Youth Ending the HIV Epidemic (YEHE)

Youth Ending the HIV Epidemic - Automated Directly Observed Therapy Pilot: Improving HIV Care Among Youth

Young adults have a disproportionately high rate of HIV infection, high rates of attrition at all stages of the HIV care continuum, an increased risk of antiretroviral therapy (ART) nonadherence and virologic failure, and a high probability of disease progression and transmission. Tracking and monitoring objective measures of ART adherence in real time is critical to strategies to support adherence and improve clinical outcomes. However, adherence monitoring often relies on self-reported and retrospective data or requires extra effort from providers to understand adherence patterns, making it difficult for providers to accurately determine how to support their patients in real time. In the proposed interventional study, the investigators aim to pilot test an automated directly observed therapy intervention paired with conditional economic incentives to improve ART adherence among youth living with HIV (YLWH) (18-29 years-old) who have an unsuppressed HIV viral load. Aim 1: Conduct a pilot study to assess feasibility and acceptability of the use of automated directly observed therapy with conditional economic incentives (aDOT-CEI) among YLWH (aged 18-29; N= 30) at AIDS Healthcare Foundation (AHF) clinics in California and Florida. Primary outcomes will be feasibility and acceptability, assessed using predefined feasibility metrics and acceptability surveys at three months. Aim 2: Explore experiences of YLWH and staff/providers with the aDOT-CEI intervention and implementation facilitators and barriers. The investigators will conduct in-depth qualitative interviews with a sample of YLWH from Aim 1 and staff/providers purposively selected from participating AHF clinics to explore intervention experiences, potential influences on ART adherence, individual-level and clinic-level barriers and facilitators to intervention implementation, and suggested refinements for a future efficacy trial. The investigators hypothesize that the aDOT-CEI intervention to improve ART adherence among YLWH will have high feasibility and acceptability.

Automated directly observed therapy (aDOT) is an innovative technology that uses artificial intelligence (AI) with computer vision and deep learning algorithms to track and support adherence through a smartphone. Additionally, aDOT provides a seamless and convenient platform for providing Conditional Economic Incentives (CEIs) because it monitors real-time adherence to automatically determine who can receive incentives. For the design and development of the mobile health app, the investigators have partnered with AiCure to use an existing HIPAA-compliant mobile health app. The investigators will invite YLWH from AIDS Healthcare Foundation (AHF) sites in CA and FL to form the study Youth Advisory Panel (YAP) and seek their input on the AiCure app. The investigators will work with AiCure to implement any required changes to the app that have emerged from formative research. The app will then be piloted with YLWH (aged 18-29; N= 30) who will use the platform for a period of 3 months (Aim 1). The app will record video of the participant taking their HIV medication in order to monitor the participant's medication adherence. Participants will complete online surveys at baseline and 3 months. The investigators will have monthly check-ins with participants which the investigators will assess app use and help increase study engagement. The investigators will measure feasibility and acceptability through app paradata (i.e., app use information) and self-report in surveys (baseline, 3 months). And the investigators will use adherence-related medical record data from AHF to compare against adherence monitored by the AiCure app. Following completion of the pilot, the investigators will conduct in-depth interviews (IDIs) with YLWH and staff /providers purposively selected from participating AHF clinics (Aim 2). Interviews will explore intervention experiences, potential influences on current and long-term ART adherence, unaddressed adherence barriers and the potential benefit of features (e.g., reminders), individual-level and clinic-level barriers and facilitator to intervention implementation, assess ease of use of aDOT-CEI, likes and dislikes, and suggested modifications for a future efficacy trial. This interventional pilot study will assess the feasibility and acceptability of aDOT-CEI and will provide preliminary data to inform an R01 to test the efficacy of aDOT-CEI in addressing disproportionately low viral suppression among YLWH.

