Advancing behavioral interventions for African American/Black and Latino persons living with HIV by integrating critical race theory, harm reduction, and self-determination theory: A qualitative exploratory study


 Background
 . Rates of participation in HIV care, medication uptake, and viral suppression are improving among persons living with HIV (PLWH) in the United States. Yet, racial/ethnic disparities in engagement along this HIV care continuum are persistent, particularly among African American/Black and Latino PLWH, signaling the need for novel approaches. We created a new conceptual model that integrates critical race theory, harm reduction, and self-determination theory, which was used to plan an intervention optimization trial to test five behavioral intervention comcompponents. The present qualitative exploratory study describes the integrated conceptual model (ICM) and explores participants’ perspectives on contributions of the ICM to study acceptability, feasibility, and impact.
Methods
 . Participants were African American/Black and Latino PLWH poorly engaged in HIV care and with non-suppressed HIV viral load in New York City. From the parent optimization trial (N=512), we randomly selected 46 for in-depth semi-structured interviews on experiences in the optimization trial. Interviews were audio-recorded and professionally transcribed verbatim, and data were analyzed using directed qualitative content analysis.
Results
 . On average, participants were 49 years old (SD=9) and had lived with HIV for 19 years (SD=7). Most were male (78%) and African American/Black (76%). All had taken HIV medication previously. Challenging life contexts were the norm comprised of poverty, poor quality/unstable housing, trauma histories exacerbated by current trauma, medical and mental health comorbidities, and substance use. Participants found the optimization trial highly acceptable. We organized results into four themes focused on the importance of 1) being understood as a whole person and in context; 2) trustworthiness and trust; 3) opportunities for self-reflection; and 4) support of personal autonomy. The salience of nonjudgment was prominent in each theme. Their experiences and perspectives reflected the ICM. Participants reported these characteristics were generally lacking in HIV care settings.
Conclusions
 . The ICM for is a new useful model for research with AABL-PLWH with potential to enhance study acceptability, feasibility, and intervention effectiveness. Future application of the ICM is warranted to reduce racial/ethnic disparities. In a future study, we will explore participants’ perspectives on the specific intervention components in which they engaged and ways components can be improved.

Second, harm reduction is a conceptual framework and set of practices that focus on the minimization of the physical, social, and legal harms that may affect people who use drugs and to society as a whole as a result of drug use [38]. At its core, harm reduction supports the dignity and autonomy of people who use drugs without judgment. Harm reduction is also a movement for social justice built on a belief in, and respect for, the rights of people who use drugs [39]. A harm reduction perspective allows clinicians and drug users to work together to establish goals and objectives to reduce drug-related harm, based on the notion of an individual's right to self-determination [40]. Harm reduction approaches have been found acceptable and effective in numerous past studies [41]. Because harm reduction is a way of viewing the impact of drug use on a person's psychosocial functioning, that is, harm reduction is a perspective, it is not bound to any one behavioral change technique or therapeutic theory of action [42].
Harm reduction was originally developed as an approach to reduce drug-related harms. More recently the harm reduction perspective has been applied to other types of public health concerns, such as mental health distress [43]. In the present study, we applied harm reduction to a range of behaviors, including HIV management. Indeed, PLWH's health behaviors exist within a constellation of what can be construed as potential harms to themselves and society, such as declining or taking long breaks from HIV medication, or idiosyncratic HIV medication dosing schedules (e.g., not taking medication on weekends). HIV and substance use treatment are similar in that treatment expectations are commonly absolute, leaving little room for individual autonomy. For example, the Centers for Disease Control and Prevention (CDC) recommends that PLWH initiate HIV medication immediately after diagnosis and then take "every dose, every day" to sustain HIV viral suppression [4]. Similarly, complete abstinence from substance use has historically been the predominant goal in most treatment settings [18]. While the present program of research certainly recognizes the value of taking HIV medication to achieve sustained HIV viral suppression for individual and public health bene t, it also recognizes great variability in PLWH's approaches to HIV management, and the need to discuss and explore those personal decisions to maximize health and wellbeing. Yet, messages consistent with harm reduction appear relatively uncommon in HIV treatment settings [44]. However, perfect adherence to HIV medication is not required to achieve HIV viral suppression [4]. Current regimens are highly effective and PLWH can achieve viral suppression with 80-90% adherence, or even lower depending on the regimen [4]. But, this more pragmatic approach is not yet incorporated into the CDC's de nition of adherence ("every dose, every day"), and the CDC exerts a powerful in uence on HIV care system policies and provider behavior. Thus, in the ICM we extended the harm reduction approach to include individuals' health care and behavioral decisions about whether or how often to take HIV medication, and how to manage other aspects of their lives and relationships in ways that might reduce or eliminate harms to the self or others, including regarding substance use.
Self-determination theory is a macro theory of human motivation and personality that concerns the innate and fundamental needs for autonomy (people need to feel in control of their own behaviors and goals), competence (people need to gain mastery of tasks and learn different skills), and connection or relatedness (people need to experience a sense of belonging and attachment to other people) [45]. The most volitional and highest quality forms of motivation emerge when these three needs are supported by the larger environment, and self-determination theory proposes that the degree to which any of these three psychological needs is unsupported or thwarted within a social context will have a robust detrimental impact on wellness in that setting [46]. The ways in which AABL-PLWH's self-determination is commonly not fostered in society and social service and health care settings are described above.
Self-determination theory is an accepted theoretical underpinning of motivational interviewing [47]. Motivational interviewing is an evidence-based directive and collaborative counseling approach for behavior change that elicits participants' values, perspectives, and questions, identi es ambivalence and discrepancies, and corrects misinformation with permission, to thereby foster durable intrinsic motivation and readiness for change [47,48]. In reviews and metanalyses, motivational interviewing interventions have been found effective at clinically signi cant levels for a range of health behaviors [49][50][51]. Motivational interviewing has been found to be particularly effective with AABL populations compared to White populations [49]. As a non-coercive, strengths-based, and autonomysupportive approach, it may have utility in particular when health beliefs and emotions such as distrust/fear impede behavior change [25,31,52]. The intervention components tested used the motivational interviewing counseling approach (described below).

