Rationale, design and methods of the Cul-CM study: a protocol for cultural adaptation of contingency management in adults with stimulant use disorder

Abstract

Introduction

Stimulant use disorder is a growing contributor to drug poisoning and mortality in the USA, with disproportionate impact on black communities driven by stimulant-opioid polysubstance use and other structural barriers to treatment. Contingency management (CM) is the most effective evidence-based treatment for stimulant use disorder but remains poorly implemented especially in minoritised communities. Cultural adaptation represents a promising implementation strategy to improve acceptability, engagement and equity of CM while maintaining fidelity to core evidence-based programming.

Methods and analysis

The Cultural Adaptation of Contingency Management (Cul-CM) study is a multiphase, mixed-methods study designed to culturally adapt CM for black adults with stimulant use disorder and evaluate its preliminary outcomes in a pilot randomised controlled trial (RCT), guided by the PEN-3 cultural model and the ADAPT-ITT framework. Phase I consists of an assessment of clinician beliefs about CM implementation using the Extended Contingency Management Beliefs Questionnaire (CMBQ-CA+, with an exploratory Cultural Adaptation subscale) and a scoping review of CM adaptation for substance use disorder. Phase II employs PEN-3-informed individual interviews and focus groups to identify culturally salient determinants of CM uptake and engagement. Findings from Phases I and II, in addition to iterative expert and community advisory board reviews and feedback, will inform the development of a culturally adapted CM (CACM) protocol, which retains core effectiveness elements and modifies peripheral delivery elements of CM. Finally, the Phase III consists of a three-arm pilot RCT (n=60) comparing the CACM, standard CM and treatment as usual following a 1 week lead-in (priming) phase. Primary outcomes include stimulant abstinence and longest duration of continuous abstinence. Secondary outcomes include treatment acceptability, satisfaction, feasibility and retention. Cul-CM operationalises cultural adaptation as a systematic, theory-driven strategy to improve CM outcomes in adults with stimulant use disorder and promote equitable uptake in community treatment settings. Cul-CM represents a replicable framework for culturally adapting CM to ensure equitable access, engagement and improve outcomes.

Ethics and dissemination

Ethical approval was obtained from the Yale University Institutional Review Board. All participants will provide written informed consent. Study findings will be disseminated through peer-reviewed publications, scientific conferences and community-engaged channels to inform culturally responsive approaches to CM for stimulant use disorder.

Trial registration number

NCT07443332.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The Cultural Adaptation of Contingency Management study operationalises cultural adaptation as a structured intervention development strategy for contingency management (CM), integrating the PEN-3 model within the ADAPT-ITT process framework through a stepwise, theory-driven approach.

• The study extends the CM Beliefs Questionnaire with an exploratory cultural adaptation subscale to assess clinician beliefs about CM implementation, adaptation and equity-relevance.

• The study demonstrates a replicable community-partnered, multi-stakeholder workflow that produces iterative drafts of evidence-based adaptation.

• The study links adaptation to implementation endpoints via a pilot three-arm randomised trial with a priori feasibility benchmarks, fidelity monitoring and mixed-methods integration to inform scale-up.

• As a pilot trial with a modest sample size, the study is underpowered for definitive effectiveness conclusions but will lay the foundation for a fully powered effectiveness trial.

Introduction

Minoritised communities have continued to experience escalating patterns of drug overdose mortality, driven in part by stimulant (cocaine and methamphetamine) use, high-potency synthetic opioid contamination of community stimulant supplies and structural drivers of inequities. Stimulant-related overdose mortality has risen in parallel with synthetic opioid (fentanyl and fentanyl analogues) exposure contributing to widening racial inequities.^{1 2} This pattern of poly-use (stimulant-opioids) was responsible for 32.3% mortality in 2021 compared with only 0.6% in 2010.²,⁴ A racial pattern of drug poisoning has also emerged, between 2013 and 2018, the presence of opioids in cocaine-related deaths increased by 135% in black populations but only 33% among white populations.² Overall, black populations experience 1.4 times the overdose rates compared with white populations.¹

This overdose mortality pattern represents a major public health crisis reflecting deep inequities shaped by social determinants, systemic and structural drivers including inequitable access to evidence-based substance use treatment. An example of such a structural determinant is the persistent 1980s race-based framing of cocaine use shaping modern public debate, despite long-standing evidence showing no significant differences in stimulant use across racial groups.^{5 6} This has contributed to worsening stigma, disproportionate involvement in the criminal justice system,^{7 8} disparities in access to treatment^{9 10} and worse health outcomes among minoritised communities.

Black people remain less likely to have access to addiction treatment and more likely to experience adverse consequences across a range of clinical outcomes. According to the National Survey on Drug Use and Health, in 2022, only 14% of black people with substance use disorder (SUD) in the past year received any SUD treatment and nearly all those who did not receive treatment reported not perceiving a need for treatment.⁶

In order to address these disparate trends of stimulant-related mortality, strategies must include scaling evidence-based treatments for stimulant use disorder (StUD) particularly among the populations most impacted.¹ While there are no United States Food and Drug Administration (FDA)-approved medications for StUD, contingency management (CM) remains the most effective evidence-based intervention,¹¹,¹⁴ yet grossly underused,¹⁵,¹⁸ reflecting multilevel policy, institutional and professional barriers to implementation and scale-up.^{15 19} CM is scarcely available in most community-based addiction treatment programmes despite four decades of robust evidence and consistent positive outcomes across a range of contexts.^{20 21} Moreover, research has shown that there is particularly low penetration of CM in black communities.²² With the exception of the Veterans Health Administration’s national implementation of CM^{11 12} and the more recent California’s Recovery Incentives Programme,²³ the limited CM adoption in mainstream healthcare delivery perpetuates treatment inequity by limiting access to a lifesaving treatment for a condition that disproportionately affects black populations.

