Determinants of opioid use disorder (OUD) and barriers to treatment

A combination of individual, social, and structural factors fuels the opioid overdose crisis. Drug supply and overprescribing of opioids are often touted as key contributors, while the intersection of racism, poverty, issues of unstable housing or homelessness, trauma, access to culturally-appropriate care, social networks, isolation, and feelings of hopelessness—among other determinants—receive less attention from policymakers and program developers [4–6]. The criminalization of drug use, particularly among the Black community, and the perpetuation of fear and stigma in the media has historically slowed efforts to confront the epidemic [4].

Medication for opioid use disorder (MOUD)—such as buprenorphine, buprenorphine/naltrexone, and long-acting naltrexone in conjunction with counseling services—are gold standard, evidence-based methods of reducing overdose incidence and mortality [7]. A Maryland-based study found the risk of overdose death to be 82% lower among people with OUD when receiving MOUD versus the alternative of no engagement in MOUD [8]. Beyond a reduction in death rates, MOUD has additional benefits including higher odds of abstinence, employment, housing, and engagement in primary health care [9–13].

Nevertheless, perceived and real barriers on the patient and provider side have created an environment in which only one in four who need MOUD receive it [7, 14, 15]. Despite recent changes in US policy to increase the availability of MOUD, locating programs and providers capable of prescribing and administering MOUD is a challenge [16–19]. Moreover, lack of motivation and misinformation about MOUD (e.g., replacing one drug or addiction with another), paired with internalized, enacted, and anticipated stigma surrounding MOUD and substance use, often keep patients from engaging in MOUD [17, 20, 21]. Further, negative views of MOUD and a lack of provider training and clinical resources may inhibit the implementation of necessary care [21, 22].

Community-based participatory research

Community-based participatory research (CBPR) is a process that engages and includes community members in all stages of program development to increase collaboration, gain community input, and increase program relevance [23–25]. The benefits of a CBPR approach are numerous, including: establishing mutual trust across community systems; gaining community perspectives in developing program components; gaining an understanding of community history, culture, and dynamics, and how program strategies used in other communities may or may not apply to the DC setting; and establishing relationships across communities to share resources and build research capacity. Given that the opioid epidemic in DC largely affects Black males ages 40–69 years, CBPR approaches may be particularly useful in understanding barriers and facilitators to accessing MOUD among this population which in turn can be used to develop strategies to increase MOUD engagement [26–30]. In this study, we use a CBPR method called “concept mapping” [31–33] to explore factors associated with MOUD engagement to develop a community-engaged response to increase MOUD uptake.

Overview and concept mapping

We conducted a pilot study called Welcome MAT that employed a mixed-methods concept mapping (CM) approach to identify facilitators and barriers to engagement in MOUD among service providers and persons with OUD in DC. CM includes obtaining a wide breadth of ideas about a topic from the general community (“brainstorming”), categorization and prioritization of those ideas (“sort and rate”), and community-based interpretation of the findings [31]. Although CM has been used to understand a variety of public health challenges and co-develop community-based programs [34–37], to our knowledge, CM has not been utilized to understand and develop strategies to improve MOUD engagement.

Study population

The Welcome MAT study identified two primary populations of interest: persons with lived experience of OUD (referred to as consumers) and service providers. OUD is defined as a chronic disorder characterized by the chronic use of opioids that causes clinically significant distress or impairment. Consumers were individuals actively injecting or using opiates (<6 months) or who were entering or had been in recovery (not used/injected in ≥6 months) and may or may not have used MOUD services previously. “Service provider” was an intentionally broad term to include those whose work focuses on substance use issues, general health care providers who care for persons with OUD, and others from community-based organizations (CBOs) who serve the population impacted by OUD. Providers included primary care providers; emergency department providers; infectious disease specialists; clinical, behavioral health, or social support staff; and community health workers/peer recovery coaches.

