Stigma towards opioid use disorder in primary care remain a barrier to integrating software-based measurement based care

We conducted a qualitative study using focus groups to better understand clinician (MD, DO, advanced practice provider) perception of opioid use and OUD treatment and whether a computerized MBC model could assist clinicians with diagnosis and management of patients with OUD. Focus groups were conducted with outpatient primary care clinicians between January and February 2020 as part of a larger study investigating the feasibility and acceptability of universal screening and MBC for OUD utilizing a software program. We used qualitative analysis to assess clinician perception of opioid use and OUD diagnosis and treatment to better understand the readiness for incorporating universal screening for OUD within outpatient primary care clinics.

Clinics (19 total) in a north Texas community were contacted about study participation via email. Of the 19 clinics, 7 did not respond, 6 indicated they did not prescribe opioid medications and as a result did not perceive a need for OUD treatment within their clinic and were thus not interested in participating, 3 were not interested in participating and did not give a reason, and 3 agreed to participate. All the non-participating clinics were in a community-based practice setting; all of the clinics that agreed to participate in the focus group sessions were clinics based within an academic medical center.

Participants in the focus groups (n = 33) were office-based primary care physicians and advanced practice providers associated with family medicine, internal medicine, and infectious disease (Table 1); infectious disease clinicians were included as these physicians provide primary care to patients who are HIV + and were thought to have an enriched sample of patients with substance use disorders. Specifically, members of these three departments were invited to attend a focus group session held during their regular department conference time. Participants were informed that the focus group was open to all regular department conference attendees and that the session’s goal was to obtain feedback from medical clinicians regarding current practices for treating patients with OUD and the potential of using computerized MBC for OUD within their clinics. Participants were compensated with lunch. Participants were informed that participation was voluntary and that sessions would be audio recorded. To maintain anonymity, the requirement for informed consent was waived by the Institutional Review Board at UT Southwestern. The protocol and all study procedures were also approved by the UT Southwestern Institutional Review Board and all procedures were conducted following the guidelines and regulations. Data were collected as part of the National Institute on Drug Abuse (NIDA) Drug Abuse Clinical Trials Network (CTN) Study “Innovatively Increasing PCP Prescribing of Buprenorphine: Measurement Based Care and Integrated Electronic Solution (MBC4OUD; CTN-0090).

The research team conducted three focus groups (one per department) in January and February 2020 (prior to the onset of the COVID-19 pandemic). Focus groups were led by one of the authors (FK) and followed a semi-structured guide prepared by two of the authors (AMD, RW) based on a review of prior literature (e.g., [6, 3, 7, 19], clinical expertise in treating substance use disorders (AMD, RW), and expertise in qualitative data collection and analysis (FK). The guide targeted the following domains: [33] current practices regarding the diagnosis and management of OUD, [25] familiarity and opinions regarding OUD treatment options, [23] positive and negative aspects of offering treatment for OUD in the clinic, and [11] reactions to proposed method of supporting the diagnosis of OUD and treatment of OUD with computerized clinical decision support. These domains represent our goal of differentiating between system-level barriers that could likely be addressed via software-based MBC with clinical decision support and attitudinal barriers that would require different interventions (e.g., education to overcome stigma and misconceptions). Each focus group lasted approximately 60 min. Focus groups were digitally audio recorded and transcribed. Anonymous demographic data (Table 1) were collected at the beginning of each focus group.

Qualitative analysis focused on identifying themes from the transcribed focus groups. This analysis [12] took place in two phases: (1) content analysis focused on uncovering general themes in the data followed by (2) systematic coding and interpretation of the data.

The first phase of analysis followed an open coding approach [30]. Data coding was first conducted independently by three authors (AMD, RW, FK), a multidisciplinary team of an addiction psychiatrist, clinical psychologist with expertise in substance use disorders, and social worker with training in public health. Data coding began by brief review of the transcripts for the coding team to familiarize themselves with the material. From this review, the coders generated a consensus list of codes and definitions. Each code had a defined list of areas that were included and excluded for inclusion in the code. During analysis, the research team met weekly to discuss coding and preliminary insights emerging from the data. This phase culminated in a list of codes and a codebook with definitions for each code; no subcodes were identified or utilized. Discussion with the full research team led to the final coding scheme.

In the second phase of analysis, the research team systematically coded the data based on the coding scheme developed in phase one [12]. Each member of the coding team (AMD, RW, FK) independently coded each transcript. Discussion was utilized to reach consensus coding of each transcript and discussion and review of coding of each transcript helped to ensure consistent coding across transcripts. A summary report was created to organize data by code across transcripts. Once the data were coded, the research team developed interpretations of the main findings.

Participant characteristics can be found in Table 1. Most of the participants were white, not Hispanic or Latino, physicians, and saw between 0 and 50 patients per week.

Six major themes emerged from the focus groups and were utilized to develop the coding terms. These themes were “distinguishing between chronic pain and OUD,” “current practices with patients using prescribed or illicit opioids or other drugs,” “attitudes and mindsets about providing screening or treatment for OUD in your practice,” “perceived resources needed for treating OUD,” “primary care physician role in patient care not specific to OUD,” and “reactions to implementation of proposed clinical decision support tool.” Each theme is summarized below and Table 2 provides a list of the themes, definitions of each theme, and a representative example selected by the authors.