Efforts to promote CM adoption alone may not be sufficient to address inequities, given emerging evidence that CM treatment response may vary by race and baseline stimulant-use severity. For example, a secondary analysis of six randomised clinical trials found that black participants with StUD who entered treatment with cocaine-positive urine toxicology did not experience significant gains in retention or other clinical outcomes from CM compared with standard care, whereas white participants who entered treatment with stimulant-positive urine demonstrated better response to CM compared with standard care.²⁴ Further, they reported that gains in terms of drug use outcomes were muted relative to whites with interaction effects persisting through the 9-month follow-up period.²⁴ These results were similar to a prior randomised trial that found that young black adults with cannabis use disorder did not experience a reduction in cannabis-positive toxicology screens following treatment with CM, while young white adults did.²⁵ A possible explanation for the failure of earlier studies to establish racial or ethnic differences in CM outcomes may be due to failure to specifically examine the interaction with abstinence at baseline (treatment entry).²⁶ Together, these findings suggest that race alone may not fully explain differential CM outcomes; rather, the interaction of race with contemporary sociocultural, structural and clinical factors may shape engagement with and response to CM among specific minoritised groups.

To address disparities in real-world effectiveness and improve implementation fit, there have been increased calls for cultural adaptation of evidence-based interventions²⁷,³⁰ including CM.^{2425 31},³³ The importance of adapting interventions to improve cultural acceptability and other implementation outcomes is increasingly emphasised in clinical practice and research.³³,³⁹ Cultural adaptation may strengthen a sense of belonging and autonomy among historically minoritised people with SUD as shown in trials of culturally adaptation of CM in Native American populations showing encouraging outcome signals.^{2238 40},⁴² However, despite evidence of disparities in CM outcomes and burgeoning evidence of CM cultural adaptation improving CM implementation, there are no known adaptations of CM specifically in black communities.

This report introduces and describes the protocol for the Cultural Adaptation of CM for adults with StUD (Cul-CM) study, the first study to culturally adapt CM for StUD in black populations to address the escalating overdose deaths related to stimulant use (including stimulant-opioid contamination). The protocol consists of multiple phases: the evidence-generation and synthesis (phase I), the intervention-development (phase II) and the testing (randomised controlled trial (RCT), phase III). The protocol is designed to have a strong public health impact by leveraging the lived experience of black individuals with StUD and the experience of their communities through engagement of multiple stakeholder groups. This approach is intended to strengthen community engagement with CM and ultimately improve the acceptability, uptake and effectiveness of CM within black communities. The objective of the initial two phases of the Cul-CM Study is to develop core components of a culturally adapted CM (CACM) intervention for black adults with StUD by:

1. Identifying relevant cultural factors and cultural narratives that shape stimulant use initiation and progression among black populations.

2. Identifying determinants of CM implementation among black adults with StUD and stakeholders.

3. Identifying and developing culturally meaningful components necessary to adapt CM for StUD in black populations.

The objective of the third and final phase of the Cul-CM study is to conduct a pilot randomised trial evaluating the CACM intervention against both standard CM (SCM) and treatment as usual (TAU) to determine if the adapted intervention improves patient outcomes.

Methods

Conceptual framework

The Cul-CM study is grounded in two complementary frameworks: the PEN-3 cultural model (figure 1),⁴³,⁴⁵ embedded within the stepwise, sequential and iterative phases of the ADAPT-ITT implementation science framework.⁴⁶ Centering culture as a core determinant of health behaviour and intervention effectiveness is instructive in this study as the black population in the USA is heterogeneous, encompassing a shared cultural experience but with multiple subcultures that may meaningfully influence intervention design, delivery and uptake.⁴⁷ Thus the PEN-3 cultural framework (figure 1) positions culture at the centre of health beliefs, health behaviours and health outcomes.⁴⁵ In the Cul-CM study, all qualitative components are guided by this framework to ensure cultural relevance and interpretive rigour.

Figure 1. The PEN-3 cultural model.

The PEN-3 cultural model consists of three interrelated and interdependent dimensions: (1) cultural identity, (2) relationships and expectations and (3) cultural empowerment.⁴³ Together, these dimensions provide a culturally grounded approach that defines health behaviours, cultural context and frames a cultural appropriateness of prevention and intervention strategies.

The Cultural Identity domain of PEN-3 consists of the subdomains of Person, Extended Family and Neighbourhood. This domain highlights the individual’s point of entry into treatment, access to care and determinants of ongoing engagement. Cultural identity emphasises that effective health interventions should extend beyond the individual and must incorporate family and neighbourhood contexts, not only as key access points but also as targets for health behaviour change and prevention.

The second domain of PEN-3 is Relationships and Expectations with subdomains of Perceptions, Enablers and Nurturers. The relationship and expectations domain is used to frame stimulant use within the intersection of individual beliefs, structural resources, family systems and cultural norms. This domain examines perceptions and attitudes towards stimulant use; societal and structural resources, including healthcare services, that may promote or discourage health-seeking behaviours and the influence of family members and social networks on decision-making related to stimulant use and treatment engagement.

Finally, the third component of PEN-3 is Cultural Empowerment with subdomains of Positive, Existential and Negative. This domain identifies culturally grounded beliefs and practices that are beneficial (positive), culturally meaningful but health-neutral (existential) and harmful (negative). Cultural empowerment domain supports the validation of protective cultural practices, acknowledgement of neutral traditions and critical examination of practices that function as barriers to health. Within the Cul-CM study, this domain informs identification of culturally congruent prevention and risk-reduction strategies, while addressing practices associated with adverse health outcomes.

