Implementation of a Dyad-Based Intervention to Improve Antiretroviral Therapy Adherence Among HIV-Positive People Who Inject Drugs in Kazakhstan: A Randomized Trial

Abstract

Purpose:

HIV-positive people who inject drugs (PWID) in Kazakhstan face many challenges to antiretroviral therapy (ART) adherence. Interventions that leverage social support from an intimate partner, family member, or friend may be effective in improving ART adherence among this population. The purpose of this paper is to describe the implementation process of a dyad-based intervention among HIV-positive PWID and their treatment support partners.

Method:

Sixty-six HIV-positive PWID and 66 of their treatment support partners will be enrolled in this pilot randomized controlled trial in Almaty, Kazakhstan, and randomized as dyads to receive an adapted version of the SMART Couples intervention or standard of care.

Results:

Several implementation strategies were used to facilitate intervention delivery, including remote delivery, training of staff, supervision, technical assistance, quality assurance, and collection of assessments through diverse sources.

Discussion:

This trial responds to a need for dyad-based ART adherence interventions adapted specifically for HIV-positive PWID.

The fastest-growing HIV epidemics globally are driven by injection drug use, but only a small percentage of HIV-positive people who inject drugs (PWID) have achieved viral suppression (Degenhardt et al., 2017; Des Jarlais et al., 2012; Godinho et al., 2004; Strathdee et al., 2010). Eastern Europe and Central Asia (EECA) are experiencing the fastest-growing HIV epidemic in the world, and Kazakhstan has seen the largest increase of new HIV cases in the region (UNAIDS, 2021). From 2010 to 2020, Kazakhstan experienced a 73% increase in HIV incidence, despite substantial declines of up to 49% in some other countries in the region (UNAIDS, 2021). The rapid increase in incident HIV cases in EECA is largely driven by opioid injection (UNAIDS, 2021). In EECA, HIV-positive PWID have low rates of antiretroviral therapy (ART) adherence and viral suppression (UNAIDS, 2021). In Kazakhstan, the prevalence of HIV among PWID (8.3%) is higher than any other key population (sex workers —1.4%, men who have sex with men—6.5%, and prisoners —4.1%) (UNAIDS, 2021). In 2020, of all HIV-positive PWID in Kazakhstan, it is estimated that 80% knew their HIV status, 46% were receiving ART, and only 38% were virally suppressed, indicating significant adherence challenges (Kazakh Scientific Center of Dermatology and Infectious Diseases, 2021). By comparison, among all people living with HIV (PLWH) in Kazakhstan, it is estimated that 83% knew their HIV status, 61% were receiving ART, and 51% were virally suppressed (Kazakh Scientific Center of Dermatology and Infectious Diseases, 2021).

There are numerous barriers to ART adherence among HIV-positive PWID in Kazakhstan. At the individual level, PWID may have misperceptions about the safety or effectiveness of ART, be prone to forgetting to take medication due to the effects of their drug use or choose not to take medication due to negative side effects (Davis et al., 2018). At the interpersonal level, PWID with few social supports and/or negative interactions with healthcare providers may have increased difficulty adhering to ART or coping with their HIV status on their own (Davis et al., 2018). Structural factors that may impede adherence include poverty, legal obstacles, disruptions in the availability of ART, and wide-spread stigma and discrimination (Davis et al., 2018).

Studies on ART adherence interventions among HIV-positive PWID in low- and middle-income countries (LMIC) are limited (Malta et al., 2008), which is problematic given the large HIV epidemics among PWID in these countries (Degenhardt et al., 2017; Mathers et al., 2008; Strathdee & Stockman, 2010). The lack of evidence-based interventions (EBIs) to promote ART adherence among PWID in the region continues to undermine effective treatment as prevention responses to the HIV epidemic. Furthermore, there are areas in many other countries globally that are similar to EECA, where volatile HIV epidemics among PWID spread due to punitive drug laws and inadequate HIV care, including in Asia, Africa, the Middle East, and the Appalachian region and other areas of the United States (Keyes et al., 2014; Malta et al., 2008). Given that the EECA has among the highest injection drug use rates and prevalence of HIV among PWID in the world (Degenhardt et al., 2017), the region is an optimal location to develop and adapt adherence EBIs for PWID. ART adherence interventions for PWID in EECA could prove beneficial in other areas of the world where PWID represent a significant share of those in need of HIV treatment.

ART adherence is critical to achieving viral suppression and reducing the spread of HIV (Cohen et al., 2011). Because HIV-positive PWID are from high-risk social networks, improving adherence to ART may reduce sexual and parenteral transmission of HIV to the broader community (Malta et al., 2008). Studies have found that HIV-positive PWID can reach similar levels of ART adherence found among PLWH who have never used illicit drugs, particularly with social support and medication-assisted therapy (MAT) for substance use (Crystal et al., 2001; Malta et al., 2008). Evidence indicates that the simultaneous combination of ART adherence interventions with linkage to MAT and needle-exchange programs has a joint effect and provides more positive clinical outcomes and a greater reduction in HIV transmission than individual strategies alone (Degenhardt et al., 2010; Malta et al., 2008, 2010; Strathdee et al., 2010).