Development of the ICM
During the planning phases of the optimization trial, our research team formed an intervention working group and carried out an iterative intervention mapping process that included a review of the literature on barriers to the HIV care continuum for AABL-PLWH as presented above, including our own past studies [14,25,27,31], and existing theories and models [35,53]. We identi ed gaps in theory, models, and interventions for this population. In particular, as noted above, structural and cultural competency are generally insu cient in existing behavioral interventions for AABL-PLWH [6, 7,54]. Cultural competency in health care describes the ability of systems to provide care to clients with diverse values, beliefs, and behaviors, including the tailoring of health care delivery to meet patients' social, cultural, and linguistic needs [55]. Structural competence is de ned as the trained ability to discern how issues de ned clinically as symptoms, attitudes, or diseases, such as medication "non-compliance," trauma, depression, or smoking, represent the downstream implications of a number of upstream decisions that shape powerful factors such as health care delivery systems, zoning laws, and urban and rural infrastructures [54]. The intervention working group took the stance that cultural and structural salience in intervention content and cultural and structural competency among staff are essential to address these racial/ethnic disparities. Thus, to address the complex multi-level barriers that AABL-PLWH experience to the HIV care continuum and improve the acceptability, feasibility, and clinical effects of behavioral intervention components, the ICM combines critical race theory, harm reduction, and self-determination theory. Critical race theory was selected to highlight structural barriers to health, and the roles of systemic racism, counter-narratives, and resilience in health behavior for AABL-PLWH. Harm reduction was selected to underscore the importance of a non-judgmental and non-coercive approach to any positive behavior change (related to HIV, substance use, or any other health behavior). Self-determination theory was selected in light of the need for autonomy supportive approaches among AABL-PLWH. When taken together, the three theories/approaches potentially have areas of congruence and complementarity, and each theory or approach has aspects that may strengthen the others, potentially combining synergistically to create a new and useful model (Fig 1). In the center of Fig. 1 we present the primary concepts and domains identi ed during the process of integrating the three theories/approaches, and the boxes on the right and left of the gure comprise secondary, but important domains that bridge two of the theories.
Goals of the present study. Although critical race theory, harm reduction, and self-determination theory are well-represented in the research literature, to our knowledge these three theories/approaches have rarely, if ever, been integrated and applied in prior intervention research. Yet, when taken together, the three theories/approaches have the potential to create a new model that can be applied to health equity research. The present study explores participants' experiences in the parent optimization trial, which was grounded in the ICM, with an emphasis on uncovering and describing the ICM's contribution to, and how the ICM contributed to, study acceptability, feasibility/engagement, and potential effects. Study ndings will inform future research efforts for populations that experience health inequities. In a subsequent manuscript we will explore the acceptability, feasibility, and potential impact of the speci c intervention components tested in the parent intervention optimization trial and ways they can be improved from the participants' perspectives.

Overview
The present study took an ethnographic/phenomenological qualitative approach [56] and used semistructured in-depth interview data collected as part of the parent intervention optimization trial. The parent study, which had the eld name "Heart to Heart 2" was carried in New York City between 2017-2021 [35,36]. As noted above, the parent study focused on AABL-PLWH with socioeconomic disadvantage who were poorly engaged along the HIV care continuum, speci cally, those who did not engage in HIV care at recommended levels and who evidenced both poor adherence to HIV antiretroviral therapy and non-suppressed HIV viral load. The parent study, described by Gwadz and colleagues [35], used a fractional factorial experimental design to test the e cacy of ve separate intervention components grounded in the ICM. It used a 2 5-1 design and was comprised of 16 experimental conditions. Participants were randomly assigned to one of the 16 experimental conditions, each of which was comprised of a unique combination of intervention components (typically participants were assigned to receive 2-4 components). HIV viral suppression assessed by laboratory report was the parent study's primary outcome. The optimization objective for the optimization trial was cost-effectiveness and thus its ultimate goal of the optimization trial was to identify the most cost-effective combination of intervention components to improve rates of HIV viral suppression in this population. To date, participants have engaged in the parent study's intervention components, and the intervention optimization process is ongoing. In addition to the main qualitative results, we also present descriptive quantitative data on participants' sociodemographic and background characteristics and on study acceptability. The parent optimization trial was registered with ClinicalTrials.gov (NCT02801747). Participants gave signed informed consent for study activities. The study was approved by the Institutional Review Board at the New York University Grossman School of Medicine.
Eligibility criteria for the parent study The parent study's inclusion criteria were 1) age 18 -65 years; 2) African American or Black race and/or Latino/a or Hispanic ethnicity; 3) HIV diagnosed for at least 6 months; 4) HIV antiretroviral therapy adherence less than 50% in the past six weeks and detectable HIV viral load based on a laboratory report; 5) sub-optimal engagement in HIV care (operationalized as less than one visit in every four month period in the past year or > two missed visits without prior cancellation in the past year; 6) resides in the New York City metropolitan area; 7) able to conduct research activities in English or Spanish; 8) willing to provide a blood specimen at screening to assess HIV viral load; and 9) willing to be randomly assigned to 1-5 intervention components. Participants were found eligible for the parent study if they met all the inclusion criteria.