The PEN-3 cultural model is used in tandem with the ADAPT-ITT framework (table 1), an implementation science process model designed to support the systematic adaptation of evidence-based interventions. ADAPT-ITT provides a sequential process for intervention adaptation that promotes stakeholder engagement, facilitates local ownership and enhances acceptability while preserving fidelity to core intervention components.⁴⁶ In this study, the ADAPT-ITT framework operationalises the translation of PEN-3-informed qualitative findings into an iteratively refined, CACM intervention.

Table 1. The sequential ADAPT-ITT framework and activities.

ADAPT-ITT phases	PEN-3 integration and activities
Assessment (A)	Systematic review of epidemiological and contextual data to identify cultural, structural and social drivers shaping stimulant use and treatment engagement.
Decision (D)	Decide which core CM elements must be retained to preserve effectiveness and identify delivery elements suitable for culturally responsive adaptations to CM delivery.
Administration (A)	Introduce individuals with StUD, community members, patients and clinicians to CM programme. Conduct PEN-3 individual interviews and focus groups. Organise and analyse data from phase I and phase II of the Cul-CM study to determine CM acceptability, cultural fit and perceived relevance. Modify CM delivery elements to ensure cultural responsiveness in response to stakeholder feedback.
Production (P)	Develop Draft 1 of the culturally adapted CM manual: language, examples, reward structures, delivery style, equity considerations and community-informed modifications.
Topical experts (T)	Engage CM and implementation science experts and community advisory board to refine Draft 1 to ensure fidelity to evidence-based intervention and cultural alignment (Draft 2).
Integration (I)	Incorporate expert and community feedback to produce Draft 3—the improved, culturally tailored CM intervention.
Training (T)	Train clinicians, counsellors, peers and community partners to deliver culturally adapted CM with fidelity, cultural humility and equity principles.
Testing (T)	Pilot the adapted CM intervention with the target population. Assess feasibility, acceptability, appropriateness and preliminary outcomes.

CM, contingency management; StUD, stimulant use disorder.

Study overview

The Cul-CM study will be conducted across three sequential phases using a mixed-methods design (figure 2). Phase I includes a systematic review and mapping of literature on cultural adaptation of CM for SUD and an online survey of addiction clinicians’ attitudes, barriers and facilitators regarding the adoption of CM in clinical practice using the extended Contingency Management Beliefs Questionnaire (CMBQ-CA+) consisting of the original Contingency Management Beliefs Questionnaire (CMBQ) and a Cultural Adaptation Subscale (CMBQ-CA). The CMBQ was developed by Rash and colleagues⁴⁸ but the Cul-CM team expanded the CMBQ to include an exploratory six-item adaptation subscale, CMBQ-CA, specifically designed to assess addiction clinicians’ beliefs regarding the cultural adaptation of CM (online supplemental file 1). The subscale was developed with critical input from experts in CM (CR), cultural adaptation (CB) and implementation science (SJB), as well as from the Community Advisory Board (CAB). The CMBQ-CA+ survey will also assess clinicians’ prior exposure to, familiarity with and experience delivering CM. The survey will include those both with and without prior CM experience thus providing valuable information regarding perceived acceptability, feasibility, training needs, implementation barriers and cultural considerations that may affect future CM adoption.

Figure 2. Overview of the phased Cul-CM study. CACM, culturally adapted CM; CM, contingency management; CMBQ-CA+, Extended Contingency Management Beliefs Questionnaire with Cultural Adaptation Subscale; Cul-CM, Cultural Adaptation of Contingency Management; SCM, standard CM; TAU, treatment as usual.

Phase II of the Cul-CM study comprises two complementary qualitative components: (a) PEN-3-guided individual interviews of adults with StUD, family members, clinicians and community members; and (b) PEN-3-guided focus groups with the same stakeholder groups. Findings from phases I and II will inform the iterative development of a CACM intervention using the ADAPT-ITT sequential process. Although prior research and extant literature review show key CM fidelity anchors,⁴⁹ a critical aspect of the ADAPT-ITT process will be also determine the CM elements that must be preserved for effectiveness and feasibility within participating settings, while also identifying peripheral delivery elements that can be modified. Based on preliminary feedback from partner programmes that have agreed to refer participants for this phase, we anticipate that established core CM fidelity anchors will be preserved. Modifiable elements will include multiple dimensions of CM delivery-intervention protocol preference and modes of delivery (digital vs traditional methods) for the adapted intervention, culturally meaningful reinforcers, communication and framing of incentives, perceived fairness and trust, family and community influences on engagement, institutional-based and community-based barriers to participation and other implementation constraints within the treatment setting.

Finally, Phase III consists of a pilot RCT to evaluate the feasibility, acceptability and preliminary outcomes of the CACM intervention relative to SCM and TAU.

Participant recruitment

Patient participants in the Cul-CM study will include black adults with StUD and key stakeholders, including family members, clinicians and local community members. Importantly, non-patient participants can belong to any racial category. Recruitment will occur in three stages. First, an online survey of addiction clinicians (n=250) including physicians, nurses, psychologists, social workers, counsellors and peer support. As in previous studies, we define addiction clinicians as those who have treated at least one individual in the past year with specific treatment goals related to the individual’s addiction.⁴⁸ The survey will be conducted using the CMBQ-CA+ assessing clinician beliefs regarding CM clinical adoption and implementation and cultural adaptation. Survey participants will be sought via targeted emails through academic departments, addiction state and national associations, agencies and other large addiction specific organisations. Participants will be compensated $20 for completed surveys.

Second, individual in-depth interviews (n=50) will be conducted with adults with StUD (n=20), family members (n=10), community members (n=10) and clinicians (n=10). Third, four (4) focus groups (n=40) will be conducted, each comprising approximately 10 participants, including individuals with StUD (n=20), family members (n=10), clinicians (n=5 comprising addiction counsellor/social worker, peer, psychologist, nurse and addiction physician) and community members (n=5).

