Protocol for the implementation of a statewide mobile addiction program

Drug overdose continues to claim the lives of individuals in MA, USA and across the country at epidemic rates. In 2020 there were an estimated 2106 deaths from opioid related overdose in MA, USA [1], nearly six deaths per day. The overdose crisis was exacerbated by the COVID-19 pandemic, with increased barriers to addiction services caused by restrictions designed to limit disease transmission as well as increased isolation due to risk of infection [2]. Evidence-based interventions proven to reduce death and disease among people who use drugs, include medications for treating opioid use disorder (MOUD) [3], distribution of naloxone (to reverse an opioid overdose) to people who use drugs and to their close contacts [4] and providing safer consumption equipment (to reduce infections such as HIV) and counseling [5]. Despite these strategies to reduce overdose death rates, people who use drugs – especially those who are unhoused – face ongoing barriers to addiction treatment and primary and preventive care. These barriers include limited access to MOUD and other addiction services [6], regulatory restrictions, competing priorities and stigma in the healthcare system experienced by people living with substance use disorder (SUD) [7,8].

In the case of MOUD, success can be measured via the ‘treatment cascade of care’ (see effectiveness component in Table 1), a framework for studying opioid use disorder (OUD) services use [9]. This framework seeks to identify transitions where patients are most likely lost to follow-up (e.g., between the receipt of prescription and prescription fill, or between filling the 2nd- and 3rd-month prescriptions) [10]. People experiencing homelessness have a high drop-off rate from office-based OUD treatment, an effective intervention to reduce mortality risk [10]. Lowering the barriers to OUD treatment, for example, by providing mobile OUD services, may lead to reduced mortality as well as building trust and greater engagement in office-based OUD treatment [10]. For health services more generally, current literature has shown that mobile health clinics are successful in reaching and delivering services to vulnerable populations and in removing barriers to care [11].

By combining elements of clinical care, harm reduction and mobility, many programs have found success at serving people living with SUDs. The Seattle-based Safe, Healthy, Empowered Clinic implemented a mobile clinic model to bring low-barrier medical care to socially marginalized women at a drop-in center, resulting in decreased emergency room services for outpatient needs and increased accessibility of medical and social services [12]. Similarly, The spot mobile clinic in Baltimore was co-located with syringe services programs (SSPs) with the objective to integrate buprenorphine (one of the three US FDA-approved MOUD) services and medical care for people who use drugs with harm reduction services already offered at the SSPs. Researchers found that individuals who received a prescription for buprenorphine at The spot mobile clinic were more likely to receive other preventative services including vaccination, naloxone distribution and HIV and sexually-transmitted infections (STI) testing [13].

Finally, the Sex Worker Promoting Action, Risk Reduction and Community Mobilization (SPARC) drop-in center in Baltimore augmented its programming with mobile outreach services, resulting in reduced incidence of HIV and sexually transmitted infections in female sex workers using drugs [14]. Despite the findings described above, concerns over mobile programs include a fear of a diminished role and funding for traditional primary care, lack of patient follow-up in mobile settings and lack of sustainability [15]. Nevertheless, mobile health services are a portal of entry for individuals at highest risk for overdose death, many of whom are not currently engaged with traditional healthcare for myriad reasons including stigma, insurance issues, lack of transportation and competing comorbidities.

In Massachusetts, the Community Care in Reach^® (CCiR) model builds on decades of mobile harm reduction and HIV services funded by the state’s Department of Public Health (DPH) and involves a mobile unit with integrated harm reduction services and clinical care for people who use drugs [16]. The infographic illustrates how CCiR combines service elements of office-based clinical care programs and harm reduction programs while eliminating barriers posed by these traditional models (Supplementary Figure 1). We note that while there are several FDA-approved medicines to treat various SUDs, we include only buprenorphine specific outcome measures. This is because OUD is a priority of this model, the restrictions on methadone provision and buprenorphine treatment is the most frequently used treatment medication provided by the programs.

The CCiR model allows for a flexible, rapid response of harm reduction and clinical services and adapts to the changing overdose geo landscape. Programs locate mobile vans in areas of greatest need based on overdose data from DPH. For example, mobile outreach sessions are often based in locations that see consistently high rates of overdose. Additionally, the schedule and location of these vans are re-evaluated when data point to a new or emerging need [16]. The model leverages its mobility, having the flexibility to alter the deployment locations and schedule to respond to the changing needs of the community in a way that traditional brick-and-mortar sites (including opioid treatment programs) cannot.

Despite the effectiveness of MOUD in preventing overdose deaths, it was estimated in 2017 over 70 percent of people who needed treatment for OUD did not receive it [17]. CCiR is designed to eliminate barriers, by mobilizing essential services such as harm reduction and clinical care including MOUD and bringing these services directly to individuals at greatest risk of near-term death from overdose. CCiR is well-suited to expand access to underused MOUD, provide a broader spectrum of services that is less frequently offered in traditional brick-and-mortar clinical care. A core element of mobile vans is the provider building a trusting relationship with the patients, thus re-engaging individuals in care who had been previously disconnected due to finances, stigma and other barriers. By prioritizing communities of color, access expansion through CCiR’s mobile programming can also reduce inequities currently seen in addiction treatment and service utilization.