Stigma associated with OUD is pervasive in the US. The belief that prescribing buprenorphine is simply substituting one addiction for another or that patients will sell or misuse their buprenorphine prescription were highlighted as potential barriers in the current study. Such inaccuratebeliefs [27]REF) are likely reflective of the stigma associated with addiction, even among those who are educated to treat patients with a wide array of medical illnesses. A recent review detailed the existence of stigma between January 1, 1998 to December 31, 2018, finding that significant stigma among providers of medications for OUD was influenced by lack of training and a belief that abstinence is preferred as well as negative perceptions of OUD patients and the associated treatments for OUD [22]. For example, in the current study, one clinician stated, “we’re adding to that failure by increasing the ability of every Tom, Dick, and Harry to prescribe Suboxone without, you know, enough training… Because pretty soon we won’t have OUD, we’ll have Suboxone use disorder.”.” In fact, OUD is one of the only chronic diseases wherein the disease itself and the treatment are stigmatized (Note: the authors would like to thank Reviewer 3 for this insightful thought).

To alleviate the impact of stigma related to OUD, and substance use disorders more generally, continued education, training, and policy changes to improve the literacy surrounding the medical basis rather than the moral basis is necessary [1, 28]. Additionally, addressing the barriers to care that patients experience and improving the healthcare system can help reduce stigma. For example, incorporating behavioral health into primary care and reducing the distinction between mental health and substance use can increase access to care, resulting in a reduction of stigma. Instruction around stigma and OUD in medical education and training is also needed. Furthermore, the screening and assessment of patients’ substance use and accompanying symptoms may alleviate or comfort some patients that may anticipate encountering stigma during a conversation about their substance use.

Previous literature has highlighted the barriers associated with providing OUD treatment in primary care settings, for example, lack of referral sources and behavioral health support [15, 20, 29]. The capacity of the primary care setting to provide evidence-based treatment for OUD has, historically, limited the implementation of OUD treatment in primary care. The current study indicates that the capacity to provide efficacious OUD treatment in primary care settings is still limited. A theme that was intertwined through our data categories was the perceived complicated nature of patients with OUD. Physicians noted that patients with OUD often have comorbid conditions and previous research has found the complicated social environment that is frequent among patients with OUD can complicate treatment (e.g., homelessness, joblessness, familial stressors, etc.) [5]. However, previous work has also demonstrated that buprenorphine treatment for OUD in primary care is highly effective [32] and that adjunctive psychotherapy provides limited benefits beyond the benefit of medication [34], though a lack of referral options has been cited as a clear issue in referring patients for higher levels of care [18]. It is unclear whether the barriers we identified were related to a lack of knowledge regarding the demonstrated efficacy of buprenorphine in primary care or whether these attitudinal barriers persist despite this knowledge.

Our project was undertaken with the foundational belief that computerized measurement-based care (MBC) may help alleviate the burden of developing a standard protocol and clinical decision-support tool for treating patients with OUD expressed by the physicians in this study. The participants in the current study expressed that the computerized MBC application could be a helpful tool to incorporate into their practice, though their hesitancy associated with the burden of implementing the application highlights the need for future research to establish efficient implementation protocols. However, this is a small price to pay to develop an implementation protocol that could screen and assist with treatment of patients within primary care - as has been demonstrated in previous research utilizing a computerized MBC application to diagnose and support treatment of depression and anxiety [15, 18].

Though this analysis revealed some expected barriers – lack of resources (i.e., time, reimbursement) and referral sources and stigma – the existing professional identities of the providers may also play a role in their current lack of acceptance and readiness for OUD screening, diagnosis, and treatment, as unexpectedly revealed during the focus group sessions. Clinicians’ comments coded under the themes of “attitudes and mindsets about providing screening or treatment for OUD in your practice” and “perceived resources needed for treating OUD” highlighted the ways in which clinicians felt they would need significantly more training, skills, and support to provide care to patients with OUD. One potential interpretation of these findings is that the role of the “clinician who treats OUD” falls outside of the professional identity of the primary care clinicians in our sample. Professional identity, which is formed and reified throughout training and practice experience, can have an influential impact on openness to new treatments. As a result, providers feel more closely aligned with institutional norms that are more consistent with their in-group professional identity [14]. However, when outside forces, such as societal changes (in this case, new treatments for OUD) challenge the groupthink and result in institutional changes (for example, reimbursement for SUD treatment), providers may be more likely to hold onto their status quo beliefs and attitudes (in this case, the perceived complicated composition of OUD patients, lack of resources, and stigma).

The results of the current study indicate that additional work is needed to adequately and efficiently overcome the perceived barriers associated with diagnosing and treating OUD in primary care settings, most significantly alleviating associated stigma and providing behavioral health support and referral sources. While significant work has been done to reduce stigma and negative perceptions and attitudes associated with problematic opioid use, medical education should continue to be reformed to educate physicians more uniformly on substance use and substance use disorders, with a focus on reducing the prevalent stigma and misconceptions of the treatment of OUD demonstrated by the primary care clinicians in our focus groups. Specifically, continuing medical education for practicing physicians focused on the management and diagnosis of OUD may be informed by interventions that address the role of stigma, attitudes and feelings and physician identity as a barrier to the adaptation of new practice habits, as the clinicians reported seeing care for OUD outside of their identities as primary care clinicians. In addition, research is needed to develop standardized implementation protocols to implement MBC succinctly and efficiently in primary care without disrupting the daily workflow.

Limitations of the current study should be considered when interpreting the results. The participants in the current study are a part of a large academic medical system in Texas. These results may be less generalizable to other primary care practices, especially those that are independently owned or located in more rural areas of the country, or those that have more experience in treating substance use disorders. We also acknowledge that the qualitative nature of this study prevents us from drawing causal conclusions. However, a qualitative study provides us with more details into the attitudes and perceptions of OUD of the participating physicians.