PWID experience high levels of stigma and discrimination from the community and healthcare providers, both for their HIV status and for injecting drugs (Davis et al., 2018). Evidence suggests access to social support from intimate partners, family members, and friends can be critical in improving ART adherence among HIV-positive PWID (El-Bassel & Remien, 2012; Kelly et al., 2001; Knowlton et al., 2006; Lucas et al., 2001; Palepu et al., 2006; Remien et al., 2005; Wood et al., 2004). Yet most HIV interventions focus on individuals as the unit of change, ignoring social dynamics and the important role others play in altering risk behaviors and HIV treatment engagement and adherence (El-Bassel et al., 2010; Jiwatram-Negrón & El-Bassel, 2014). Dyad-based interventions have several advantages over individual approaches, including (1) promoting mutual responsibility in protecting each other from HIV transmission and improving each other’s health and ART adherence; (2) highlighting social relationship contexts and their connection to HIV acquisition and disease progression; and (3) creating a safe environment to discuss sensitive topics, such as relationship conflict, sexual concurrency, needle-sharing, and medication adherence (Jiwatram-Negrón & El-Bassel, 2014).

Evidence has shown that dyad-based interventions are effective in reducing HIV acquisition, HIV risk behaviors, overdose, and increasing the proportion of prescribed ART doses taken (El-Bassel et al., 2014; Jiwatram-Negrón & El-Bassel, 2014; Remien et al., 2005). However, despite the promising potential of dyad-based social support interventions, PWID are often excluded from dyad-based intervention trials and studies on couple-based interventions remain limited globally (Jiwatram-Negrón & El-Bassel, 2014). In focus groups conducted with 58 HIV-positive PWID in Kazakhstan, participants reported a strong desire for a social support intervention and cited partner support as a key facilitator of ART adherence, stating that partners provided needed support for coping with their HIV status (Davis et al., 2018). Despite the desire for a partner-based approach, there is only one dyad-based ART adherence EBI (SMART Couples) (Remien et al., 2005) on the Centers for Disease Control’s effective behavioral interventions website (Centers for Disease Control and Prevention, 2022), but it has not been adapted for PWID. The integration of substance use risk reduction components, the incorporation of linkage to MAT and harm reduction services, and the addition of new biomedical measurement tools that provide an objective measure of adherence and clinical utility could substantially improve adherence among HIV-positive PWID, especially in resource-constrained settings (Binford et al., 2012; Thompson et al., 2012).

There is also limited implementation research on intervention delivery in Kazakhstan and EECA. Implementation data are necessary to assess how to deliver interventions in limited-resource settings and determine what strategies are necessary to facilitate delivery and sustain them. This is particularly important for socio-behavioral interventions that are delivered by social workers, psychologists, and other mental healthcare providers, as these professionals navigate other challenging social and environmental factors, such as the criminalization of substance use, lack of basic human resources, and past trauma, as they work to improve patient health and promote adherence to ART. This paper describes the study protocol for adapting and implementing the Kazakh (KZ) SMART Couples intervention, a dyad-based medication adherence intervention for people with HIV, among HIV-positive PWID attending the Almaty City AIDS Center and their treatment partners (intimate partners, family members, and friends).

Method

Overview of Study and Study Aims

The KZ SMART Couples Study is a pilot efficacy trial that involves adapting the SMART Couples ART adherence intervention for HIV-positive PWID and their treatment support partners, and testing the efficacy of the adapted intervention. Study aims include:

• Aim 1: Identify participants’ perceptions of the core components of the SMART Couples intervention and other existing ART adherence intervention strategies and assess their appropriateness and feasibility as an integrated, dyad-based ART adherence intervention for HIV-positive PWID living in Kazakhstan through in-depth interviews with HIV-positive PWID, their partners, and healthcare providers.

• Aim 2: Adapt and refine SMART Couples and identify augmentative intervention strategies guided by the Collaborative Intervention Planning Framework (CIPF) to create an integrated, dyad-based ART adherence intervention for HIV-positive PWID and their partners using results from Aim 1.

• Aim 3: Pilot test of the resulting dyad-based intervention among 66 PWID and their treatment support partners (n = 66) in Kazakhstan through a randomized control trial (RCT) to assess the safety, feasibility, and acceptability of the intervention and obtain preliminary estimates of adherence outcomes (i.e., viral load, tenofovir-diphosphate blood levels, electronic monitoring device (EMD), and self-reported adherence) in the intervention arm versus standard of care (SOC).