Staff and training
Study staff were diverse with respect to age, gender, race/ethnicity, and HIV status and had bachelor's or master's degrees in the social sciences (e.g., social work, public health, and psychology). All had past research experience with the population of AABL-PLWH. To promote high rates of engagement in the study, all staff members were trained in the ICM and the study ethos that aligned with the ICM which emphasized respect for autonomy and personal decisions regarding HIV medication and on providing an overall high-quality experience (e.g., staff members remembering participants' names, exibility in rescheduling, prompt compensation, acknowledgement that participants' time is as valuable as staff time, and refreshments provided in the waiting area and during sessions or groups). Thus, staff were trained to be both structurally and culturally competent to foster engagement and trust and foster the behavior change process.
Brief description of the components. The intervention components were Component A: motivational interviewing sessions was designed to speci c motivational interviewing techniques to address salient, including culturally salient health beliefs (e.g., outcome expectancies, self-e cacy, medical distrust) and emotions (e.g., concerns/fears of HIV medication) to foster durable intrinsic motivation for behavior change. Component B: pre-adherence skill building was designed to help participants build behavioral skills to manage HIV medication adherence such as habits while attending to cultural and structural factors that can impede adherence (e.g., lack of private living space). Component C: peer mentorship was facilitated by a "successful" peer mentor (i.e., a PLWH demographically similar to study participants who had consistently engaged in care and was taking HIV medication with high levels of adherence) and sought to provide peer modeling of and shape peer norms regarding HIV management (primary goals), and provide social support and combat stigma (secondary goals). Peer mentors elicited and addressed the types of barriers AABL-PLWH experience to engagement. Component D: focused support groups were designed to provide social support and reduced stigma regarding care and HIV medication use, including culturally salient factors that impede engagement such as medical distrust and fear. Component E: navigation was designed to identify and ameliorate structural barriers to care and HIV medication. In this factorial design, all components had two "levels." Components A-D's levels were off/on (participant did not receive/participant did receive), and Component E's two levels were short navigation (three months) vs. long navigation (six months). The study overall emphasized non-judgment and the project's tag line, found on the project logo and recruitment materials, was "No pressure, no judgment." Thus, the project took an active and explicit approach to non-judgment.
Procedures for the parent optimization trial Recruitment into the parent optimization trial. The recruitment approach for the parent study comprised a hybrid sampling strategy that included peer-to-peer recruitment; direct recruitment by study staff members in HIV service, HIV housing, and other community-based organizations; and advertisements in a local free newspaper. Peer recruitment was the primary sampling approach. Peer recruitment was tracked with a coupon system that linked the recruiter to the recruit, and recruiters received modest compensation for recruitment ($15/recruit). Most enrolled participants were recruited by peers (75%); 9% were recruited through newspaper ads, and 16% through other means.
Screening, enrollment, intervention activities, and follow-up assessments Participants were screened for eligibility after providing informed consent. Screening included assessment of HIV viral load levels via laboratory report obtained from a commercial laboratory. Those found eligible for the parent study provided signed informed consent, and completed a structured baseline assessment battery lasting 60-90 minutes. The baseline was conducted in the Research Electronic Data Capture (REDCap) platform. REDCAP is a cloud-based platform for data capture designed for clinical research [57,58]. After completing the baseline assessment, participants were randomly assigned to an intervention condition. Participants were randomly assigned to one of 16 different experimental conditions, each comprised of a different combination of the ve intervention components [35]. Most experimental conditions were comprised of 2-4 intervention components. All participants received a core health education session, the standard of care. The intervention components were individualized and exible, and did not assume that participants wished to or were ready to take HIV antiretroviral therapy at the present time. The period during which participants engaged in intervention activities ranged from 4 to 8 months. Regarding the timing of intervention activities, navigation was provided rst, to begin to address structural barriers to engagement along the HIV care continuum (all participants received either 3 or 6 months of navigation). Counseling sessions were scheduled next for those randomly assigned to receive that component. Peer mentorship and focused support groups could be administered in the same time period (although not on the same days), and pre-adherence skill building was provided last for those assigned to receive it.
Participants received follow-up assessments and 4-, 8-, and 12-months post-baseline, comprised of a structured interview and HIV viral load test at a commercial laboratory (at 8-, and 12-months). Thus, participants were enrolled in the Heart to Heart 2 study for 12 months. In-person study activities took place in con dential o ces at a project eld site in lower Manhattan in New York City. Participants were compensated $15 for a screening interview, $15 for providing the blood specimen for HIV viral load testing, $25 for baseline and follow-up assessments, and $25 for each intervention session or activity, along with funds for local round-trip public transportation.

Effects of the COVID-19 pandemic on the parent study
The rst case of COVID-19 was diagnosed in New York City on March 1, 2020. On March 12, 2020, inperson activities with human subjects were suspended at New York University, although virtual activities could continue with IRB-approval. At this point in the study, 241/512 (47%) had completed participation in the study, with the remainder still scheduled to attend intervention activities and/or follow-up assessments. With the exception of focused support groups, which were not feasible given participants' lack of smartphone and computer access [59], intervention components and follow-up assessments were carried out in a virtual format. Because we could not escort participants to a commercial laboratory because travel to laboratories was restricted, we requested that participants provide a recent laboratory report from their clinic for which they would be compensated $30. Yet because early in the pandemic HIV services and travel were disrupted due to the public health order to remain at home, it was commonly challenging for participants to carry out laboratory tests and obtain reports. The COVID-19 pandemic resulted in some delays in engagement in study activities and reduced the proportion that provided HIV viral load results modestly.
Procedures and materials for the present study Selection of participants for qualitative interviews. A total of 2-4 participants from each of the 16 experimental conditions were randomly selected for two qualitative, semi-structured, in-depth interviews, one early in the study (within 5-7 months of enrollment) and another at study completion. The rst qualitative interview focused mainly on participants' general experiences living with HIV and with engagement along the HIV care continuum, and the second on perspectives on the parent study's activities. Random selection for the qualitative interview, rather than a selection method more common in qualitative research such as purposive sampling, was intended to serve the needs of the parent study. Nonetheless, random selection and purposive sampling for maximum variability have similar goals; namely, variability in participant characteristic and experiences. The qualitative interviews sought to understand participants' experience in the range of intervention conditions and who received a range of intervention components. The qualitative interviews were audio-recorded and professionally transcribed verbatim. Individuals were compensated $25 for each qualitative interview, along with funds for local round-trip public transportation. A total of 46 participants engaged in the rst qualitative interview and 32 of these also completed the second interview (thus 70% participated in both interviews). Interviews were conducted in-person at the study eld site prior to the COVID-19 restrictions, and on the phone after restrictions on in-person activities were implemented.

Qualitative semi-structured interview guide
The in-depth interviews were guided by a semi-structured interview guide developed by the research team, which included experts on AABL-PLWH and the HIV care continuum. The guide was based on a review of the literature and guided by the ICM. The guide was pilot tested prior to administration and re ned. It was structured as a series of questions and prompts, starting with more general questions and moving to more speci c ones. Throughout the interview process, the interview guide was updated to re ect newly emergent concepts (e.g., feeling pressured to take HIV medication and its effects). The guide had four sections: 1) general experiences with the study (e.g., To start off, what was it that led you to agree to participate in the study? What stands out to you most about the project so far? What have you like? Disliked? Have you been involved in similar projects? How is this study similar/different? What do you think about the staff in general?; 2) emotional or behavioral effects of study participation or recent changes concurrent with study participation, if any (e.g., Have you taken HIV medications since you joined the Heart to Heart 2 study? It's OK if you haven't. We just want to understand what's going on with you now. Why or why not?; What factors played a role in your deciding to take HIV medications at this time, whether related to the Heart to Heart 2 study or other factors?, Since you've been involved with the study, has anything changed about the way you think about HIV medication?); 3) exploration of the speci c intervention components the participant was randomly assigned to receive (e.g., What do you remember about this component?, Can you tell me what kinds of things you discussed? Describe whether this component useful to you or not useful?); and 4) ways the study could be improved (e.g., We want to ask a few more questions about how we can improve. What do you think should be included in the Heart to Heart 2 study that wasn't included?). The present study focuses on data from the rst two sections of the guide, and data collected in response to the second two sections will be presented in a future study.
The interview guide is provided as Supplemental Material.