All YLWH who choose to enroll in the study will receive access to AiCure, the mobile health application. The participants will be asked to use the app for 3 months, during which the investigators will assess the feasibility and acceptability of AiCure.

For the design and development of the mobile health app, we have partnered with AiCure to use an existing HIPAA-compliant mobile health app. The app will record video of the participant taking their HIV medication in order to monitor the participant's medication adherence and track incentives for this behavior in real-time.

Mobile app data from the AiCure application (paradata) will be used to measure the duration of each participants participation in the trial. Rate of Participant Retention will be calculated as the percent of time the participant participated in the trial versus the total trial duration.

Mobile app data from the AiCure Application (paradata) will be used to measure the number of participant logins per week and thereby calculate mean logins per participant per week.

Mobile app data from the AiCure Application (Paradata) will be used to measure the number of minutes each participant spends in that app each day, thereby calculating the mean number of minutes in app per day per participant.

The AiCure mobile application platform uses an artificial intelligence or AI platform to recognize dosing patterns recorded by users that do not correspond to what has been defined as 'normal' dosing by the AI. The AI accordingly flag's any video recording featuring abnormal dosing for review by the AiCure Video Review team. Upon confirmation of abnormal dosing, the AiCure Video Review team notifies the research coordinator in the form of a "Red Alert". Each of these "Red Alerts" message will be counted as an event of intentional nonadherence, and the mean number of times a participant may have "falsified" medication taking will be calculated and compared against the number of 'normal' doses.

The acceptability of the AiCure Mobile application will be through a system usability scale. 10 items and 4 scale points. For odd items: subtract one from the user response. For even-numbered items: subtract the user responses from 5. This scales all values from 0 to 4 (with four being the most positive response). Add up the converted responses for each user and multiply that total by 2.5. This converts the range of possible values from 0 to 100 scale, with a higher score corresponding to greater usability. Scores above 68 considered above average and acceptable, and scores at or below 68 considered below average and not acceptable.

Client satisfaction will be measured using a Client Satisfaction Questionnaire (CSQ-8); 8 items. An overall score is calculated by summing the respondent's rating (item rating) score for each scale item. Scores range from 8 to 32, with higher values indicating higher satisfaction.

Recommend study to a friend. 7-point Likert Scale of how likely participants are to recommend the study to a friend. Scale values from 1-7 with lower values corresponding to less likely and higher values corresponding to most likely.

Client satisfaction with adherence monitoring, conditional economic incentives, app reminders reminders, security and privacy, support. 7-point Likert Scale values from 1-7 with lower values corresponding to least satisfaction and higher values corresponding to greater satisfaction.

Adherence data will be abstracted from the automated directly observed therapy platform within the AiCure mobile health application. The app tracks and validates if and when a client has Adherent: ≥80% of doses taken Nonadherent: <80% of doses taken.

Survey data. How much did participation in this study help you improve adherence to your medications? 3-item self-reported measure of ART adherence. Adherent: no missed doses Nonadherent: any missed doses.

How easy/difficult was it to use your personal phone; use adherence monitoring; receive incentives? Measured using a 7-point Likert Scale ranging from easy to difficult. Higher values correspond to greater ease of use, lower values correspond to greater difficulty.

Did you ever have trouble accessing app, using adherence monitoring, receiving reminders, receiving incentives, or finding a private place? 7-point Likert of frequency ranging from often to never with high value corresponding to app related issues occurring with greater frequency and lower value corresponding to app related issues issues occurring with lesser frequency.

Inclusion Criteria: 18 to 29 years of age Have access to a smartphone Can speak and read English Client is at an AHF clinic in California or Florida Client has an unsuppressed HIV viral load at least 3 months post HIV diagnosis Client is on once daily oral antiretroviral therapy Consent to participate in the proposed study Exclusion Criteria: YLWH who are newly HIV diagnosed in the past 3 months Smartphone is a Samsung galaxy s21 or iPhone 5SE (app is not supported by these phones)