We also formed a Welcome MAT community advisory board (CAB), whose members were consulted prior to the initiation of CM and later in the interpretation phase. The CAB consisted of individuals at various levels of involvement with persons with OUD in DC, including those with lived experience; people who provide harm reduction, treatment, and social services to persons with OUD; and providers and other professionals (e.g., DC Department of Behavioral Health, community-based organizations, peer recovery coaches, and academic researchers). These key stakeholders assisted in 1) establishing the current state of MOUD and other harm reduction programs in DC, 2) refining the research question, focus prompt, and CM methods, and 3) interpreting the study findings. CAB members were compensated for their time and expertise. The CAB included representatives from organizations that have been working with persons with OUD in the Washington, DC area for more than 30 years.

Recruitment

The core activities of this study—brainstorming, sorting, and rating—were conducted from June 9, 2021-August 10, 2022. Several methods were used to recruit study participants to each activity. Flyers were distributed to CAB members, community-based organizations that work directly with consumers, and other local listservs (e.g., DC Center for AIDS Research, area MOUD provider lists, and primary care physicians). Additionally, the study was advertised via Craigslist, targeted social media posts, and institution-specific registries of potential study participants. As an incentive, we conducted a raffle of five $10 e-gift cards to participants who took part in the brainstorming phase using a random number generator. Consumers who participated in the sorting and rating phases received $50 gift cards.

Concept mapping procedures

CM incorporates qualitative and quantitative analysis techniques through a modified four-step process [31]: 1) community brainstorming; 2) community sorting and rating; 3) multivariable statistical analyses; and 4) community interpretation and utilization of concept maps.

Sample description

In the sort and rate phase, the study sample consisted of 23 consumers (sort and rate) and 34 providers (rate-only). Most (95%) consumers were middle-aged Black men (53.8 years, SD = 10.9). More than half (52.4%) were Veterans, as indicated by their use of Veterans Administration insurance coverage. Prior experience using or injecting opioids was a pre-requisite to be in the study (whether in active addiction or recovery), and heroin (57.1%), marijuana (57.1%), and opioids (47.6%) were the most used substances among participants in the six months prior to sampling. Methadone and buprenorphine were the most common treatment options for those currently in treatment (23.8%) or with any history of treatment. All consumers who responded to the survey had been part of a drug treatment program at least once, with nearly one in five (19%) engaging in treatment “five or more” times. Consumers received treatment primarily through community clinics and hospital emergency departments. Within the prior 12-month period, 43% of consumers had experienced a heroin- or opioid-involved overdose.

Providers in the sample were 44.7 years old on average (SD = 12.5) and identified across different racial groups (45% non-Hispanic white, 35% Black, 15% Hispanic, 10% Asian, and 5% AI/AN). More than half (60%) identified as female. They reported working primarily in hospitals and Federally Qualified Health Centers. Their expertise also reached into other settings such as mental health centers, needle exchange programs, and pharmacies. Providers noted that buprenorphine and naltrexone were the most common treatment options offered in their clinics and workplaces, with one having offered methadone. At the time of sampling, only one in four were waivered to prescribe buprenorphine, which likely represented the mix of clinicians and non-clinical personnel in the study. Demographic characteristics for each group are shown in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Sample demographics.

https://doi.org/10.1371/journal.pone.0306931.t001

Concept mapping results

Brainstorming, sorting, and rating.

The 70 unique statements generated during the brainstorming phase provided direct insight into the facilitators and barriers to engagement in MOUD. After reviewing the results with CAB members, our research team determined that seven clusters best encapsulated the data, outlining the following topics: 1) availability and accessibility, 2) hopelessness and fear, 3) basic needs not being met, 4) characteristics of treatment program, 5) understanding and awareness of treatment, 6) personal motivations, attitudes, and beliefs, and 7) easier to use drugs. Fig 1 displays the cluster map overlaid with the point map, and the corresponding statements grouped by cluster are shown in Table 2. The mean ratings of importance in Table 2 are inclusive of both consumer and provider rating data (n = 57). The subject matter of each cluster is explored below, in descending order from highest to lowest importance (i.e., mean cluster rating).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Cluster map.