The evaluation of this intervention is designed to capture the cumulative effect of combining harm reduction services, clinical addiction care and mobility to serve populations at high risk for overdose. Whereas each of these evidence-based programmatic elements have been evaluated individually [4–6,11], there is a dearth of information on the cumulative impact of combining all three elements in one service model. We outline here an analysis that uses the RE-AIM framework to measure the impact of CCiR on its target population as well as the experience of the program staff in implementing the model. The results will be essential in understanding the benefits of reducing service silos by using a more comprehensive approach to low-barrier addiction care.

Based on the success of the Boston pilot [16], the CCiR model was expanded starting in 2020 to four programs across Massachusetts (Figure 1). After signing contracts that include model requirements, the new programs received funding from the Massachusetts Department of Public Health, Bureau of Substance Addiction Services (MDPH-BSAS) and technical support from The Kraft Center for Community Health at Massachusetts General Hospital (Kraft Center). Each program secured a specialized vehicle equipped to provide medical care with a sink, medical supplies, Wi-Fi and a private medical examination area. Preparation of implementation began in 2021. Data collection began in 2022 and will run through 2024.

In keeping with the CCiR model, each program’s team includes at least one clinician practiced in buprenorphine treatment and as many as three outreach workers trained in harm reduction strategies, among other support staff. Outreach workers reflect the diversity of the communities they serve and hiring decisions have prioritized staff members with lived experience with SUD. Programs that are not already SSPs themselves partner with a local agency to ensure staff receive necessary naloxone and other harm reduction trainings on an ongoing basis. The harm reduction workers and clinical practitioners provide services both near the parked mobile unit and on board the vehicle depending on the needs of the individual requesting services. Services include the prescription and initiation of MOUD such as buprenorphine; the distribution and collection of safer consumption supplies, such as syringes and glass pipes; distribution of naloxone kits; preventive care such as immunizations and screenings; provision of community drug checking services, such as fentanyl test strips; chronic disease management such as HIV treatment, hypertension management and diabetes management; skin and soft tissue infection care; and referrals for social and other medical services, including specialty care and behavioral health services [18–20].

The four programs implementing the CCiR model prioritize cities, towns and neighborhoods experiencing consistently high rates of substance use and overdose. Data for siting clinic locations comes from a combination of publicly available, quarterly Massachusetts overdose reports from the Massachusetts Department of Public Health, municipal data on emergency medical services transports for narcotic related events and anecdotal information acquired from SSP staff and other addiction and homeless support services. The Kraft center receives data from Boston emergency medical services in a very timely manner through a partnership with the City of Boston. The quarterly MDPH data has a lag of approximately 4 months.

Any individuals experiencing barriers to primary care and addiction services are eligible to participate in the CCiR model, including those who are not currently engaged with the healthcare system in any capacity and those whose current care providers do not offer access to addiction services. There is no specific age requirement. Adults with SUD not currently engaged in regular care represent the core population served by the program. The staff in mobile vans identifies potential participants through: regularly canvassing areas around clinical sites; raising awareness of the program with local stakeholders and community leaders who provide referrals; and encouraging word-of-mouth promotion of the program. The CCiR model also emphasizes equity, ensuring clinical sites include locations in communities of color and that services are provided by staff that reflect the diversity of their patient population.

Implementation of the CCiR involves a several key players (Figure 1). The Kraft Center, based on its involvement with the pilot program in Boston, provides technical assistance. The Kraft Center is housed within Massachusetts General Hospital. It’s primary mission is to expand access to high quality healthcare for vulnerable populations by developing innovative programming, piloting innovations locally and disseminating effective programming state and nationwide. Kraft Center owns three of the four vehicles and leases them to the programs. The Kraft Center hosts virtual monthly technical assistance sessions to provide guidance on program specific issues, foster cross-program learning, share best practices, troubleshoot problems, hear expert speakers and explore common themes and issues. These sessions foster a learning collaborative and a supportive network. Learning collaboratives have been used for many initiatives and can improve outcomes by facilitating collaboration through sharing of resources and ideas among participants [21–25].

MDPH-BSAS funds the services provided on the vans and provides data needed for the evaluation. Brandeis University is conducting the evaluation component. Data collection tools and protocols are standardized across the four programs to allow for cross-program analysis and evaluation of fidelity to the model. Mobile van programs complete a standardized intake form on program patients for MDPH, as well as a monthly online survey (Qualtrics Survey software) for the Brandeis evaluation team that captures aggregate data for outreach encounters and clinical services provided. Once every 6 months, also via Qualtrics Survey software, programs will report on measures related to initiating and engaging patients on OUD medication treatment, (the ‘OUD treatment cascade of care’, see Table 1).