Study Design

The study is a 5-year, three-phase project implemented by research and clinical staff in Almaty, Kazakhstan. The project aims to adapt, refine, and test a dyad-based ART adherence intervention for HIV-positive PWID using the steps of the CIPF (Cabassa et al., 2014) (Table 1). Following community-based participatory research (CBPR) principles, the framework of this study brings together researchers and community stakeholders through a community consultant group (CCG) and a community advisory board (CAB), who provide valuable input on social norms, cultural values, and local practices to tailor interventions to community realities (Devieux, 2005; NIMH Multisite HIV/STD Prevention Trial for African American Couples Group, 2008). In Phase 1, Formative Work, we conducted in-depth interviews with HIV-positive PWID, their partners, and AIDS Center healthcare providers. This formative qualitative work assessed the needs and attitudes of PWID and their partners regarding various intervention components and the attitudes and capacity of the AIDS Center clinic staff who would implement the intervention. In Phase 2, Intervention Adaptation, we recruited four PWID and four of their partners for the CCG and established a CAB comprised of key stake-holders. We adapted the intervention for PWID and for the local setting through focus groups with the CCG and working groups CAB, which drew on findings from our formative work. We included an examination of implementation issues, with attention to logistics and attitudes about psychosocial interventions, stigmas, and provider roles. Intervention sessions and components were reviewed with the CCG and CAB in order to adapt and refine the intervention. In Phase 3, Pilot Testing, 66 HIV-positive PWID and their partners will participate in an RCT of the three-session, adapted, dyad-based intervention developed from the previous two phases. We trained psychologists and social workers to deliver the adapted intervention to assess safety, feasibility, acceptability, retention, and preliminary estimates of adherence to directly inform a future, full-scale efficacy trial. Using a computer-based random number generator stratified by sex of the index case, dyads are randomized to the intervention or SOC arms by a research staff member and assessed with repeated measures at four time-points on ART adherence, behavioral outcomes (i.e., substance use and sexual risk behavior), social support processes, and biological measures from baseline to 6 months follow-up (Table 2). Dyads are blinded to the arm to which they are assigned. Electronic monitoring device (EMD) and weekly mobile app data are being collected throughout the intervention and follow-up periods.

Table 1.

Summary of the Collaborative Intervention Planning Steps for Intervention Adaptation With a Community Collaborative Group (CCG) and Community Advisory Board (CAB).

Step	Objectives	Activities
1. Setting the stage	Foster partnership and collaboration; Introduce project aims, intervention adaptation process and SMART Couples intervention; Presentation of results from pilot data	Review SMART Couples outline, summarize formative findings, discuss gender issues, mental health, substance use, and sexual risk
2. Problem analysis and needs assessment	Identify the needs of HIV + PWID; Discuss how the intervention may or may not address these needs; Identify areas for adaptation based on pilot results and CCG discussions	Brainstorming exercises, group discussions, overview of session content and homework activities, discussion of gender roles and gender issues in this context
3. Review of intervention sessions	Review the objectives, methods, and materials of the intervention by session; Identify specific adaptations to intervention content or delivery	Group discussions and role-play during intervention sessions, suggested adaptations to the intervention
4. Development of intervention adaptations	Incorporate adaptations into the intervention manual and materials; Finalize adapted intervention	Review of intervention manual and materials and evaluation of suggested adaptations

Table 2.

Measures According to Theoretical Constructs and Times of Collection.

Constructs	Measures	Timea
Health outcomes
Medication adherence	BETRCap data	Daily
	Self-report measure for medication adherence—collected by weekly app assessment (3 items, α = 0.83)	Weekly
Clinical outcomes	Viral load—collected by nurse	BL, 6 mo
	TFV-DP levels (DBS)—collected by nurse	IPI
	ART levels hair sample—self-collected	BL, IPI
	Urine Tenofovir POC test—self-collected	BL, IPI, 3 mo, 6 mo
Medical appointment attendance	Clinic attendance records—extracted from medical records	BL, 6 mo
Couple interactions
Social support	Medication specific social support (8 items, α = 0.92); Social support and relationship satisfaction—weekly app assessment	BL, IPI, 3 mo, 6 mo
Relationship satisfaction	Global distress scale—marital satisfaction inventory revised (22 items, α = 0.95)	BL, IPI, 3 mo, 6 mo
Self-regulation
Knowledge of ART	HIV treatment knowledge inventory (12 items, α = 0.70)	BL, IPI, 3 mo, 6 mo
Knowledge of treatment	Knowledge of personal treatment regimen	BL, IPI, 3 mo, 6 mo
Adherence attitudes	Attitudes about HIV Medications (11 items, α = 0.70)	BL, IPI, 3 mo, 6 mo
Substance use	Alcohol use disorders identification test (AUDIT) (10 items, α = 0.85); drug abuse screening test (DAST) (10 items, α = 0.94); drug treatment history	BL, IPI, 3 mo, 6 mo
Self-efficacy	Adherence self-efficacy (13 items, α = 0.96)	BL, IPI, 3 mo, 6 mo
Risk behaviors	Sexual risk and substance use risk behaviors	BL, IPI, 3 mo, 6 mo
Contextual influences
Background	Demographics	BL
Environmental stress	Economic and food insecurity, stressful life events	BL, IPI, 3 mo, 6 mo
Stigma	Social impact scale for HIV (24 items, α = 0.85); drug use stigma (9 items, α = 0.80)	BL, IPI, 3 mo, 6 mo

^aBL = baseline, IPI = immediately post-intervention (1 month), 3 mo = 3-month follow-up, 6 mo = 6-month follow-up.

Phase 1: Formative Work.

We conducted in-depth interviews with 20 HIV-positive PWID, 18 of their intimate partners (11 HIV-positive and 7 HIV-negative), and 7 Almaty City AIDS Center healthcare providers to understand barriers and facilitators of ART adherence and attitudes and preferences for different ART monitoring methods. In order to be eligible for participation, for HIV-positive PWID, participants had to (1) be at least 18 years of age, (2) have injected drugs in the past year, (3) have been diagnosed HIV-positive, and (4) fluently speak Russian. For partners of HIV-positive PWID, to be included in the study, they had to (1) be at least 18 years of age, (2) have been in a relationship with an HIV-positive PWID for at least 6 months, and (3) fluently speak Russian. For healthcare providers, participants had to (1) be at least 18 years of age, (2) have worked at the AIDS Center as a healthcare provider for at least 6 months, and (3) fluently speak Russian. HIV-positive PWID participants and their partners for Phase 1 were recruited through a nongovernmental organization (NGO) that works with PWID and people living with HIV (PLWH). This NGO has an office located in the Almaty City AIDS Center where individuals can go to receive additional information and support about HIV and drug use. Individuals visiting the NGO were informed about the study by trained staff members, and informational flyers were also available in the office. Healthcare providers were recruited from the Almaty City AIDS Center. Providers were informed about the study by research staff, and providers who expressed interest in participating were scheduled for an interview.