Patient participants for Phase II will be recruited from the New Haven community using networks from the CAB and from four substance use treatment settings: Medication for Addiction Treatment Clinic, Connecticut Mental Health Center, the Substance use and Addiction Treatment Unit and the APT Foundation.

All participants will be aged at least 18 years old. Specific inclusion criteria for patient participants will include: (1) self-identification as black or African American; (2) a diagnosis of StUD as documented in the patient’s medical record, treatment plan or structured clinical interview for Diagnostic and Statistical Manual of Mental Disorders (DSM)-5 diagnostic interview and (3) stimulant use in the past year. Patient participants who are actively suicidal, acutely intoxicated or otherwise unable to provide informed consent will be excluded. Other participants in this phase of the study (ie, clinicians, family members and community members) do not have to belong to any specific racial category. Clinician participants may hold a range of professional job functions including (physicians, social workers, licensed drug and alcohol counsellors, nurses, peer counsellors and psychologists). All participants will be compensated $50 for individual interviews and $50 for focus group.

Community Advisory Board

The CAB will consist of five community members residing in the New Haven area who represent one or more of the target stakeholder characteristics: (a) person with lived experience of SUD; (b) clinician involved in community-based addiction treatment or (c) involved in community SUD-related advocacy. The CAB will meet quarterly and at key decision points throughout the study including evaluation of ongoing activities and guide study adaptations as needed (table 2). CAB members will receive compensation of $50 per meeting. Prior to study initiation, CAB members will participate in a 1 day orientation meeting to review the study design, procedures and timeline. Subsequent meetings will occur in person or virtually. CAB members may contribute to naming the study, recommend modifications to recruitment materials, review and refine questionnaires, provide feedback on the feasibility and cultural relevance of any components of the study and facilitate participant recruitment. CAB activities will be integrated across study phases and mapped to key implementation functions, including adaptation, engagement and dissemination (table 2).

Table 2. Mapping of Community Advisory Board (CAB) responsibilities to implementation functions.

Domain	CAB responsibility	Operationalisation in Cul-CM study
Study design and adaptation	Review of study materials	Iteratively review interview guides, surveys and intervention adaptation drafts to ensure cultural relevance, clarity and alignment with community norms.
Recruitment and engagement	Recruitment and retention strategy guidance	Provide input on culturally responsive recruitment channels, messaging and retention approaches to optimise participation among target populations.
Cultural validity	Oversight of cultural appropriateness	Assess study procedures and intervention components to ensure cultural congruence and acceptability within the community context.
Community partnership	Bidirectional communication	Serve as a conduit for ongoing, bidirectional communication between the research team and community stakeholders, ensuring community perspectives inform study decisions.
Study oversight	Monitoring and milestone feedback	Review study progress at predefined milestones and when needed to provide structured feedback and refine procedures and address emerging challenges.
Dissemination and translation	Dissemination support	Co-develop and assist in dissemination strategies, including translation of findings into accessible formats for community audiences and stakeholders.

Cul-CM, Cultural Adaptation of Contingency Management.

Patient and public involvement

Patients and members of the public were involved in the initial design of this study through a CAB. The CAB contributed to refining the research question, shaping the study procedures and finalising the overall study design. Community stakeholders will continue to be involved throughout the study, including in implementation, interpretation of findings, reporting and dissemination planning. Their involvement is intended to ensure that the study remains grounded in community priorities and responsive to the lived experiences of individuals and communities affected by stimulant use disorder.

Study instruments

The Extended Contingency Management Beliefs Questionnaire (CMBQ-CA+)

The CMBQ is a 32-item instrument rated on a five-point Likert scale that assesses provider beliefs regarding the adoption of CM for SUD treatment.⁴⁸ The CMBQ measures key barriers and facilitators to CM implementation, including general barriers, training-related issues and pro-CM attitudes.⁴⁸ For the Cul-CM study, an additional exploratory cultural adaptation subscale (CMBQ-CA) was developed to assess provider beliefs regarding the cultural adaptation of CM. The CMBQ-CA consists of six questions: five closed-ended items rated on the same five-point Likert scale, assessing perceived effectiveness, acceptability, cultural fit, equity-related impact and feasibility of culturally adapting CM, and one open-ended item eliciting respondents’ perspectives on potential cultural adaptations in practice. The CMBQ-CA was developed iteratively with input from one CM effectiveness researcher (CR), one CM implementation researcher (SJB), a cultural adaptation expert (CB) and feedback from the CAB (table 2). The subscale was designed to align conceptually with the original CMBQ while minimising respondent burden.

PEN-3 guided qualitative interviews and focus groups

Following eligibility determination and informed consent, patient participants will complete demographic measures (eg, age, income, education, gender, housing status) and assessments of psychosocial, substance use, psychiatric, and medical history. Interviews will be analysed using a hybrid deductive-inductive approach (see Phase II analyses below). The interview guide is informed by the PEN-3 framework (onlinesupplemental file 2AB).

Focus groups will begin with a brief video demonstration of the components, mechanics, processes and structure of SCM protocols (voucher and prize-based CM).⁵⁰ Guided discussion will then explore behavioural risks, perceived need for prevention, barriers to care, perceptions and preferences for specific CM protocol and intervention design and delivery. Focus group discussions will remain open until thematic saturation is reached, defined as the point at which no new concepts or perspectives emerge, and participants begin to repeat previously identified themes. In addition, interviews and focus groups will elicit discussion of potential adaptation targets not typically included in SCM protocols but identified as salient to the target population, informed by prior adaptation work by Hirchak and colleagues.³⁸ These targets include but are not limited to:

1. Specific professional job function (eg, licensed drug and alcohol counsellor, social work, psychologist, nursing, case management, peer recovery support and physician) and racial and ethnic composition of CM staff, given evidence that provider-level characteristics may influence CM dissemination, implementation strategies, adoption and uptake.⁵¹

2. Cultural appropriateness, type and size of incentives.

3. Intervention setting (eg, clinic (specialty)-based vs community (non-specialty)-based locations such as community centres).