Qualitative Data Analysis
The strategy used a directed content analysis approach that was both inductive and theory-driven [60]. First, a primary researcher trained in medical anthropology analyzed interview transcripts and developed an initial start-code list and operational de nitions for each code, informed by the theoretical and conceptual perspectives guiding the study [61]; namely, the underlying ICM approach and the theory of triadic in uence, which organizes barriers to and facilitators of engagement along the HIV care continuum as structural-, social-, or individual/attitudinal-level in uences. Thus, codes included those related to culture and race/ethnicity (e.g., experiences of discrimination, medical distrust, counternarratives), substance use management, and autonomy, competence, and relatedness, as well as about other factors that promote or impede engagement along the HIV care continuum (e.g., housing, mental health distress, and poverty). Then, the primary analyst coded approximately 20 transcripts using the start-code list. Next, two additional trained qualitative researchers coded a subset of the interview transcripts and met frequently with the primary data analyst. Codes were further re ned and elaborated upon, and discrepancies were resolved by consensus. After resolution of discrepancies, each transcript was then recoded using the nal coding frame. Then, in an iterative process and in collaboration with an interpretive community made up of members of the research team, codes were combined into larger themes and sub-themes [62, 63].
Regarding positionality and methodological rigor, the research team was made up of men and women from White, African American/Black, Asian, and Latino/a backgrounds. The primary data analyst was a member of the research team trained as a medical anthropologist and experienced with HIV research, including with this subpopulation of AABL-PLWH. Positionality challenges related to sex, gender, race/ethnicity, power, health, socioeconomic status, and privilege were intentionally addressed throughout the data collection process through re ection and training, which focused on how these types of issues might impact the interviewing process and data analysis [64,65]. Although we used the random sampling method for the qualitative interview based on the demands of the parent study, we attended to issues of maximum variation in sample characteristics [66] as one aspect of trustworthiness [67]. Methodological rigor of the analysis was further maintained through an audit trail of process and analytic memos and periodic debrie ng with the larger research team, which included PLWH and experts in long-term HIV survivorship and HIV medication adherence, as well as member checking with AABL-PLWH; feedback from the member checking was incorporated back into the results [56].

Quantitative measures
We assessed age, sex assigned at birth, gender identity, sexual minority status (i.e., identi es as gay, lesbian, bisexual, queer, or other non-heterosexual identity), race/ethnicity, housing status, history of incarceration (yes/no), indications of extreme poverty (how often unable to pay for necessities in the past year and food insecurity) with structured instruments developed for populations in high-risk contexts [68]. The Adverse Childhood Experiences Scale-revised (ACES-R) 14-item scale was to assess early life experiences such as peer victimization, neighborhood disorder, physical abuse, neglect, and sexual abuse [69]. We used a version of the HIV Cost and Services Utilization Study (HCSUS) [70] instrument to assess years since rst HIV diagnosis; years since rst initiated ART; number of months since last HIV medication dose (if not on HIV medication at screening). HIV viral load was assessed with a laboratory report and suppressed viral load was coded as < 200 copies/mL. Substance use patterns were assessed by the World Health Organization Alcohol, Smoking and Substance Involvement Screening Test (WHO ASSIST) [71]. Using established thresholds, symptoms of depression were assessed with the Patient Health Questionnaire depression module (PHQ-9) and coded as likely depression (yes/no) [72]. The Generalized Anxiety Disorder scale (GAD-7) was used to assess symptoms of anxiety and coded as coded as likely anxiety (yes/no) [73]. The Primary Care PTSD Screen to assess symptoms of PTSD, coded as likely PTSD (yes/no) [74]. Study acceptability was assessed using the 12item Client Satisfaction Survey [75]. Feasibility was de ned as the proportion of participants who completed study assessments.

Quantitative data analyses
We used descriptive statistics to summarize socio-demographic and background characteristics and study acceptability.

Results
Participants' sociodemographic and background characteristics are found in Table 1. Participants were 49 years old, on average (SD=9 years). Most (78%) were assigned male sex at birth. Approximately a third (33%) were sexual and/or gender minorities. The majority (76%) were African American or Black and the remainder were Latino. Rates of adverse childhood experiences ranged from 0-14 (mean=4, SD=3 experiences). Indications of low-socioeconomic status and extreme poverty were common: Only 17% were employed, nearly half (46%) had run out of funds for necessities at least monthly in the past year, and most (85%) experienced food insecurity often or sometimes in past year. Less than half (52%) were not stably housed. Participants had been diagnosed with HIV 19 years ago, on average (SD=7 years). All had taken HIV medication in the past. The longest duration of sustained HIV medication use was 45 months (SD=63 months). Current substance use was common: approximately half (54%) used alcohol at a moderate-to-high-risk level based on The World Health Organization ASSIST measure scoring criteria, 61% used cannabis at a moderate-to-high-risk level, and 63% used cocaine or crack use at a moderate-tohigh-risk level. Less than 10% injected drugs in their lifetimes or the past three months. Most (78%) engaged in substance use treatment in the past. Approximately one third of the sample or less reported likely depression, anxiety, or post-traumatic stress disorder (PTSD). A total of 40% evidenced suppressed HIV viral load assessed via a lab report at the 8-and/or 12-month follow-up period. In Table 2 we present participants' acceptability ratings of aspects of the study overall at the nal followup assessment for the entire sample. Acceptability ratings were high (> 70%). Regarding feasibility, despite disruptions due to the COVID-19 pandemic as described above, assessment follow-up rates were high: 83.4% completed the 4-month, 81.1% completed the 8-month, and 80.7% completed the 12-month follow-up assessment.