Interviews were conducted in a private room in the field office of the Global Health Research Center of Central Asia (GHRCCA) or a private room at the Almaty City AIDS Center. A trained research assistant (RA) explained all study procedures and obtained written informed consent from the participants. Interviews were conducted in Russian and lasted approximately 45–90 min. All interviews were digitally recorded and transcribed verbatim into Russian. No personally identifiable information was contained in the transcripts. Interview questions followed a guide with core questions and prompts that covered specific topics related to participants’ history with ART adherence or providers’ perception of patients’ ART adherence, barriers, and facilitators to ART adherence, and attitudes toward and willingness to use different ART adherence measures (Davis et al., 2018, 2019; Sarang et al., 2013; Wolfe, 2007). Participants were reimbursed 6,000 tenge (~$15 USD) for travel and time spent.

Data were analyzed using NVIVO (version 11, Doncaster, Australia). Transcripts were analyzed using thematic analysis (Guest et al., 2012). Three members of the research team (the principal investigator and two trained RAs) developed a set of preliminary descriptive codes based on an initial review of the transcripts and pre-identified domains. Each trained qualitative analyst coded the same two interview transcripts in Russian using these codes to ensure reliability. Any discrepancies in coding were discussed between the three analysts and resolved by consensus. Once reliability was established and the codebook finalized, all transcripts were divided between the three analysts for coding. Excerpts were selected to illustrate the main findings and expand upon the selected themes (Results of Phase 1 presented in Davis et al. (Davis et al., 2019)).

Phase 2: Intervention Adaptation.

We adapted KZ SMART Couples from the original SMART Couples intervention (Remien et al., 2005) using international collaboration and CBPR methods. The original SMART Couples intervention was developed in the United States as a couples-based HIV medication adherence intervention and comprises four sessions guided by Social Action Theory (Ewart, 1991; Remien et al., 2005). The intervention originally consisted of four sessions approximately an hour in length delivered by a nurse practitioner over a 5-week period (Remien et al., 2005). The sessions involve psycho-education and structured discussions as well as problem-solving and couple-communication exercises (e.g., model problem-solving process, tell partner things you like about them, set goals) (Remien et al., 2005). Key components target (1) the importance of medication adherence to avoid the development of the resistant viruses and maintain health; (2) identification of non-adherence patterns; (3) partner support, communication, and problem-solving strategies to overcome adherence barriers; and (4) partner’s confidence in achieving and maintaining improved adherence. Another key component of the intervention included tracking patients’ adherence using an EMD (Medication Event Monitoring System Cap—MEMS Cap) and discussing adherence challenges and patterns during the sessions (Remien et al., 2005).

Given cultural differences between the United States and Kazakhstan and the differing needs of PWID compared to people who do not inject drugs, we sought guidance and insight from the local community to inform the adaptation of SMART Couples for Kazakhstan. The adaptation process included multiple focus groups conducted with the CCG composed of four HIV-positive PWID and their intimate partners to provide feedback on the intervention content and identify areas that needed to be modified. The inclusion criteria for the focus groups were the same inclusion criteria as those for HIV-positive PWID and their partners in Phase 1. The moderator and assistant moderator took field notes during all focus groups to record information about important contextual issues, such as group dynamics and body language. Each focus group was audio-recorded and transcribed into Russian. A transcript-based analysis was conducted by two coders (principal investigator and project director) using an a priori list of codes and allowing for emergent themes. Important areas of adaptation identified during the coding process were implemented into the intervention manual, and the updated version of the manual was presented in future focus groups. We also received feedback on intervention adaptation from a CAB consisting of behavioral and social scientists from the United States and Kazakhstan, local clinical providers from the Almaty City AIDS Center and drug treatment clinic, HIV-positive PWID, and NGO staff members who work with PWID and/or PLWH.

Initial meetings of the CAB defined local contextual issues, such as the need for culturally tailored intervention materials, integration of substance use information and resources, identification of appropriate intervention facilitators, and compact intervention length. Through iterative meetings, the U.S. and Kazakh teams identified critical themes related to ART adherence in Kazakhstan, including key barriers (e.g., substance and alcohol use, psychiatric distress, HIV and substance use-related stigma, and misperceptions regarding treatment) and key facilitators (e.g., social support and problem-solving). Current Kazakhstan national guidelines for ART were also reviewed. This process identi-fied components of the original SMART Couples intervention that needed to be modified for use in Kazakhstan among HIV-positive PWID. The adaptation process resulted in a three-session ART adherence intervention designed to be delivered in person by a psychologist or social worker at the Almaty City AIDS Center. SOC is an appointment scheduled every 3 months at the Almaty City AIDS Center with clinic staff.