4. Other relevant contextual factors, including language and cultural and local norms.

Component analysis and development of the CM adaptation plan

Component analysis of the CM adaptation will involve synthesis of Phases I and II and subsequent systematic refinements to intervention content and programming guided by ADAPT-ITT Phases 4–6 while preserving the fidelity anchor of CM (figure 3). Component analysis will clarify the goals of adaptation, inform the structured development of the adaptation plan and culminate in a final adapted intervention draft (Draft 3). ADAPT-ITT Phase 4 (Production) will involve generating Draft 1 of the CACM programming based on findings from Phases I and II. During this stage, any CM adaptations that were recommended as high priority that cannot be feasibly incorporated into the protocol based on partner preferences will be documented as candidate delivery strategies for future studies. Phase 5 (Topical Experts) will include review and feedback from implementation science, contingency management and cultural adaptation experts to refine Draft 2. Phase 6 (Integration) will synthesise expert and CAB input to produce Draft 3, the finalised CACM protocol for pilot testing. Phase 7 (Training) will include preparation of the study principal investigator (OOJ) to facilitate the CACM intervention, and develop training procedures for CM interventionists and clinicians, and create study assessments and data management processes. Phase 8 (Testing) will consist of implementation of the pilot RCT.

Figure 3. Sequential mapping of ADAPT-ITT phases to the iterative development of the culturally adapted contingency management intervention (Drafts 1–3), culminating in pilot randomised controlled trial (RCT) testing.

Phase III: pilot RCT

Cul-CM Phase III consists of a three-arm, parallel-group pilot RCT (n=60) designed to evaluate the feasibility, acceptability and preliminary clinical outcomes of a CACM intervention compared with SCM and TAU. Participants will be randomised in a 1:1:1 ratio (n=20 per arm). As a pilot study, this trial is not powered for definitive efficacy testing; rather, the sample size was selected to estimate effect sizes, feasibility indicators, recruitment and retention rates and implementation parameters to inform a future fully powered trial.

Participants will be recruited from the partner clinics (see above) as described in Phase II and the community. Phase III includes a 1-week lead-in phase (described below), before the 12-week randomised intervention phase. Randomisation and allocation concealment will be conducted using the Research Electronic Data Capture (REDCap) randomisation module hosted by Yale University.⁵² Eligibility will be determined based on same criteria for individuals with StUD as outlined for Phase II, but participants will be required to have used stimulants in the past 30 days. Although, participants and research assistants cannot be blinded to treatment assignment, outcome assessors and data analysts will remain blinded to group allocation where feasible. Participants in all three arms may continue to access routine clinical services available through their treatment programme. Study-delivered CM will be provided only to participants randomised to the SCM and CACM arms, whereas participants randomised to the TAU arm will receive routine clinical services without study-delivered CM.

Lead-in (priming) phase

All eligible participants will complete a 1 week lead-in phase prior to randomisation. During this phase, participants will attend clinic twice and will receive $20 per visit only for providing urine samples regardless of urine toxicology results. In addition to providing urine, participants will provide other study-related data (table 3). The lead-in phase is intended to familiarise participants with study procedures, establish engagement, and reduce post-randomisation attrition. Participants who complete the lead-in phase will then be randomised.

Table 3. Diagnostic assessments and psychosocial measures in the pilot trial.

Domain/construct	Measure (instrument)
Diagnostic assessment	MINI International Neuropsychiatric Interview (MINI)75
Substance use (past 30 days)	Timeline Follow-Back (TLFB)76
Psychiatric symptom severity	Brief Symptom Inventory (BSI)77
Substance use severity	Addiction Severity Index–Lite (ASI-Lite)78
Ethnic identity	Multigroup Ethnic Identity Measure–Revised (MEIM-R)79
Quality of life	12-Item Short-Form Health Survey (SF-12)80
Medical comorbidity	Charlson-Deyo Comorbidity Index81
Discrimination	Everyday Discrimination Scale82
Social determinants of health	Protocol for Responding to and Assessing Patients’ Assets, Risks, and Experiences (PRAPARE)83

Standard contingency management

Participants randomised to the SCM arm will receive CM consistent with the prize-based (Petry) CM protocol, as recently revised to keep up with inflation.^{33 42 53 54} This prize-based CM protocol was selected a priori based on the preferences of participating treatment programmes as it is an evidence-based approach with comparable effectiveness to voucher-based CM at lower cost. Participants will receive an escalating schedule of reinforcement for submitting stimulant-negative urine at each clinic visit over the course of the 12-week intervention. Per SCM protocol, there will be a cap at eight draws at week 4 of continuous abstinence.⁴²⁵³,⁵⁵ Participants will submit urine samples twice weekly during intervention visits, stimulant urine tests will be conducted using an office urine drug screening kit that provides results within 10 min. Participants who provide a stimulant-negative urine sample will earn reinforcement using a variable-magnitude prize-based ‘fishbowl’ system including 250 non-winning slips (Good job), 209 small prizes ($5 value), 40 large prizes ($25 value) and 1 jumbo prize ($100 value). This distribution produces frequent small reinforcements with a low probability of high-magnitude rewards, maintaining cost-efficiency while preserving behavioural salience. The expected value per week approximates $55, consistent with established CM trials for StUD.⁵⁵ Total potential earnings during the 12-week treatment phase will depend on sustained abstinence and escalation of draws with a maximum expected value of $660, consistent with prior prize-based CM protocols.⁵⁵

When consecutive negative urines are broken by a stimulant-positive urine sample, an unexcused absence, or refusal to provide a sample, no draws (zero) are earned for that visit and the draw schedule resets to its original starting value of 1 draw for the next stimulant negative urine, escalating by 1 draw up to a cap of 8 draws. Each stimulant positive urine, refused sample or unexcused missed visit will trigger a reset.