Informed consent for the evaluation is not being obtained since personal data are not collected by the Brandeis evaluation team. This protocol has been approved by the Institutional Review Boards for Massachusetts Department of Public Health and Brandeis University.

The RE-AIM framework [26] informs all aspects of this evaluation (Table 1). RE-AIM identifies the components of a program that produce valued outcomes and reach the target population, while also detecting practices that may hinder or reduce health inequalities in a community [27]. The framework includes five core dimensions: ‘Reach’, ‘Effectiveness’, ‘Adoption’, ‘Implementation’ and ‘Maintenance’. Each dimension evaluates the impact of the program on either the individual and/or organization level. ‘Reach’ and ‘Effectiveness’ are individual-level measures representing the size and characteristics of the population participating in the program and highlighting whether the program impacts the health outcome. ‘Adoption’ is a measure at the organizational level that analyzes the uptake and acceptance of the program protocol among staff members, practices and settings, while ‘Implementation’ focuses on whether staff members deliver the program as intended. ‘Maintenance’ is measured at both the individual and organizational level and analyzes the extent to which the program is sustained over time [28].

Using both quantitative and qualitative methods, RE-AIM will help to evaluate the implementation of the CCiR model by the four organizations and determine its effectiveness in expanding access to addiction services, specifically MOUD in communities of color and high-risk individuals. It will also help to gain an understanding of whether buprenorphine can be successfully prescribed to the target population in a mobile setting.

Data will include all individuals seeking medical or harm reduction care and each type of encounter is recorded differently. Data collection among people who use drugs and alcohol, who are often transient, and who can be skeptical of sharing personal information poses several challenges. To address this, MDPH-BSAS and evaluation staff conducted a preliminary interview of the programs implementing the CCiR model to capture their current data collection systems and best practices. Since priority for this vulnerable population is placed on building trust, developing rapport and providing ultra-low-barrier, evidence-based services that improve public health and save lives, staff would only be able to collect a limited amount of data on individuals utilizing harm reduction services. Thus, collecting variables such as sociodemographic data in a harm reduction encounter may inhibit utilization of services [29].

All encounters of clients seen for the more specific clinical care will result in generating a record in the agency Electronic Medical Record and will have those services recorded as usual in the agency Electronic Medical Record. CCiR programs will complete an intake form in a system maintained by MDPH-BSAS. This intake form includes demographic information such as gender, ethnicity, race, education, housing status, income and insurance type. No subsequent patient-level data will be collected during follow-up clinical visits for this evaluation project. MDPH will send deidentified data to the Brandeis evaluation team quarterly. No data from individuals served will be connected on an individual level to data provided by MDPH-BSAS.

Collectively, the data collection tools specifically developed for this project allow the Brandeis evaluation team to track the population accessing clinical and harm reduction services across the four programs. In addition, data specialists from the four programs will submit aggregate monthly metrics via an online survey tool developed specifically for this project by the Brandeis evaluation team. Data elements include hours of van operation, frequency of harm reduction services accessed, frequency of clinical services accessed, sexually transmitted infection tests administered, buprenorphine services provided, services most frequently requested and referrals to regular care in more traditional healthcare settings. Again, the evaluators kept the number of metrics to a minimum to maximize trust between care giver and this population. These metrics will be available to implementing programs, Kraft Center and MDPH-BSAS to see progress via a monthly ‘dashboard’ report, including the reach of both harm reduction and clinical services.

The online survey tool will also track the OUD treatment cascade of care. Programs will report on clients with new OUD diagnoses; the number that initiated medication; and the number that subsequently remained in treatment at 28, 60, 90 and 180 days. These metrics will be completed by the implementing programs every 6 months, starting in January 2023 to allow a full 180 day runout period for clients served in the first reporting period (i.e., January–June 2022).

In addition to quantitative data, the evaluation will include qualitative data. Brandeis will conduct interviews with staff from each of the four programs. The interview protocol will follow the framework outlined in RE-AIM. The evaluators will inquire about how patients are referred in and out, how clinical and harm reduction activities are offered, barriers and facilitators to implementation of the model and perceived community reception. These questions will elucidate metrics such as fidelity to the model and community reception on both the intra-organizational level (e.g., “In what ways has the van influenced services provided by the parent organization?”) and an extra-organizational level (e.g., “How would you describe the communities’ response to your services?”).

Brandeis will also conduct a visit to each van as a non participant observer, to better understand the logistics and processes of treating people in a mobile van setting. In addition, the evaluators will attend monthly TA sessions where all four programs gather in a ‘learning collaborative’ environment. In TA sessions, program staff report their impressions and feedback regarding implementation issues at each site. For example, discussion at the TA sessions can inform the evaluators regarding difficulties in staffing issues, personal safety concerns and challenges arranging entry of study data. The evaluation team will record notes from these activities to use in qualitative data analysis.

The evaluation will use a thematic analysis approach to the data to highlight lessons learned, best practices, as well as information about staffing and community engagement. The overall qualitative analyses will help us to understand reasons for differences between the four programs, unexplained variables in the quantitative analysis, and any shifts in the implementation process.