Transition to Online Format.

The pilot RCT was scheduled to begin in March 2020. Due to the COVID-19 pandemic, non-essential in-person research procedures were not permitted in Kazakhstan due to both local and Columbia University regulations, and the intervention had to be modified for online delivery. Members of the research team and CAB met remotely to discuss the adaptation of materials into an online format, protocols and guidelines for conducting intervention sessions via video call, technological challenges, and logistics. Intervention facilitators engaged in the practice of online delivery of intervention sessions with research staff members through role-plays. Intervention sessions were then pilot tested among three PWID and their partners. Adaptations to the intervention delivery process were made and additional guidelines were provided based on the results of the pilot tests (e.g., procedures for conducting intervention sessions when couples are on different devices instead of the same device).

Phase 3: Pilot RCT.

We launched a pilot RCT of the KZ SMART Couples intervention in Almaty, Kazakhstan, in October 2020 to determine feasibility and acceptability and obtain preliminary estimates of efficacy in increasing ART adherence. The study team includes U.S. and Kazakhstan researchers and program staff based at the Almaty City AIDS Center. The study is conducted in close collaboration with AIDS Center leadership. All research activities involving human subjects received approval from the Institutional Review Board at Columbia University and Al-Farabi Kazakh National University. Thus far, we have enrolled 43 dyads and 19 dyads have completed the study. Upon completion of the study, findings will be shared with key stakeholders, including CAB members, AIDS Center, NGO, and other organizational staff, and community members.

Intervention Facilitators.

The primary implementers of the KZ SMART Couples intervention are psychologists at the Almaty AIDS Center and local social workers. Although we had initially planned on having AIDS Center nurses deliver the intervention, the CCG and other PWID community members provided feedback during Phases 1 and 2 of the study expressing a strong preference that the intervention be delivered by psychologists and/or social workers. PWID wanted trained professionals who had the expertize to help them with the psychological difficulties associated with an HIV diagnosis and with using drugs and felt that nurses did not have the skills to support them in this area. PWID also mentioned the potential for peers to serve as facilitators since they have shared lived experiences, but PWID also felt peers would not have the training or skills to help them solve psychological challenges and worried about the potential loss of confidentiality. We thus changed our intervention facilitators to match the preferences of the community. All intervention facilitators received training in human subjects protection, as well as intervention protocols.

PWID Participants and Treatment Support Partners.

We are recruiting HIV-positive PWID who have been on ART for at least 6 months, but are not virally suppressed, from the Almaty City AIDS Center. To be eligible for study participation, HIV-positive PWID also need to report injecting any drug in the past year. AIDS Center staff conduct chart review to identify potentially eligible participants, who they contact to ask if they are interested in the study and if they can share the individual’s contact information with research staff for eligibility screening. A trained RA then contacts interested individuals to conduct eligibility screening and informed consent. Eligible PWID are asked to identify a partner (e.g., spouse/intimate partner, family member, or friend) who could provide treatment support and participate with them in the study. Dyads are eligible to participate if (1) both partners are ≥18 years old, (2) both identify each other as their intimate partner, family member, friend, or other person within their social network, (3) the relationship has lasted at least 3 months, (4) both report feeling safe with their partner in the study, (5) neither report any severe physical or sexual violence perpetrated by the other partner in the past year, (6) both are able to provide informed consent and follow study procedures, and (7) both are fluent in Russian. Treatment support partners may or may not be HIV-positive and may or may not inject drugs.

Sample Size.

Power analyses were conducted with G*Power (v3.1.0) (Faul et al., 2007) using repeated measure within-between ANOVA approach and α = .05. We calculated that with 30 dyads per arm, we would have 80% power to detect a medium effect size (f = .15). Thus, we plan to recruit 66 couples with the aim of having 60 for analysis.

Results

KZ SMART Couples Intervention and Implementation Strategies

Intervention and Delivery.

KZ SMART Couples is a dyad-level intervention aimed to improve PWID patients’ adherence to ART by fostering the support of a treatment support partner. The adapted intervention also includes referrals to local harm reduction and substance use services and incorporates materials discussing issues related to substance use and HIV (e.g., the importance of using clean needles, concerns about substances used interacting with ART). Originally designed to be an in-person intervention, the intervention is being administered via Zoom video call to each dyad by a clinical psychologist or social worker. The intervention consists of three 60–75 min sessions to be completed within a 1-month period. The session content includes structured discussions and instruction, as well as specific problem-solving and dyad communication exercises. Facilitators also discuss EMD adherence data and help participants identify times when adherence is more difficult and identify potential causes for lapses in adherence. Key components include education about the importance of adherence to avoid viral resistance and maintain health, identifying patterns of non-adherence, developing communication and problem-solving strategies to overcome adherence barriers, optimizing partner support, and building confidence within the dyad for achieving and maintaining improved adherence (Remien et al., 2005). Participants are reimbursed 4,000 tenges (~$10 USD) per partner for completing each intervention session.

Implementation Strategies

We are also collecting data on implementation outcomes to inform implementation processes for larger follow-up studies. The integration and assessment of implementation strategies and outcomes are illustrated in our implementation logic model based on the design developed by Smith et al. (Smith et al., 2020) (Figure 1). Three of the primary implementation strategies for KZ SMART Couples are:

Figure 1.

Implementation logic model for KZ SMART couples.