Overview of results
Participants described managing a con uence of recurring challenges and crises, all exacerbated by chronic poverty, throughout the time they were enrolled in the Heart to Heart 2 project. These commonly included homelessness and/or poor quality or unstable housing, relationship instability, involvement with probation and parole systems, histories of trauma exacerbated by current trauma, underlying physical health comorbidities in addition to HIV, challenges with substance use management and in some cases, hazardous substance use, the need to sell HIV medication to meet basic needs, and severe mental health distress. This attention in the present study to structural factors such as poverty and housing is consistent with the ICM which emphasizes structural competence. Nonetheless, even in this challenging context, the majority of participants were still able to re ect on their own strengths and resilience, perhaps re ecting the ICM, which incorporated a strengths-based approach. Further, we found participants maintained both the desire and ability to make changes in health behavior and other aspects of their lives consistent with their own values, including within the "safe space" that they reported the project provided.
Participants overwhelmingly emphasized that the Heart to Heart 2 project provided a space within which they felt welcomed, individually cared for, and, ultimately, within which they were able to re ect on both their emotional and physical health, including their management of substance use, anxiety and depression, personal relationships, HIV, underlying health conditions, and in many cases their willingness, ability, and/or desire to re-initiate HIV medication or to increase HIV medication adherence in order to achieve HIV viral suppression. Even when participants did not elect to initiate HIV medication or increase the number of HIV medication doses they were taking during the project, they reported that engagement in project activities typically resulted in other types of psychosocial, emotional, or tangible improvements.
Consistent with the ICM, participants were encouraged to stay engaged in the study even if they elected not to focus on or work toward the study's primary outcome, HIV viral suppression, without experiencing pressure or judgment from project staff. Generally, participants engaged in frank discussions of their health decisions (e.g., not taking HIV medication) and contextual challenges (e.g., selling HIV medication, substance use) with project staff, including discussing those behaviors not typically considered socially desirable or socially acceptable, which may re ect the ICM and its emphasis on non-judgment, harm reduction, and personal autonomy.
We organized results into the following four inter-related themes: the importance of feeling understood and validated as a whole person and in context; experiences of trustworthiness and trust; opportunities for self-re ection on a range of topics and its effects; and support of personal autonomy and its effects on motivation and decisions. The importance of nonjudgment was prominent in each theme. We also provide ndings on the context of participants' lives in the sections that follow and highlighted the ways the results re ect the ICM. In reporting the qualitative results, we present ndings pertaining to participants' experiences in the project as whole, with some references to speci c intervention components for clarity. Gender-neutral pronouns (they/them/theirs) were used in the sections that follow because we did not assess which pronoun series participants used to describe themselves. We used pseudonyms and changed or obscured identifying details to maintain participants' con dentiality.
The importance of feeling understood and validated as a whole person and in context Participants emphasized the important role that the contexts of their lives, as noted above, and the characteristics of institutional settings they commonly engaged in, had on behavior and wellbeing. First, they emphasized that social isolation was both chronic and extreme. They reported being generally unable to connect with healthcare, mental health, and other service providers in a manner that was nontransactional, non-judgmental, and meaningful, with one exception; namely, support groups at local community-based organizations oriented toward working speci cally with PLWH. Overall, participants felt unwelcome and devalued in health care settings, and were typically hesitant to openly discuss issues such as mental health and substance use with healthcare providers and social service agencies. Participants reported that in many service settings, and particularly in HIV care settings, they frequently experienced feelings of invalidation and lack of individualized care, describing instances wherein they felt like a number, and were treated as "less than human." This typically resulted in an erosion of their intrinsic motivation to prioritize themselves or their health, which in many cases led to feeling less motivated, rather than more motivated, to improve their HIV medication adherence patterns after engaging in health care settings.
In contrast, when asked to re ect upon their experiences during the Heart to Heart 2 project, participants typically discussed feeling understood as an individual, and, in many cases, as a whole person. They commonly reported that the project was one of the rst times they felt viewed in a professional setting as a complex person with well thought-out and worthwhile perspectives and needs. As shown in Fig. 1, the ICM emphasizes, in part, individualized care, participant dignity, and reduction of stigma. Participants indeed did commonly experience relationships with project staff as de-stigmatizing, mutually respectful, possibly dignity enhancing, and, in many cases, caring. This, in turn, fostered a clinical context in which they could explore aspects of their lives they generally experienced as underappreciated, disregarded, and stigmatized in most professional settings, in relative safety. Simone had not taken HIV medication for over a year at the time they enrolled in the study. They described their experience in Heart to Heart 2 as "personalized," in contrast to the typical medical setting, and these personalized experiences contributed to their deciding to access treatment for depression and, sometime later, re-initiate HIV medication.
Hank, a Black, gay, cisgender male in their early 50s, who had been living with HIV for approximately 10 years, contrasted their experiences in the research project with typical health care setting encounters.
They noted a marked lack of social and structural stigma in their interactions with Heart to Heart 2 project staff, consistent with the ICM: Relatedness (i.e., social relationships) is one important aspect of the ICM. Like with Hank, many participants described the ability to develop meaningful, ongoing connections with project staff as a critical part of their experiences with the project and the foundation, in turn, for their being willing and able to address a broad range of HIV medication-related struggles in a holistic manner.
On the other hand, participants commonly expressed frustration that their primary care providers and other social service providers were not better able to see their HIV care and HIV medication adherence struggles in a larger context. In particular, they experienced their providers as "dismissive" of a number of issues participants considered vital, such as housing insecurity, personal relationships, substance use, and legal problems. Participants reported providers appeared to see these types of concerns as peripheral or even unrelated to HIV management. Yet, participants were clear these issues were directly related to HIV care and medication, and the lack of such recognition commonly left participants feeling overlooked, frustrated, or rejected. In contrast, the ICM highlights the importance of understanding a participant or client's larger context, including structural factors that may affect health behavior. Jared, a Black, heterosexual, cisgender male in their early 60s, who was diagnosed with HIV at the age of 40 and who was struggling with medication adherence and substance use issues while in the Heart to Heart 2 project, drew a stark contrast between the connections forged during their time with the project and with other settings. They repeatedly stressed the importance of feeling seen and heard by all Heart to Heart 2 staff members alike: Because the basic thing here with you all is that you listen. All of you all. Even you. I'm looking at you. Thus, Jared highlighted the importance of being "listened to" as a precursor to receiving feedback that might assist them in achieving their health goals. This experience of being listened to and heard may re ect the ICM in that participants' larger contexts are considered a vital part of the clinical encounter. We found that when participants felt they were being genuinely listened to, the level of investment they felt by staff translated to the participant developing a deeper level of investment in themselves.
It follows that since the ICM highlights the importance of eliciting and understanding a participant as a whole person and in context, participants are more than just an "HIV positive person." Results indicated that having a space where participants experienced individualized services that took the larger context into consideration, while also supporting autonomy, commonly fostered their abilities to see HIV and HIV medication as aspects of their lives, and therefore, to reduce the sense of being de ned solely by their HIV status. Steven was a Black, heterosexual, cisgender male in their early 50s, who had been living with HIV for 20 years, and for whom "going to a doctor is a reminder" of their HIV status. For Steven, the ability during the Heart to Heart 2 project to have an "open dialogue" with chances to "branch off" into other areas of their life presented an "unexpected opportunity" to connect the painful emotions related to living with HIV and their experiences in typical HIV service settings more directly with their ongoing HIV medication adherence challenges. In the following quote, they refer to the study in general, and to Component D: focused support groups: You come to a place where [you can engage with] people with the [HIV] virus and shit. And you're offered a chance with people with the virus to talk and feel relaxed and comfortable. There's not that many places that you can feel human, because you've got to understand: There are times you feel less than human, which is why a lot of people don't take their medication and everything, because they want to forget. They want to be -I'm just normal. I'm like everybody else. So, you don't take medication, and you don't present to doctors and do what you want to do to live that fantasy of being everybody else. And you put yourself in jeopardy. Truthfully, you're putting more people in jeopardy.
As with most participants, feelings of validation that their life was seen in context, that they were valued as a whole person, and of connectedness with the project staff allowed for more open and honest conversations about a range of issues directly and indirectly associated with HIV care and HIV medication adherence. Further, the sense of acceptance and connectedness that resulted from the intervention activities grounded in the ICM were reported to increase participants' view on the acceptability of intervention components, as well as the likelihood that participants would continue to actively engage with the Heart to Heart 2 project over time; that is, study feasibility.