Culturally adapted contingency management

Participants randomised to the CACM arm will receive the culturally adapted version of CM as developed through the PEN-3 informed ADAPT-ITT process (Draft 3). The CACM intervention will preserve core mechanisms of the selected protocolised CM reinforcement structure (prize or voucher), including objective verification of stimulant abstinence, immediacy of reinforcement, an escalating reinforcement schedule and reset processes while incorporating culturally tailored modifications. For comparability to SCM condition, the CACM will also consist of twice weekly sessions of 12 weeks. Potential adaptable elements include but are not limited to the CM interventionists’ professional job function and personal characteristics; language and framing of the CM protocol, and delivery workflow; other mechanistic details of the selected CM protocol including reinforcement schedule, escalation, reset protocol and elements of fishbowl mechanism; type and cultural relevance of reinforcement offered; engagement supports and communication strategies including trauma-informed communication around urine toxicology testing, purpose of testing (therapeutic monitoring vs punitive surveillance), address concerns about privacy and stigma, and communicate results in a respectful and nonjudgmental manner.

Treatment as usual

Participants randomised to TAU will receive standard Intensive Outpatient Programme services at one of the partner programmes. TAU consists of 2 weekly visits, typically including one group counselling session lasting approximately 60–90 min and one individual counselling session lasting 30–60 min. Participants will continue other psychiatric care as clinically indicated (including prescribed medications) and will be encouraged to participate in mutual-help groups such as 12-step meetings.

Intervention fidelity monitoring

Intervention fidelity will be monitored to ensure preservation of core CM mechanisms across both CM conditions. Intervention fidelity will be monitored throughout the 12-week trial using a structured CM fidelity checklist (online supplemental file 3) for both the CACM and SCM arms to ensure consistent delivery of core intervention components across study conditions. Assessments will be conducted by the principal investigator (OOJ) or other trained research assistant who is not the CM interventionist for the session being reviewed. To ensure early procedural consistency, all CM sessions delivered during the first 2 weeks of implementation will be reviewed. Thereafter, a random sample of approximately 20% of sessions will undergo fidelity assessment, with additional targeted review of sessions in which protocol deviations are suspected. Fidelity assessments will be done by direct observation and structured review of urine toxicology records, reinforcement logs, prize administration records and escalation/reset tracking forms. Fidelity findings will be reviewed regularly in supervision meetings to identify drift and guide corrective feedback or retraining. Adapted components unique to CACM will be monitored separately to ensure that cultural modifications are delivered as intended while maintaining fidelity to core CM principles thus limiting contamination and overlap between the adapted and SCM conditions.

Outcomes and assessments

The primary outcomes of the trial include percentage stimulant-negative urine samples during the 12-week trial and the longest duration of continuous abstinence achieved during treatment. Secondary outcomes include: (1) treatment retention at 12 weeks; (2) treatment acceptability post intervention measured using the Acceptability of Intervention Measure (AIM), a four-item measure of perceived intervention acceptability⁵⁶; (3) feasibility of intervention measured post intervention using the Feasibility of Intervention Measure (FIM)—a 4-item instrument to assess perceived intervention feasibility⁵⁶ and (4) treatment satisfaction post intervention measured using the Client Satisfaction Questionnaire (CSQ-8), an 8-item, self-report measure developed to assess global satisfaction with health and behavioural health services.⁵⁷ Urine toxicology will be collected at baseline, throughout the treatment phase and during follow-up assessments.

Additional measures collected at baseline and monthly (weeks 4, 8 and 12) will include substance use severity (Addiction Severity Index—Lite), other drug use, psychiatric symptom severity (Brief Symptom Inventory), quality of life and perceived discrimination. All-cause emergency department visits and psychiatric hospitalisations will also be assessed. Each participant across all three-arms will receive $20 for completion of each follow-up assessment and interviews.

Harms and adverse events

Although CM is a low-risk behavioural intervention, adverse events will be monitored throughout the study. Potential risks include psychological distress during assessments or discussions of substance use. Any adverse events or unanticipated problems will be documented and reported to the Yale Institutional Review Board in accordance with institutional policies. Participants requiring clinical support will be referred to appropriate services.

Ethics and consent

The study protocol has been approved by the Yale University Institutional Review Board (IRB, HIC #2000040284). All participants will provide written informed consent prior to participation in any phase of the study. Participant confidentiality will be protected through assignment of unique study identifiers. Identifiable information will be stored separately from research data on secure, encrypted servers. Only authorised study personnel will have access to identifiable information.

Dissemination

Study findings will be disseminated through peer-reviewed publications, conference presentations and community forums. The CAB will assist in translating findings into accessible formats for community stakeholders.

Data analytic plan

Data management

Study data will be collected and managed using REDCap hosted at Yale University. Access will be restricted to authorised study personnel through password-protected accounts. Identifiable information will be stored separately from analytic datasets. Regular data quality checks will be conducted to ensure completeness and accuracy. De-identified datasets will be used for analysis. All research records will be maintained in accordance with IRB guidelines and securely stored for 12 months following study completion. During this period, physical documents will be kept in locked cabinets within secure research offices, while electronic data will be stored on encrypted, password-protected systems with access limited to authorised personnel and routinely backed up. After the retention period, identifiable records will be securely destroyed through shredding of paper files and permanent deletion of digital data. Any long-term storage will require prior IRB approval and will use de-identified data whenever possible to ensure participant confidentiality.