Each point on the cluster map corresponds to a participant-generated statement, listed in Table 2. The proximity of one point to another indicates that participants (i.e., consumers) sorted them together more frequently and found them to convey similar meanings.

https://doi.org/10.1371/journal.pone.0306931.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Statements by cluster.

https://doi.org/10.1371/journal.pone.0306931.t002

Pattern match.

We next generated a pattern match to explore the relative importance of these seven themes and what level of agreement existed between consumers and providers (Fig 2). All participants generally found the subject matter of the statements to be important, rating them 3 and above on the scale (i.e., “moderately important,” “important,” or “very important”), but the pattern match illuminated discrepancies between the two groups’ rating data. The y-axis for consumers demonstrated that the average importance rating for the seven clusters was between 3.81 and 4.22. For providers, ratings were slightly lower, between 3.22 and 3.96. Both consumers and providers shared the same top three clusters (“hopelessness and fear,” “availability and accessibility,” and “basic needs not being met”). The order in which participants prioritized them was not a perfect match. Notable inversions in the figure include “personal motivations, attitudes, and beliefs” and “understanding and awareness of treatment.” Consumers placed personal motivation higher on the list of importance and awareness at the bottom, while providers gave more weight to awareness/promotion in improving access to and engagement in MOUD.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Pattern match.

The pattern match shows the level of agreement between consumers and providers by separating the two groups’ rating data. The plotted points along the y-axes represent the aggregate cluster ratings using a 1–5 scale of importance (in response to “Rate this item in terms of importance to accessing MAT”). Both groups shared the same top three clusters (“hopelessness and fear,” “availability and accessibility,” and “basic needs not being met”), although the order of prioritization varied.

https://doi.org/10.1371/journal.pone.0306931.g002

Summary of findings

In this study, we identified seven themes based on participant self-reports of barriers to MOUD initiation and use. These themes encompassed various domains including withdrawal, fear, hopelessness, basic needs, and characteristics of treatment programs. Study findings showed that there is an urgent need for treatment programs to adapt certain strategies including extended clinical hours, expanded telemedicine and mobile services, an increased presence of peer recovery coaches, and improved coordination of services for consumers (e.g., housing assistance, food vouchers, and flexible take-home medication doses). Study findings also highlighted social-structural factors such as poverty and stigma, which contribute to addiction and challenge engagement in MOUD among our predominantly Black sample. These factors undergird Black men’s experiences with MOUD and care, and underscore the need for culturally-appropriate care to address them [4–6, 41]. Collectively, findings from this study demonstrated the complexity of opioid use and engagement in MOUD, and further support the importance of community-engaged approaches as a fundamental component of the public health response to the opioid crisis.

This study builds upon existing research on barriers to MOUD, specifically focusing on three key areas: accessibility of MOUD in various settings, including clinical spaces and consumer hotspots, lack of provider training and clinical support services for people seeking MOUD, and stigma surrounding MOUD and substance use [20–22, 42, 43]. Additionally, findings revealed that misinformation and disinformation about MOUD may be a growing problem among consumers, which not only hampers utilization of MOUD but also perpetuates negative attitudes towards treatment. For example, brainstorming statements included several misconceptions endorsed by consumers such as the belief that MOUD is ineffective, difficult to use, and mimics withdrawal symptoms. While many studies emphasize the need for improved access to MOUD and enhanced provider training [44–46], few reported on the consequences of misinformation and disinformation among consumers. Future research is needed to develop targeted messaging and marketing strategies for both providers and consumers. These strategies should be designed to dispel myths and misinformation about MOUD particularly related to efficacy, potential side effects, and costs.

