Repeat and prolonged treatment interruption (TI) is common and the major threat to HIV epidemic control in eastern and southern Africa. The proposed project will test an innovative long-term dynamic choice intervention for ART clients experiencing TI in Malawi. Findings will provide essential information on how to improve sustained retention among TI client, a critical step to curbing the HIV epidemic. TI clients need long-term, responsive interventions. There are no one-size fits all intervention to support long-term care for TI clients because clients experience vastly different and changing barriers to care. While health facilities do have limited capacity for adding new services, existing services can be packaged differently to meet clients' needs.

Long-term, dynamic choice of services is one way to provide responsive services and promotes client ownership over care. The investigators propose to give TI clients long-term, dynamic choice of what services they receive and how they receive it (drawing from key building blocks of DSD). Long-term, dynamic choice puts clients in the driver's seat and may be the best practical strategy to provide long-term and responsive TI interventions that are tailored to clients' evolving life circumstances. Dynamic choice is frequently used for HIV prevention and family planning products, whereby clients select the type of health product that works best for them (i.e., condoms, injectables, etc.). Choice of these services is strongly associated with improved outcomes.

The goal of this clinical trial is to determine if CHOICE can improve outcomes in TI clients, compared to standard of care (SOC). Participants will be randomly assigned to either the CHOICE or SOC group, and follow them for 12 months. The primary outcome will be viral suppression at 12 months.

Repeat and prolonged treatment interruption (TI) is common and the major threat to HIV epidemic control in eastern and southern Africa. The investigators define TI as >28 days late for last ART appointment. Each TI episode is associated with rapid viral load rebound and long-term decreases in CD4 count. Up to 46% of ART clients in the region experience TI in the first years after treatment initiation. 30-50% of clients returning from TI experience repeat TI <6 months after returning to care. TI and repeat TI account for the majority of HIV transmissions in Malawi. Strategies to provide ongoing support to TI clients are urgently needed.

Current TI services do not work in the long-term. The few evidence-based TI interventions that exist are extremely limited in duration (usually only ~1-3 intervention visits) and services offered (counseling alone or counseling + home-based initiation). Yet reasons for TI are different for each client and change over time. TI is rarely a one-time event. Short term interventions can improve return to care, but do not address repeat TI. The investigators' recent R01 (ENGAGE; R01MH122308-05, n=735) and BMGF (IDEAL; n=569) trials with men experiencing TI in Malawi show that short-term person-centered counseling + variations of short-term home-based ART dramatically improve return to care after TI (~96% re-initiated vs 67% in standard of care - SOC; p<0.001). But rates of continued ART engagement at 6-months after re-initiation (4-5 months after interventions completed) was unacceptably low (~64% were continuously in care without repeat TI vs 48% in SOC, p<0.001).

Recent trials found that clients with repeat TI had many unexpected and changing barriers that require long-term solutions. In-depth interviews with repeat TI clients in ENGAGE and IDEAL trials (n=67) showed that clients want long-term services responsive to their changing needs, and the ability to choose services.

TI clients need long-term, dynamic choice of differentiated service delivery (DSD) options. There are no one-size fits all interventions for TI clients because they experience vastly changing and different barriers to care, from distance to facility to lack of social support to limited HIV-related literacy. Ongoing person-centered counseling and choice of how services are delivered facilitates responsive services and promotes client ownership over care. Long-term and dynamic choice of DSD may be the best practical strategy to provide long-term, responsive TI interventions. There are no evidence-based dynamic choice interventions for TI clients.

The investigators piloted a novel intervention, called CHOICE, that provides person-centered counseling (building on the success of our prior trials) + ongoing, dynamic choice of DSD services to TI clients. The pilot (n=125) found that counseling + dynamic choice of DSD was feasible, acceptable, and showed signs of efficacy to reduce repeat TI (15% vs 50% at 3-months). A full trial is needed to test the impact on viral suppression long-term (at 12- and 24-months) and cost-effectiveness.

The investigators propose to evaluate CHOICE in a randomized control trial (RCT) with TI clients in Malawi. TI clients will receive person-centered counseling + ongoing, dynamic choice of how services are delivered (drawing on DSD models) over 24-months. Clients can combine multiple choices for service delivery at any given time to create personalized intervention packages. Choice of service delivery include: 1) ART dispensing intervals (1, 3, 6months), location of ART distribution (facility, home, or community), and peer mentorship (varying frequency and meeting location). Clients can adjust their choice at each ART visit or via hotline throughout study period. The investigators will compare CHOICE to SOC (1-3 counseling sessions + routine facility-based services - no choice available <6-months after TI. Specific aims are:

Aim 1. Test the effectiveness of CHOICE versus SOC on 12-month viral suppression among TI clients. The investigators will conduct an individually randomized trial at 12 health facilities (n=800 individuals). Primary outcome is viral suppression (<50copies/mL) at 12-months (study collected sample). Secondary outcomes are: presence of repeat TI, time to repeat TI, and ART coverage (days with ART in possession) across 24-months.

Aim 2: Systematically evaluate the implementation of CHOICE. The investigators will use mixed-methods to understand barriers, facilitators and needed improvements to HCW implementation of CHOICE, clients' ability to choose DSDs, and equity in intervention implementation and outcomes.

Aim 3. Estimate cost and cost-effectiveness of CHOICE. The investigators will use a micro-costing approach to estimate the distribution of care costs by study arm, differences between arms, and incremental cost effectiveness. The investigators will draw on patient-level resource use and unit costs developed from clinic and implementation financial data.

CHOICE will inform strategies for sustained retention among TI clients regionally and globally. The study is timely, of high-impact, and highly feasible within the Malawi health system and with our team. Malawi has a strong track record of implementing scientifically tested innovations into SOC that are high-impact and scalable, including several guidelines informed by our previous trials.

Inclusion Criteria:

• 15 years of age or older
• living with HIV;
• initiated ART for the first time>3-months ago (i.e., not a new initiate);
• non-pregnant;
• experienced Treatment Interuptions during their most recent ARTappointment (>28 days late).

Exclusion Criteria:

• pregnant women because the ART/antenatal care (ANC) programs are integrated and require additional considerations and linkages with antenatal services.

An established repository will be used to ensure access that facilitates FAIRness (Findable, Accessible,Interoperable, and Re-usable) of the data. The metadata and other supporting documents will be submitted to the National Institute of Mental Health Data Archive (NDA) to facilitate scientific collaboration and data sharing.

Per NDA guidelines, each participant will be assigned a global unique identifier (GUID) at trial enrollment, and informed consent will use appropriate language to convey data access and policies within NDA. All manuscripts and other dissemination materials will reference the Digital Object Identifier (DOI) associated with the appropriate data or document on NDA. Scientific data can also be found using standard indexing tools that optimize data locating via search engines. Keyword phrases for search engine optimization include: client empowerment for HIV services; HIV treatment interruption; Malawi; eastern and southern Africa.