Phase I: survey analyses

Participant characteristics and survey responses will be summarised using descriptive statistics. Continuous variables will be reported as means and SD, and categorical variables as frequencies and percentages. For both the CMBQ and CMBQ-CA, item-level distributions, means and SD will be calculated. Internal consistency of the CMBQ-CA subscale will be evaluated using Cronbach’s alpha, with α ≥0.70 considered acceptable for exploratory purposes. Item-total correlations will be examined to assess individual item performance and coherence with the overall construct.

Construct validity of the CMBQ-CA will be evaluated by examining correlations with conceptually related CMBQ subscales, including perceived effectiveness, acceptability and feasibility. Exploratory factor analysis will be conducted, contingent on adequate sample size, to evaluate the dimensionality of the CMBQ-CA. Differences in CMBQ and CMBQ-CA scores across provider characteristics (eg, discipline, years in practice, prior CM experience) will be examined using independent-samples t-tests or Analysis of Variance (ANOVA) for continuous measures and χ² tests for categorical comparisons, as appropriate. Qualitative data from the single open-ended CMBQ-CA question eliciting potential cultural adaptations will be exported and aggregated into a list of potential adaptations, which will be combined with the data elicited in Phase II and analysed qualitatively (described below).

Phase II: qualitative analyses

Qualitative data generated from the PEN-3-guided individual interviews and focus groups and the single open-ended question from phase I (see above) will be analysed using applied thematic analysis techniques.⁵⁸ All interviews and focus groups will be audio-recorded, transcribed verbatim and reviewed for accuracy. A hybrid inductive-deductive analytic approach will be employed. Initial coding will be informed deductively by the PEN-3 framework, while allowing inductive themes to emerge from participant narratives. A preliminary codebook will be developed and refined iteratively through team consensus.

Two trained research assistants will independently code transcripts. Each transcript will be read multiple times to ensure familiarity with the data prior to generating initial codes. Coding discrepancies will be discussed and resolved through consensus during weekly team meetings. A subset of transcripts will be double coded to enhance analytic rigour. NVivo software will be used to support data management and coding. Themes and subthemes will be identified through iterative comparison across interviews and focus groups, with attention to patterns that recur across participants and stakeholder groups. Identified codes will be organised within the PEN-3 domains (Cultural Identity, Relationships and Expectations and Cultural Empowerment). Narrative synthesis will be refined iteratively until thematic consensus is reached.

Phase III: data analytic plan (pilot RCT)

Preliminary descriptive analyses

Baseline demographic and clinical characteristics will be summarised descriptively by treatment arm using means and SD for continuous variables and frequencies and percentages for categorical variables.

Primary outcome analysis

The primary outcome is stimulant abstinence, operationalised as percentage stimulant-negative office-based urine toxicology screens collected at each clinic visit during the 12-week treatment phase and the longest duration of continuous abstinence achieved during treatment. Because urine results are repeated and time-varying, generalised estimating equations (GEE) will be used to model abstinence over time while accounting for within-subject correlation. Comparisons will focus on the CACM arm relative to SCM and TAU.

Secondary outcomes include (1) treatment retention at 12 weeks; (2) treatment acceptability—measured using the AIM; (3) feasibility of intervention measured using the FIM and (4) treatment satisfaction post intervention measured using the CSQ-8.

Longest duration of abstinence will be analysed using linear regression models. Time-to-event outcomes (eg, time to return to stimulant use) will be analysed using Cox proportional hazards regression models. Effect estimates will be reported with 95% CIs using a two-sided alpha of 0.05. Given the pilot nature of the trial, emphasis will be placed on estimation of effect sizes and feasibility metrics rather than hypothesis testing alone.

Missing data and handling of incomplete assessments

Missing data are anticipated given the study population and study design. All primary analyses will follow an intent-to-treat framework. For urine toxicology outcomes, missed visits or failure to provide a urine sample will be conservatively coded as stimulant-positive in the primary analysis, consistent with prior CM trials. GEE models will use robust SEs to account for within-subject correlation over time. Sensitivity analyses will be conducted to examine the impact of alternative missing data assumptions, including models that treat missing urine samples as missing-at-random and, where appropriate, multiple imputation for secondary continuous outcomes. Patterns of missingness will be examined descriptively to assess whether missing data differ systematically by treatment arm or participant characteristics.

Data monitoring

Given the pilot nature of this trial and minimal risk to participants, a formal Data Monitoring Committee will not be established. However, study oversight will be conducted by the Principal Investigator (PI), study team and CAB, with regular review of recruitment, retention, protocol adherence and data quality. Any protocol deviation will be identified, addressed in study source documents and reported within five working days reported to the study sponsor and the IRB.

The study may be discontinued prematurely for several reasons, including but not limited to unexpected adverse events, safety concerns or a determination by the PI, sponsor or IRB that continuation of the research is no longer ethical or feasible. If such a decision is made, the PI will promptly notify the IRB and any relevant stakeholders, including funding agencies or collaborators as required. In the event of early termination, all study-related activities will be halted in an orderly and ethical manner. Participants will be informed about the discontinuation, and, if applicable, appropriate follow-up will be offered to ensure their well-being. Study data will be securely archived or disposed of in accordance with institutional policies, and a final report will be submitted to the IRB summarising the reason for discontinuation, number of participants enrolled and status of collected data. This approach ensures transparency, protects participant interests, and upholds the integrity of the research process.