Several clusters identified in this study described known social-structural determinants of OUD, such as poverty, food insecurity, and housing instability [4–6]. Given that this study was conducted during the height of COVID-19, legal barriers, limited access to technology, and the aforementioned social-structural determinants may have contributed as significant barriers to treatment and recovery [46, 47]. Although other studies indicated that people are more likely to sustain housing, employment, and recovery once they receive MOUD, the lack of stability and security in these areas greatly hinders treatment engagement [9–13]. Participant statements further confirmed the need for MOUD programs to be coupled with initiatives that provide housing stability and fulfill basic needs. For instance, the Housing First model or community-based case management interventions could be adapted to fit different settings, local needs, and treatment modalities [48, 49]. To improve nationwide access to care, it will be vital to continue to reduce barriers in treatment policies, reduce out-of-pocket expenses for insured individuals, and increase financial resources to support uninsured people throughout the treatment process. Moreover, although the provision of MOUD through telemedicine improved retention and health outcomes during the COVID-19 pandemic, inequities persist, with lower utilization of such services among people who are Black, AI/AN, and Asian or Pacific Islander [50]. Additional research is needed to identify best practices in delivering MOUD through telemedicine and to address broader barriers to MOUD among racial and ethnic minority populations.

Stigma towards people with OUD, which leads to experiences of isolation and discrimination, emerged as a key barrier in our study. There was a widespread belief reflected in the statements collected (e.g., statements 38 and 58) that MOUD simply replaces one drug addiction with another, which is consistent with other findings [21, 51]. Our CAB further described how these beliefs are commonly held within the broader community, exacerbating the stigmatization of people engaged in MOUD and ultimately serving as another barrier to treatment.

The CAB’s interpretation of the clusters and statements emphasized the critical need to develop innovative strategies that raise awareness about MOUD and the treatment process among the broader community including family members and employers. In addition, effective dissemination of accurate information could have a positive impact on consumers by increasing knowledge about the effectiveness of MOUD and fostering a supportive environment that encourages help-seeking behavior. For providers, there may be numerous opportunities to combat stigma towards populations that use drugs, including targeted education and training that avoids blaming individuals for opioid use [51]. The statements collected in this study further highlight the importance of addressing other aspects of patients’ lives, including their mental health, and connecting them to additional resources. This approach allows providers to demonstrate care for the consumer, recognize the consumer’s humanity, and establish a treatment environment that is more inclusive and destigmatizing.

Strengths and limitations

Community engagement was a considerable strength of the study. By listening directly to consumers with lived experience of addiction and providers operating in treatment environments through each phase of the study, from brainstorming to interpretation, we were able to elicit meaningful and relevant findings and recommendations. Further, the most salient themes to emerge from the results aligned not only with CAB members’ experiences but also with existing literature on facilitators and barriers to MOUD. Another strength was the use of a mixed methods CM approach which contextualized, added depth, and resulted in an interpretable visualization of facilitators and barriers to MOUD.

The study had some limitations. The sorting phase for CM was complex, and the results did not fully capture the nuanced context within which this phase of the study occurred (i.e., conversations between participants and the research team regarding why statements were placed in certain piles over others). Nonetheless, CM has been successfully used with substance-using populations, and we followed best practices for data collection with this group [35, 52, 53], and the interpretation session with the CAB confirmed the validity of clusters. Another limitation was the sample size, which ideally would have been larger if not for the restrictions imposed by the COVID-19 pandemic. Nevertheless, the sample size was adequate for CM procedures [40]. Lastly, while the characteristics of our sample (e.g., majority Black men, veterans, engaged in drug treatment previously) reflect a portion of the consumer population of DC [2, 3], it does not reflect the population entirely and caution should be used when interpreting results with regards to sample bias. We acknowledge the rates of overdose and the makeup of persons with OUD in DC are unique, with higher rates than the national average and a population skewing older, which may limit the generalizability of study findings. Future studies should employ CM and CBPR to gain a better understanding of perceived facilitators and barriers to treatment among younger people with OUD, a growing demographic in the opioid crisis.