Experiences of trustworthiness and trust
As noted above, distrust of health care systems, HIV medications, and counter-narratives about the causes of HIV and its treatments are very common among AALB-PLWH, while distrust in actual health care providers is secondary. Yet, participants' trust in the project and the project staff are critical for perceived acceptability of services, ongoing engagement and thus study feasibility, and effective clinical service provision. Trust is a central precursor to effective counseling interventions and productive patientprovider relationships, as noted above. The ICM was designed in large measure to create a research project that was worthy of trust and to build trust. In part this was addressed by eliciting and understanding the valid reasons for medical distrust and counter-narratives among AABL-PLWH, since medical distrust and distrust of research projects are related. Guided by the ICM, the project staff sought to build participants' trust and communicate their trust in the participants to make their own health decisions, even in the context of medical distrust (which the project sought to elicit but not necessarily to question or change). Thus, to some extent trust between the project and participants was built by eliciting, discussing, and understanding the valid reasons for distrust and counter-narratives, and by the project expressing trust that participants would make the right decisions for themselves.
Participants frequently discussed the importance of trust and honesty in medical care and social services. They provided numerous examples of times when they either had or had not trusted or been honest with others, and also times when they had and had not been honest with themselves. Consistent with the intentions of the ICM, participants commonly experienced the Heart to Heart 2 project as an environment where they could speak openly and honestly about matters related to their physical and mental health care that, in the past, they typically felt were off-limits or that they avoided out of an abundance of caution. Of vital importance, participants noted they experienced being trusted and being seen as trustworthy in the project. For example, they experienced project staff as assuming that they (the participants) were experts on their own health and could and should make their own health-related decisions. This perspective, in turn, contributed to feelings of con dence and self-e cacy with respect to these health decisions. In contrast, participants reported commonly experiencing great pain and frustration in health care settings in cases where they were not trusted to be expert on their own health.
This theme of the importance of trust and trustworthiness was noted in discussions about relationships with HIV medical providers, project staff, and other support staff in participants' lives. Upton, introduced above, described a long history of signi cant mental health concerns and complicated substance use patterns, coupled with pervasive homelessness, which began when they disconnected from their family decades prior. Upton explained that all of these experiences made it very di cult to remain in medical or mental health care on a continuous basis, particularly over the past ve or six years. They discussed how they had been selling their HIV medication when they needed money for necessities, and sometimes for substances. Upton described being able to discuss these types of experiences honestly and without judgment during the Heart to Heart 2 project, suggesting Upton found the project trustworthy. Similarly, Ronald, introduced above, described the project as a place where they learned a great deal about themselves, their decisions, and their motivations. They noted the environment provided space for them to be honest with the project staff, which led Ronald to become more honest with themselves, and also more honest with their health care providers and others in their life. They described feeling trusted by staff to be a reliable reporter of their own experiences, including related to substance use problems, and that experiencing that trust helped them to prioritize their own health: [Heart to Heart 2 has] taught me a lot, man. Talk about your medication. But they don't see past that.
Thus, in addition to project staff seeking to build trust with participants and convey trust for participants' decisions, intervention activities grounded in the ICM commonly fostered participants' own trust in and honesty with themselves. This honesty with oneself was described by a number of participants as a critical aspect of their behavior change decisions and processes. We examine the role of a related concept, self-re ection, in more detail in the theme described below.
Opportunities for self-re ection on a range of topics and its effects In some cases, self-re ection, an intended outgrowth of the ICM, was directly linked to a new awareness of the importance of autonomy and self-determination in health care decisions. Samuel, a Black, gay, cisgender male in their late 30s, who had been living with HIV for nearly 20 years, said they experienced a major shift in the way they looked at their life during the study: The counselor, she's a real sweetheart. She never judged me, and never made a decision for me. She just asked me, "Well, how do you feel that decision has affected you?" And I never really thought about that. It was just a life-changing moment. Something just clicked. Like, yeah, that [the decision] is still not helping me. So, I want to do everything I can to help me, even though I still have to go through these dilemmas.
I'm going to make sure that I'm all right.
Further, non-judgement may be essential for self-re ection in this context. Indeed, the importance of nonjudgment was present in each of the themes described in this section.
Emmanuel was a Latino, bisexual, gender uid male in their early 40s, who had been living with HIV for approximately 10 years, and who had struggled for most of those years with remaining in HIV care and adhering to HIV medication. They explained the complex nature of their di culties remaining in HIV care prior to joining the study, which included di culties accepting their HIV status, feeling judged by multiple providers due to their sexual orientation and gender identity, a perceived lack of con dentiality in some service provider settings, and extensive substance use over the years. As they re ected on their time in the Heart to Heart 2 project, they said they experienced, "…a warm welcome when I came. For Emmanuel, therefore, positive relationships with the project and project staff and non-judgment were preconditions for engagement in clinical services that, in turn, spurred self-re ection. Further, selfdetermination, an aspect of the ICM, was clearly a deeply held value for Emmanuel.
William, a Black, heterosexual, cisgender male in their early 50s, who had been living with HIV for 20 years, discussed developing a pervasive substance use "addiction" at an early age, which led them to leave home and to become estranged from their family for more than 30 years. During that time, and due largely to their substance use challenges, William was diagnosed with HIV and struggled to remain in HIV care and on HIV medication. Further, they subsequently experienced homelessness for several years, but, more recently, was stably housed with a family member they had reconnected with. William looked back on their time before starting the Heart to Heart 2 study and re ected on the changes they had experienced in their life since starting the study, related in large measure to self-re ection that took place during study activities: But this place has done a lot for me, looking at me, looking at the things that I'm doing, the things I've suffering, but that they sometimes got in the way of "being there for myself". They described experiences in the study as a time for them to "to sit down and talk about these things out loud so that you can actually get it, process it," and "I started listening to what I was saying." These clinical experiences helped them to understand their own decisions and behaviors in a new and different way, which appeared to foster Wallace's more serious engagement in the behavior change process, rather than passive participation. This, thereby, helped them re-evaluate their behavior patterns (heavy substance use, selling HIV medication) and make changes: It's one thing to just think things or hear someone else talk to you about it. But when you have a conversation about it where you're involved in the conversation or a part of the conversation, it makes a difference. You process things a little differently, and it becomes more important to you.
Further, Wallace's quote above underscores the importance of the ICM in eliciting discussion of behaviors such as heavy substance use, and selling/diverting HIV medications. These behaviors are not generally considered socially acceptable, but candid acknowledgement of and self-re ection about them are vital aspects of the behavior change process for many.