Discussion

Inequities in treatment access and clinical outcomes reflect not only structural barriers but also limitations in how current evidence-based interventions—including CM were developed, tested and implemented—often with clinical trials that are not powered to examine heterogeneous sociocultural contexts. Indeed, emerging literature suggests that SCM protocols are not equally effective or acceptable across populations, underscoring the importance of cultural adaptation.⁵⁹

The need to culturally adapt CM for black adults with StUD is an equity-based proposition necessary due to multivariable and persistent racial disparities in uptake, access, engagement and treatment outcomes particularly given the emerging era of stimulant-opioid polysubstance use.⁶⁰,⁶³ The Cul-CM Study addresses this implementation gap through systematic cultural adaptation of CM for black individuals with StUD. Cultural adaptation involves a structured modification of evidence-based practices to align with the language, values, norms and lived realities of specific populations while preserving fidelity to core intervention components.^{64 65} In this study, cultural adaptation of CM is developed as an iterative and theory-informed process involving local needs assessment, community engagement, co-creation of knowledge, pilot testing and structured refinement prior to broader implementation. This approach builds on prior cultural adaptation efforts across a range of evidence-based interventions adapted for diverse populations and settings such as motivational interviewing, cognitive behavioural therapy, community reinforcement approaches, mindfulness-based relapse prevention and skills-based prevention programmes.⁶⁶,⁷¹

Prior CM adaptation research has been scant with limited contextual scaffolding and theoretical grounding, often focusing on demographic predictors of CM response rather than proactive community-informed adaptation processes. While some studies have addressed modifications across domains such as language and literacy, cultural values, contextual relevance and programme delivery,^{63 70 72} most of the available literature on CM adaptation is concentrated within American Indian and Alaska Native communities,^{24 73} offering limited guidance for adaptation across other ethnoracial groups. Although previous CACM interventions have demonstrated promising outcomes, gaps remain in adaptation rigour, generalisability and implementation strategy specification across diverse geographic and sociocultural contexts.^{22 51}

Further, while prior studies lack theoretical foundation, by embedding the PEN-3 cultural model within a structured ADAPT-ITT implementation framework, the Cul-CM study advances cultural adaptation beyond superficial modification to include a systematic, theory-driven implementation strategy. The integration of stakeholder perspectives, including individuals with StUD, family members, clinicians and community representatives, positions cultural adaptation as both a methodological and ethical imperative in the pursuit of equitable intervention delivery.

Other key areas of innovation in this study include:

1. Engagement of a high-risk population and multiple stakeholder groups (people with StUD, family members of people with StUD, clinicians and community stakeholders) as equal partners in the research process.

2. Use of complementary methods to inform intervention development including qualitative interviews, elicitation approaches, quantitative surveys and focus groups.

3. Explicit focus on culturally grounded determinants using the PEN-3 model.

4. Fidelity to a sequential, iterative, theory-driven adaptation framework (ADAPT-ITT).

5. Systematic generation of iterative adaptation drafts, incorporating expert review and CAB input at each stage.

Limitations

This study has several limitations. First, this study is conducted within a single geographic region and among black adults with StUD in community-based treatment settings in New Haven, Connecticut, which may limit generalisability to other regions, healthcare systems or racial groups. Moreover, the study is designed to generate a locally grounded, stakeholder-informed adaptation of CM for the participating clinical context, rather than to produce a universally generalisable adapted CM model. However, the use of a structured adaptation framework, detailed documentation of stakeholder input and systematic tracking of adaptation decisions will allow future studies to assess transferable components, which with local tailoring may provide a scaffold to determine how the adaptation process may be replicated in other contextual, geographic and service settings. Second, as a pilot RCT with a modest sample size (n=60), the study is not powered to detect definitive differences in clinical outcomes, and findings will primarily inform feasibility, acceptability and effect size estimation. Third, although the adapted CM intervention preserves core behavioural mechanisms, the separation between CACM and SCM conditions may be influenced by contextual factors within participating clinical settings, including provider characteristics and site-specific implementation dynamics.

Fourth, the disproportionate representation of black communities in the criminal justice system may elicit strong negative reactions and even traumatising experiences to urine-based testing. However, the study mitigates this by centering the experience of participants, clinician cultural humility training and leveraging a trauma-informed approach to urine collection.⁷⁴

Finally, while qualitative methods are used to inform cultural adaptation, the adaptation process may not capture the full heterogeneity of black communities across diverse cultural, socioeconomic and regional contexts. Future research will be needed to examine scalability, sustainability and effectiveness in broader and more diverse settings.

Conclusion

CM is one of the most effective evidence-based interventions for StUD. Yet its uptake and uneven outcomes remain limited underscoring persistent gaps between efficacy, implementation and equity. The Cul-CM study addresses this implementation chasm by conceptualising cultural adaptation not as mere cosmetic modification, but as a structured, theory-driven strategy grounded in community engagement, cultural relevance, while preserving fidelity to core CM behavioural mechanisms.

By integrating the PEN-3 cultural framework within the sequential ADAPT-ITT process, this study centres the lived experiences of black adults with StUD while preserving the reinforcement principles that underlie CM effectiveness. The mixed-methods design, iterative development of adaptation drafts and active engagement of stakeholders are intended to enhance feasibility, acceptability and scalability within real-world community treatment settings.

The proposed pilot randomised trial will generate critical data on feasibility, implementation outcomes and preliminary clinical effects, establishing an empirical foundation for future hybrid effectiveness-implementation trials. More broadly, Cul-CM contributes to the growing implementation science literature by offering a replicable framework for culturally adapting evidence-based behavioural interventions in ways that advance equitable access, engagement and outcomes. The Cul-CM study may inform scalable strategies for delivering CM and other behavioural interventions to populations disproportionately impacted by stimulant-related harms.