Training of Staff.

Clinical psychologists at the Almaty City AIDS Center, local social workers, and GHRCCA staff received intensive training on the study protocols, intervention materials, and delivery procedures. Training of staff was conducted remotely by the project director and trained RAs. We used Zoom to conduct the remote training to help facilitators become comfortable with the new platform for intervention delivery. Staff were provided a brief orientation covering the use of the Zoom platform, audiovisual device setup, and contingency planning for technical difficulties/disruptions. We also disseminated the modified intervention manual to the facilitators ahead of the training to allow time to review changes, submit questions in advance, and/or provide suggestions for further modifications.

Staff were taught the fundamentals of remote intervention delivery to enhance their understanding of the intervention components and methods of delivering them via Zoom. Trainers and staff role-played delivery of the intervention in different scenarios, including how to deliver the intervention (1) when patients are connecting via computer or phone, (2) when the dyad is sharing the same device or when they are located on different devices, (3) when a patient has an unstable internet connection or loses internet connection, and (4) when a patient or facilitator has other technical problems (e.g., sound or video not working). Intervention facilitators practiced delivering the intervention online in role-play scenarios over 1-month period. Based on their training experiences, study protocols were further refined to provide additional clarity and facilitate easier remote delivery of the intervention. A GHRCCA IT staff member was appointed to provide technical support to the intervention facilitators, who can contact IT staff at any point before or during an intervention session to assist with resolving technical problems.

Almaty City AIDS Center nurses were trained in person prior to the COVID-19 pandemic on how to recruit participants and collect dried blood spot (DBS) samples from individuals and store them for future testing. Nurses were provided with an instruction manual for reference. GHRCCA staff are available to answer questions as needed and monitor the collection of samples.

GHRCCA staff were trained in study procedures, including recruitment and enrollment, intervention delivery, collection of biological samples, remote set-up of the EMD, and monitoring and collection of data. Original study protocols were designed to be conducted in person and had to be modified to remote delivery due to the COVID-19 pandemic (although some procedures are now able to be conducted in person again as the world has adapted to COVID-19). Columbia University and GHRCCA investigators met remotely via Zoom to discuss changes in study procedures. A draft of remote study procedures was developed and procedures were pilot tested among three dyads. The study protocol and procedures were further refined based on feedback from pilot testing (e.g., provision of additional follow-up training on using the EMD and assisting with the conduct of surveys by phone).

Supervision.

Ongoing supervision is provided to AIDS Center staff. GHRCCA staff members provide ongoing supervision and training to intervention facilitators in order to build local capacity and ownership. Intervention sessions are recorded and reviewed by a trained and experienced GHRCCA staff member, who provides feedback and guidance to intervention facilitators on KZ SMART Couples components and delivery procedures. Intervention facilitators also complete a self-rating sheet after each intervention session to rate the extent to which key components were delivered and the responsiveness of participants. GHRCCA staff provide individual consultations to facilitators to address questions and provide support as needed. GHRCCA staff also provide ongoing supervision and support to nurses conducting recruitment and blood sample collection procedures. Nurses can contact GHRCCA staff as needed via phone, WhatsApp, or email, and a GHRCCA staff member checks in with the nurses on a bi-weekly basis.

Technical Assistance.

GHRCCA staff are available to respond to technical assistance requests, such as troubleshooting problems with assessment instruments or intervention materials, providing additional training or information on core elements, assisting with technical challenges in intervention delivery, troubleshooting challenges with EMDs, and providing guidance in challenging clinical situations. Technical assistance requests are logged, and modifications are made to study materials as needed.

Assessments

In order to capture outcomes related to the study aims described above, the following assessment tools are utilized:

EMD and Mobile App Data Collection System.

To provide a more objective measure of medication adherence, we are using an EMD system (BETRCap^™) that includes an electronic pill bottle cap that connects via Bluetooth to a mobile app on participants’ phones to monitor ART adherence. The electronic pill cap sends a signal to the app via Bluetooth connection whenever the device is opened as a proxy for ART adherence. Participants can view their own adherence data and history on the BETRCap^™ mobile app on their phones. Data are also uploaded in real-time to an online dashboard system accessible by authorized research staff. If Bluetooth connection problems prevent real-time monitoring, openings are stored and uploaded to the app and dashboard when a connection is re-established.

After enrollment, a package containing a study smartphone (to ensure compatibility with the BETRCap^™ EMD system) with the BETRCap^™ app downloaded and a BETRCap^™ pill bottle are delivered to all HIV-positive participants. After participants receive the package, they complete a call with GHRCCA IT staff to set up and learn how to use the EMD system. IT staff teach participants how to fill the device with pills, open it for dosing, and pair it with the app. Before ending the session, IT staff confirm that the EMD device is properly set up and recognized by the server and that device openings are being recorded properly. Multiple call sessions are provided for participants who need to troubleshoot problems or need additional instruction on how to use the EMD system. If 7 days pass and device openings are not seen at all on the server, the IT staff contacts the participant again to see if there are any device problems.

Participants also complete a brief weekly survey on the BETRCap^™ mobile app. The app contains a self-reported adherence scale (Wilson et al., 2016) and questions about participants’ substance use and mental health. Participants are compensated 500 tenges (~$1 USD) for every weekly survey they complete.