Support of personal autonomy and its effects on motivation and decisions
As described above, participants generally experienced a lack of support for their personal autonomy in health care settings, particularly with regard to HIV care and decisions related to HIV medication. In particular, participants recalled feeling unable to advocate for themselves effectively in primary medical care and other social services environments. Although participants valued their individual providers, for the most part, HIV primary care settings were described as locations where they did not feel su ciently valued or listened to, and participants discussed feeling "stuck" and unsure if there were other options available to them; for example, regarding where else they might be able to receive care. In contrast, consistent with the ICM, participants noted the development of a sense of personal autonomy throughout their time in the Heart to Heart 2 project. Some recalled feeling unexpectedly empowered to change medical providers if they felt they were not receiving the care they needed and deserved. They also expressed renewed feelings of autonomy regarding their ability to take HIV medication regularly, remain in HIV care, and reduce substance use or substance use-related harms. Samuel, introduced above, recalled that prior to starting in the Heart to Heart 2 project they were feeling discouraged by the bureaucratic nature of the health care system, particularly at the primary care clinic they attended. They noted that experiences with the clinic, such as frustrations making appointments, made it extremely di cult to maintain their HIV care, leading them to stop caring about their health, which in turn contributed greatly to their stopping their HIV medication regimen. Samuel then described the changes they experienced in navigating their health care and HIV medication use after participating in the Heart to Heart 2 project for a year: When asked how the project helped with aspects of HIV and other health management, Samuel described the following: Okay, well basically the whole program to me -they were never judgmental, and they never made a decision for me. They just informed me of where I was and what decisions -and began to help me put in perspective what decisions were most important to me, and how in that plan I would respond differently when faced with different dilemmas.
Samuel's experience, therefore, provides support for the importance of non-judgment and personal decision making, both of which are aspects of the ICM. Olivia, a Latinx, gay, transgender female approaching the age of 50 years, who had been living with HIV since they were a teenager, explained that they go "back and forth" with taking HIV medication regularly, since "I get tired of taking them after a while". Olivia did not start HIV medication until they were in their early 40's. They attributed this decadeslong period of not taking HIV medication to negative experiences they had in the hospital when they were rst diagnosed with HIV in the early days of the epidemic, along with a general disdain for taking pills and feeling like HIV medication was a constant emotional reminder of their HIV status. In re ecting on their experiences in the study, Olivia said they felt more inclined to decide to take HIV medications as a result of their time in the study: Because you get information from each other here and, you know, sometimes when you do better things, when you go to better places and you do better things, it makes you feel better and you want to do better. Marcus was a Black, gay, cisgender male in their early 50s who was diagnosed with HIV when they were a teenager. Marcus noted "the majority of people that come [to the project], they come here from SROs [single-room occupancy residences], from drug usage, from prostitution, from the whole low income, promiscuous lifestyle." As a result, they suggested, participants like themselves were generally reticent to speak openly with health care and social service providers about some of the more heavily stigmatized challenges they face. Marcus described their experience of being in medical environments where they felt like, "another piece of cattle just going through the chute." They described this feeling as pervasive and added that it led them to stop attending appointments with multiple providers over the years. Marcus also discussed their substance use experiences extensively and also their challenges discussing substance use with providers: I don't talk to them [providers] about substance use. And the reason why I don't is because I believe that relapse is a part of almost everyone's story, and a doctor that tries to insert their will would be a problem Marcus, therefore, highlighted the importance of autonomy support and harm reduction, both aspects of the ICM, including because a "100 percent abstinence" model is not useful to them. Similarly, Simone, introduced above, described their time in the project as valuable. Simone had a history of very di cult experiences with medical providers who Simone described as pressuring them to engage in certain health behaviors but not truly understanding their larger context or needs. According to Simone, during the project they were able to traverse through a frustrating bureaucracy and nally nd a clinic where they felt comfortable and could develop a good relationship with their provider: In many instances, participants emphasized that while participating in the Heart to Heart 2 project was useful, it was they themselves who carried the burden of managing HIV and they who must make the decision to take HIV medication, underscoring the importance of self-determination theory and autonomy support in the ICM. As Hank, identi ed above described,

