate number of youth in care and similarity of youth patient populations. We will randomly select 4 facilities to receive the intervention and 4 facilities to serve as control sites. Eligible participants will be youth aged 16-24 living with HIV, who have daily access to a mobile phone (their own or shared) that can access the intervention technology platform. The stage of transition at which youth become eligible will be based on their score on the Transition Readiness Assessment Questionnaire (TRAQ), with the threshold for eligibility to be determined in Aim 1. If Aim 1 indicates that peer chat groups should be tailored based on youth characteristics (e.g., route of transmission, developmental age), we will limit the trial to the subgroup with the largest number of youth, but will also assess interest among youth from other subgroups in participating in a future peer intervention. We will enroll 50 participants from intervention facilities (~12 at each facility) to receive the intervention for 12 months and 50 participants from control facilities who will be followed for 12 months without intervention. Based on data from the ongoing ATP study, we estimate 20% of facilities nationally have at least 50 patients in this age group engaged in care annually. We expect to complete enrollment within 3 months.

Evaluation of intervention impact: A screening questionnaire will be used to assess eligibility and collect sociodemographic characteristics such as age, school attendance and technology access of all screened individuals. In order to evaluate the intervention's reach and uptake, differences in these characteristics between screened and enrolled participants will be assessed by chi-squared test for categorical and t-test for continuous variables. At baseline (enrollment) and endline (12 months), participants will complete a standardized questionnaire to evaluate intermediate outcomes in our intervention framework. Participants who do not return for their 12-month visit will be traced by phone call and home visit. Participants will be provided refreshments and KSh 400 to compensate for their time and transportation expenses to each study visit, provided to participants at the conclusion of each visit.

Mediating and moderating factors: Based on our conceptual model linking the intervention to elements of social support that promote the care transition, we expect the intervention will affect successful transition through multiple mechanisms. While the pilot trial is not powered to assess impact on primary transition outcomes, we will focus on its impact on hypothesized mediators.

Inclusion Criteria:

• Aware of HIV status
• Involved in youth HIV care
• Ready but not yet transitioned to adult HIV care
• Has access to mPACT intervention messaging platform

Exclusion Criteria:

• Unable to provide informed consent