EMD-measured adherence is operationalized as the percentage of days for which a device-opening signal is received in the past month. For self-reported adherence, item responses for the three adherence items will be linearly transformed to a 0–100 scale, with 0 being the lowest adherence and 100 being the highest adherence (Wilson et al., 2016). The mean of the three individual items will then be calculated.

Pharmacological Samples.

Three different pharmacological sample types are collected from participants: DBS samples, hair samples, and urine samples. This is to test the feasibility and acceptability of adherence sampling types and to provide more objective measures of adherence. DBS samples are collected by an AIDS Center nurse from HIV-positive participants who are on tenofovir-based regimens during their regular AIDS Center visits. A nurse collects DBS samples by venipuncture, using a pipette to spot blood directly onto Whatman 903 ProteinSaver cards. The cards are air-dried, individually packaged in a plastic bag with desiccant, and transferred to a −20 to −80°C freezer for storage. Once all samples are collected, samples will be shipped to the University of Colorado Antiviral Pharmacology Lab for assay. Results detect ART adherence over the past couple of days and the past couple of weeks.

HIV-positive participants on any ART regimen collect hair samples at baseline and 1 month after baseline assessment (immediately post-intervention). Participants are given a sample collection kit with required materials and instructions and are also given real-time instructions on sample collection via video call. To collect hair samples, participants cut a small thatch of hair (~50 strands of hair) close to the scalp from the bottom layer of their hair. They then affix their hair to a piece of tinfoil with the scalp end denoted and place it inside a Ziploc bag labeled with their study ID and date of sample collection. Participants then ship samples to GHRCCA research staff. For participants completing in-person assessments, samples are collected by an RA. GHRCCA research staff will batch ship samples to the University of California-San Francisco Hair Analytical Lab for assay once all samples are collected. Results detect ART adherence over approximately the past month.

At each assessment period, HIV-positive participants on tenofovir-based regimens are mailed a self-collection kit and instructions to self-collect urine samples for point-of-care ART adherence testing. Participants notify GHRCCA staff once the sample has been collected, and samples are delivered via courier to the GHRCCA office. Upon receipt of the sample, trained GHRCCA research staff conduct the point-of-care assay according to manufacturer protocols. Results detect whether participants have been adherent to ART regimens in the past 48 hr.

Survey Assessments.

All participants receive the same schedule of survey assessments, which are delivered by an RA via phone call and answers are logged into Qualtrics. Surveys are approximately an hour in length and contain items on sociodemographics, adherence, mental health, substance use, access to healthcare services, social support, and COVID-19 issues. There are four main repeated assessment occasions: baseline “pre-intervention,” one-month post-baseline (immediately post-intervention), and 3-months and 6-months post-baseline.

Chart Review of AIDS Center Records.

Almaty City AIDS Center staff provide GHRCCA research staff with electronic medical record data to determine participant eligibility and assess intervention outcomes (e.g., viral load levels). Electronic medical record data are de-identified and assigned a unique study ID before they are provided to GHRCCA research staff. Almaty City AIDS Center’s Electronic HIV Case Management System (EHCMS) is a government-approved computer system for the collection, storage, transfer, and analysis of epidemiological, laboratory, and clinical data on all registered cases of HIV infection in Kazakhstan. Within this system, a file is created for each client who registers for treatment at the AIDS Center. All visits, tests, results, and treatment information for each client are entered by epidemiology and treatment department staff. Data extraction from this database is done by authorized AIDS Center staff. Data extracted and provided to GHRCCA research staff include (1) HIV status, (2) PWID status, (3) current ART regimen and ART regimens over the past year, (4) length of time on ART, and (5) viral load test results over the past year.

KZ SMART Couples Intervention and Implementation Strategies Delivery Data.

A number of program materials and reporting forms are used to assess intervention delivery and implementation. After each intervention session, intervention facilitators complete a form to self-assess their fidelity to intervention implementation. Facilitators also report on challenges and barriers faced during implementation. Approximately 10% of audio sessions were reviewed by a GHRCCA staff member for quality control and independent verification. At the end of the study, facilitators will also be asked questions about their satisfaction with various aspects of the intervention (e.g., content, complexity, delivery modality, and comfort) to assess provider acceptability. Participants receive acceptability questions during the 6-month survey/exit interview assessment.

Research staff complete documentation related to the number of hours of training delivered, feasibility and acceptability of different study procedures, and barriers and challenges to implementation, which will be used to evaluate implementation processes for this study and inform implementation procedures for future studies. GHRCCA staff also track technical assistance requests about challenges faced by participants or staff during study implementation. Reports encompass what technical assistance is needed to support implementation, requests for clinic support, challenges with electronic devices, content clarification, skills-building, and requests for additional information or training. Technical assistance data include who asked for assistance, what assistance was requested, and how assistance was provided.

Policy and Environment Monitoring.

GHRCCA research staff document contextual changes that may impact study implementation, including new protocols or policies regarding HIV, substance use, or other related services, new domestic or international funding sources for HIV-related programing (or the cessation of funding), regularly scheduled activities of the AIDS Center or other organizations that may conflict with study programing, and trends in staff turnover. Research staff regularly seek information about such changes and record this information. Changes are reviewed by the study team to discuss how they may affect ongoing study activities.

Quality Assurance.