Discussion
The present qualitative and exploratory study focuses on a population of AABL-PLWH with socioeconomic disadvantage that experiences considerable barriers to HIV care and medication adherence, and, as a result, is either commonly not engaged, generally poorly engaged, or only intermittently engaged along the HIV care continuum. In response to the con uence of risk factors that impede HIV management in this population, and in the interest of ultimately creating new acceptable, feasible, culturally and structurally salient, and effective behavioral interventions to support sustained HIV viral suppression among AABL-PLWH, our research team developed a new ICM, which served as the foundation for clinical behavioral intervention components tested in an intervention optimization trial.
Overall, results from the present study provide evidence for the ICM's important role in the acceptability and perceived utility of the study activities, and also in fostering engagement and therefore, study feasibility. Participants also reported a range of positive effects on behavior in response to the clinical behavioral intervention activities grounded in the ICM. In particular, the study results highlight how the ICM contributes to productive therapeutic and clinical processes between participants and interventionists, which, in turn, has potential to lead to improvements in quality of life and/or to behavior change.
Trust and trustworthiness, power and privilege AABL-PLWH are often caught in cycles of distrust and partial self-disclosure with their health care and social service providers, in part related to pre-existing medical distrust and counter-narratives, which most care settings do not address or are not well-equipped to address [25]. Further, relationships between AABL-PLWH and providers are shaped by treatment perspectives that prioritize total abstinence from substance use, and similarly, the expectation of taking HIV medication "every dose, every day." Indeed, although harm reduction has gained some traction in HIV clinical settings, the abstinence-based perspective, applied to substance use and HIV management, has deep, historical roots [18]. These characteristics of health care settings and AABL-PLWH's perspectives on these settings certainly impede engagement along the HIV care continuum. Moreover, trust between medical or social service professional and AABL-PWLH is not easily established [76,77]. But, trust is vital to clinical and therapeutic interactions and self-change processes and fosters openness and self-re ection, but many research and clinical settings create environments and provide services that contribute to or perpetuate distrust [78][79][80]. Findings from the present study suggest that aspects of the clinical behavioral intervention activities carried out in the present study increased participants' trust in the research project and study staff in many cases, although overall medical distrust and counter-narratives may have persisted. Nonetheless, AABL-PLWH can certainly engage along the HIV care continuum even while experiencing medical distrust and counter-narratives, and commonly trust their providers more than health care systems [24,25].
As shown in Fig. 1, the ICM requires cultural and structural competence and comprises constructs that seek to communicate such cultural and structural competence, along with non-judgment and the support of autonomy, which in turn, is intended to foster trust between AABL-PLWH participants and the research project and its staff. Generally, we found participants engaged in frank discussions of their health decisions (e.g., not taking HIV medication) and contextual challenges (e.g., selling HIV medication, substance use) with project staff. This included participants making the decision to discuss with project staff those behaviors not typically considered socially desirable or socially acceptable. Participants contrasted their experiences in the Heart to Heart 2 project with typical health care encounters. For example, substance use and substance use problems, both past and present, are very common among AABL-PLWH, but challenging to discuss in typical health care encounters, in part due to stigma and the short interval of time patients have with providers [81]. The Heart to Heart 2 project activities grounded in the ICM, with its emphasis on non-judgement and including harm reduction, may have played a role in this contrast between the research project and typical care settings. Participants in the present study were from very low socio-economic status contexts, while the research study was located in a well-funded academic institution and study staff were, by and large, more privileged than participants. In such cases, power imbalances are marked and can serve as impediments to trusting relationships [82]. We speculate that the ICM approach helped project staff and participants work collaboratively across these marked differences, since the ICM explicitly recognizes participants as the ultimate experts on their health decisions. Indeed, themes related to the importance of non-judgment from project staff, a lack of pressure to take HIV medication or achieve any other speci c outcome de ned by the study, and the important role that open and honest communication had on self-re ection (which often spurred behavior change) were prominent in the analysis.
Health care settings are heavily in uenced by the larger structural and policy context The public health system and the individual HIV care providers within that system are charged with providing an effective and highly tolerable lifesaving treatment to individuals with a life-threatening condition, with the knowledge that such treatment can improve their wellbeing, provide them with a normal lifespan, and effectively eliminate the chances of their passing HIV [5]. Study ndings suggest current medical and social service settings may not be adequately prepared to address the con uence of serious barriers that some AABL-PLWH experience to HIV care continuum engagement. Indeed, there may be frustration on both sides: AABL-PLWH express dissatisfaction with the ways they are viewed and the treatment and care they receive, including feeling pressured to take HIV medication, and HIV care settings and providers may be frustrated when patients do not take or adhere to HIV medication regimens [83]. Yet these service setting characteristics serve as direct and indirect barriers to engagement along the HIV care continuum: AABL-PLWH may avoid service settings they experience as insu ciently supportive of autonomy and that do not provide dignity-enhancing care, and, as we found in past research, these stresses, strains, insults, stigmas, and hassles are commonly internalized by AABL-PLWH over time, with grave adverse effects on one's sense of self-worth, motivation to stay healthy, and HIV management behaviors [14]. The present study highlights the need for structural changes in order to end the HIV epidemic, along with enhancements to HIV care settings consistent with the ICM, as we discuss in more detail below.
Autonomy support and motivational interviewing in health care settings Although there are exceptions, in the present study HIV care and social service settings are generally described as dismissive of issues that AABL-PLWH commonly consider relevant to HIV management, such as poor housing quality and relationship instability, but that providers may not see as directly related to engagement along the HIV care continuum. Further, AABL-PLWH are involved in numerous institutional systems and settings that restrict their autonomy, as we describe above. Thus, AABL-PLWH nd individualized, autonomy-supportive, and non-judgmental approaches largely lacking in the care and service settings in which they are engaged, although they tend to be satis ed with their health care providers [ Cultural and structural salience of components The resonance of harm reduction and self-determination theory, aspects of the ICM, were prominent in the analysis. The role of critical race theory in the ICM were not as apparent in this analysis as aspects of the other theories or approaches, but because critical race theory was integral to the type, structure, and content of the study activities and intervention content (e.g., the role of poverty and housing, signaling the need for structural and cultural competence and salience, centering the project on AABL-PWLH), we do not consider it any less important. Indeed, throughout the project activities, participants were made aware that the Heart to Heart 2 project was focused on and interested in issues that shape the lives of AABL-PLWH such as structural racism, medical distrust, and counter-narratives about the causes and treatments of HIV. Thus, we assume that we cannot interpret ndings such as the resonance of the autonomy supportive approach without acknowledging the contribution of critical race theory to the ICM.

Limitations
The study has limitations including the possible in uence of social desirability bias on ndings. We sought to minimize social desirability bias during the interview process by asking general questions rst and reminding participants they could and should feel free to decline to answer any question without penalty. The primary qualitative interviewer was not someone participants had previously met or worked with, as a further means of reducing social desirability bias. The present study does not evaluate the effectiveness of the intervention components or the most cost-effective combination of intervention components, which will be the focus of future research. Instead, we explored participants' overall experiences with the Heart to Heart 2 project. Further, as noted above, the acceptability, feasibility, and potential impact of the speci c intervention components tested in the parent intervention optimization trial will be presented in a subsequent study, along with ways the components could be improved from the perspectives of participants. The clinical behavioral intervention components grounded in the ICM appear highly acceptable, but the Heart to Heart 2 project may have been better resourced than many HIV care settings. Indeed, the components were designed to be easy to access, exible, and easy to navigate, but not all service settings have the resources to take such an approach. and unstably housed clients [90]. Yet, as described throughout the present study, gaps remain.

Implications
In Table 3 we present a number of implications for policy, health departments, and HIV care delivery models drawn from the present study. These implications fall into two main categories: implications for the larger context in which AABL-PLWH are located, including structural changes, and implications for HIV care settings. The present study indicates it will be necessary to simultaneously address structural barriers and improve HIV care delivery models to engage this subpopulation of AABL-PLWH along the care continuum consistently. For example, chronic poverty is a core cause of disengagement along HIV care continuum [91], along with poor-quality housing [92]. The present study provides insights into potential alternate approaches that can be carried out within settings. Co-located substance use and HIV services, funding for high-quality supportive housing, and collaborative patient-provider relationships could improve sustained viral suppression among populations experiencing constellations of challenges such as substance use, poverty, and long-term HIV [93]. The present study also suggests there would be utility in some cases for a comprehensive, top-to-bottom designed or re-design of HIV care delivery models, HIV care settings, and social service settings for PLWH guided by the ICM or similar model, including involving AABL-PLWH in the planning process. Fox and colleagues (2014a, 2014b) describe such a model; namely, an urban transitions clinic serving formerly incarcerated persons that evidences promising engagement and health outcomes. Yet, similar to the present study, Fox and colleagues note that that access to medical care is necessary but not su cient to control chronic health conditions for this population with structural barriers to heath [94,95].

Conclusions
Study ndings indicate that the clinical intervention activities grounded in the ICM and delivered as part of the optimization trial were experienced by participants as highly acceptable, culturally and structurally salient, and useful. Further, the grounding of the study activities in the ICM contributed to participants' engagement in the study, and, therefore, to high rates of retention and study feasibility. Thus, as we hypothesized at the time the optimization trial was designed, the three theories/approaches that comprise the ICM, namely, critical race theory, harm reduction, and self-determination theory, appear to combine synergistically to create a new and useful model. The ICM may have applications for policy, to HIV care delivery models, and to other AABL populations in high-risk contexts, and warrants further study. Depiction of the integrated conceptual model that resulted from the combination of critical race theory, harm reduction, and self-determination theory