GHRCCA staff monitors intervention progress and fidelity on an ongoing basis through a review of supervision records and recordings. The research team monitors data collection on an ongoing basis to observe the accuracy and challenges of the different assessments and troubleshoot potential problems that may arise. Data are also monitored for completion to minimize participant loss to follow-up. If assessments are not completed within a timely manner, staff follow up with participants to remind them to complete their assessments. During each reporting period, chart review reports from the AIDS Center are checked for accuracy.

Discussion and Application to Practice

The KZ SMART Couples intervention offers a systematic and collaborative approach that can be delivered by social workers and clinical psychologists to help HIV-positive PWID work with a support partner to improve their adherence to ART and connect to substance use services. Intervention activities are interactive and can be customized to individual dyads as needed. This trial has a number of strengths, including the implementation of a social support intervention, a multi-faceted and comprehensive data collection system that includes biological and socio-behavioral measures, the ability for remote delivery, feedback from local community members and healthcare providers, and multiple implementation strategies.

The study is ongoing and a full data analysis will be conducted upon study completion to assess feasibility, acceptability, efficacy, and implementation outcomes. However, initial, anecdotal findings suggest high acceptability and feasibility by patients, partners, and providers. Clinic staff have reported that many patients who participated in the intervention now come to the AIDS Center with their partners for their appointments, even after completion of the intervention and participation in the study. Patients and partners report improved communication with each other and receiving greater social support. Intervention facilitators have reported that the intervention is easily incorporated into their regular clinical practice schedules without much disruption. Facilitators report wanting to use intervention content in their standard practice once the study is over.

There have been a number of challenges to intervention implementation, primarily as a result of the COVID-19 pandemic onset immediately prior to launching the RCT phase of the study. Not all of our intervention facilitators had previous experience conducting remote intervention sessions. Many of the participants and some intervention facilitators had never used Zoom before, so additional time was required for training. Furthermore, it was more difficult to train participants on how to use the EMD system via video call than during an in-person session. It has not been uncommon for participants to have an unstable internet connection, which sometimes makes it difficult to conduct video calls. Facilitators sometimes are required to use Zoom for video and WhatsApp for audio. Conducting intervention sessions with partners who do not live together is also sometimes challenging due to the need for multiple stable internet connections. Additionally, 16 phones in our study have been lost, stolen, or sold (with some participants experiencing multiple losses), which results in pauses in data collection for some participants. Social workers and clinicians aiming to implement this intervention into their practice or organization should be aware of these technical challenges and allow for additional time and effort to complete extra technology training, develop contingency plans for unstable internet access (e.g, audio-only calls, provision of internet services for low-income clients), plans for managing couples in separate locations, and plans for lost or stolen phones. Furthermore, although anecdotal findings suggest high acceptability, it is not clear how long any effects of the intervention will persist and how quickly participants and providers will drift from using skills they learned in the intervention. Future booster sessions or other follow-ups may be needed to encourage sustained implementation of intervention strategies over time.

There have also been challenges related to the contextual environment in which PWID in Kazakhstan live. Because the sale, transportation, and use of illicit drugs are illegal in Kazakhstan, a number of our participants have been arrested during the study and are in and out of jail. The provision of ART medications in the criminal justice system is highly inconsistent, and many individuals in the criminal justice system are unable to access adequate ART medication while they are imprisoned, which can impact their viral load levels and have severe adverse consequences on their health. Social workers and clinicians who work with PWID or other clients who have a high risk of interaction with the criminal justice system should develop plans with their clients to maintain access to their necessary medications in the event of arrest. Additionally, few participants use metha-done-assisted therapy and harm reduction services. Sadly, a few study participants have died due to a fatal drug overdose or other health issues (e.g., stroke), illustrating that this population is faced with a number of health challenges beyond ART adherence that need to be addressed. In countries where naloxone is available, social workers and clinicians should try to help PWID clients obtain naloxone to reduce the risk of overdose.

In summary, KZ SMART Couples is a promising social support intervention to improve ART adherence among HIV-positive PWID. Initial data preliminarily suggest acceptability and feasibility among patients and providers. If formal analysis of the full dataset reveals similar results and shows efficacy, the intervention could potentially be scaled across Kazakhstan and other Central Asian countries. Given the disproportionately high and rising rates of HIV among PWID in the region, there is a critical need to improve ART adherence among this population in order to improve health outcomes and reduce the spread of HIV. This study provides valuable insights into leveraging social support for ART adherence, remote delivery of interventions in Kazakhstan, and novel methods of measuring ART adherence in the region. KZ SMART Couples is a promising intervention that could be implemented by social workers and psychologists in resource-constrained settings. The intervention helps social workers and psychologists explain how HIV works, how medications help control the virus and lead to better health, and how substance use relates to HIV transmission and HIV care. It facilitates a review of patients’ medication regimens and enables counselors to correct patients’ misunderstandings about their treatment. The emphasis on treatment support partners and problem-solving maximizes the potential of the intervention to empower patients to overcome barriers to adherence. The EMD system allows patients to monitor their own adherence progress and health details (e.g., treatment regimen, prescription refill schedule, and healthcare provider contact information). Given that substance use is often a barrier to adherence, the intervention also incorporates substance use information, including referrals to local substance use services. The use of readily implementable social support interventions, such as KZ SMART Couples, in social work practice could improve individual and public health outcomes, particularly in LMIC and other resource-constrained settings.