3.1 Survey Results
3.1.1 Survey Respondent Characteristics
Out of the 127 physicians included in the analysis, the majority of respondents were allopathic doctors (83%), male (59%), white (67%) and nonhispanic (93%). Additional File Table 1 lists respondent specialties. Respondents most frequently specialized in family medicine (34%), addiction medicine (25%), and anesthesiology (15%). These providers serve approximately 263 unique patients in an average month (n=105).
Most respondents primarily practiced in outpatient primary care clinic settings (25%) or outpatient specialty clinics (14%), while only 5% practiced in an OTP. 35% reported that their practice facility was affiliated with a not-for-profit health center or hospital, 23% with an academic medical center, and 23% with a for-profit health center or hospital (Additional File Table 2).
45% of respondents indicated that they had a DEA buprenorphine waiver, although a small proportion were not currently using it (n=6/104). 40% of physicians with DEA waivers could serve up to 100 patients. 76% of prescribers reported that they had not obtained the Risk Evaluation & Mitigation Strategy (REMS) certification to implant Probuphine® as treatment for OUD and did not plan to in the future (n=77/101); 11% had the certification but were not currently implanting Probuphine® (n=11/101); and no respondents had the certification and were implanting Probuphine.
45% of respondents indicated that no one in their practice currently prescribed extended-release naltrexone; only 22% indicated that they or someone else in their practice prescribed the medication. Only 16% of respondents answering the question indicated they often or always referred patients with OUD for methadone treatment, while 48% said they “sometimes” and 29% “never” did so.
3.1.2 Provider Attitudes and Beliefs about MOUD Efficacy
Survey respondents had overall positive impressions of buprenorphine, extended-release naltrexone, and methadone for OUD treatment. However, there were some distinctions in beliefs about efficacy of the specific MOUDs. Table 1 depicts the comparison of respondent perceptions of the efficacy of buprenorphine and extended-release naltrexone. Respondents believed that buprenorphine, to a greater degree than extended-release naltrexone, decreases opioid cravings (paired t(52)=4.474, p<0.001)., decrease the risk of fatal opioid-overdose (paired t(51)=3.413, p=0.001), decreases return to opioid misuse (paired t(49)=2.078, p=0.043), and works well in patients with co-occurring mental health disorders (paired t(49)=2.461, p=0.017).
Table 2 shows the comparison of provider-perceived efficacy of extended-release naltrexone and methadone to treat OUD. Respondents believed that methadone, to a greater degree than extended-release naltrexone, decreases opioid cravings (paired t(51)=3.759, p=0.000), decreases risk of fatal opioid-overdose death (paired t(50)=2.349, p=0.023), decreases return to opioid misuse (paired t(49)=2.780, p=0.008), and works well in patients with co-occurring mental health disorders (paired t(49)=2.322, p=0.024). When comparing physician perspectives about buprenorphine and methadone to treat OUD (Table 3), respondents believed that buprenorphine is slightly more effective than methadone in decreasing the risks of opioid-overdose death (paired t(67)=2.147, p=0.035).
When comparing beliefs about MOUD efficacy among physicians with and without a DEA waiver, some significant differences emerged across medications. Waivered physicians agreed less strongly that buprenorphine is effective in treating opioid dependence in pregnant women, as compared to non-waivered physicians (paired t(67)=-3.911, p=0.000, Additional File Table 3). Waivered physicians believed that extended-release naltrexone treatment decreases the rate of return to opioid misuse to a greater degree than did non-waivered physicians (paired t(49)=2.143, p=0.037, Additional File Table 4). Finally, waivered physicians, as compared to non-waivered physicians, believed less strongly that methadone decreases risk of opioid-overdose death (paired t(71)=-3.097 , p=0.003, Additional File Table 5); decreases opioid cravings (paired t(70)= -3.203, p=0.002, Additional File Table 5), decreases rates of return to opioid misuse (paired t(62.573)= -3.668, p=0.001, Additional File 5), and is effective in treating OUD in pregnant women (paired t(65)= -4.397, p<0.001, Additional File Table 5).
3.1.2 Provider Perceptions of Barriers to Office-based MOUD Prescribing
Figure 1 summarizes prescriber beliefs about barriers to prescribing buprenorphine and extended-release naltrexone in office-based settings, using percentages to reflect the differing number of respondents for the two questions. The most common barrier to prescribing buprenorphine, according to DEA waivered physicians (n=47 respondents), was insurance prior authorization requirements (22%), followed by insufficient staff support (16%). Lack of support by managers/administrators at the practice was most commonly identified as a non-barrier (73%), followed closely by insufficient training (69%). As with buprenorphine, a commonly cited barrier to prescribing extended-release naltrexone (n=97 respondents) was insurance prior authorization requirements, as well as the lack of community resources for patient withdrawal management (each 16.5%). Concern about diversion was the most commonly identified non-barrier to prescribing extended-release naltrexone among all prescribers surveyed (42%). Table 4 shows the comparison of provider-perceived barriers to buprenorphine and extended-release naltrexone use for OUD. Respondents were statistically significantly more likely to be concerned about professional licensing board oversight (paired t(34)= 3.311, p=0.002) for prescribing buprenorphine as compared to extended-release naltrexone.
Table 5 depicts the comparison of DEA-waivered and non-waivered prescribers’ perceptions of barriers to extended-release naltrexone. Waivered providers, as compared to non-waivered ones, were more concerned about the following with respect to extended-release naltrexone to treat OUD: insufficient training (paired t(50)=4.076, p=0.000), insufficient time (paired t(51)=5.476, p=0.000), insufficient staff support (paired t(51)=3.762, p=0.000), insufficient experience (paired t(52)=5.175, p=0.000\1), insufficient resources for patient psychosocial support (paired t(53)=5.855, p=0.000), and insufficient resources for patient withdrawal management (paired t(53)=5.375, p=0.000).
3.2 Focus Group Results
Focus group participants were made up of 7 physicians (MD) across different states (PA, IL, FL, MO, ME, WA, CT), four of whom had a waiver to prescribe buprenorphine at the time of the focus group. Three identified as female and four as male. Participants in the focus groups provided more detail regarding 6 key themes identified in the coding process: MOUD efficacy, financial barriers to medications for OUD (provider- and client-side), treatment capacity, processes and procedures for treatment, provider competencies, and stigma. A list of selected quotations by theme is available in Table 6.
3.2.1 MOUD Efficacy
With respect to MOUD efficacy (Theme 1 in Table 6), focus group participants noted disparities in the evidence base for different MOUDs. According to one focus group participant, “[t]he evidence base behind [extended-release naltrexone] right now is actually really limited. And it’s one of the things that makes me the most nervous when we talk about [MOUD], lumping them all together.” This perception was borne out by survey results that indicated greater belief in the efficacy of methadone and buprenorphine as compared to extended-release naltrexone.
3.2.2 Logistical and Financial Barriers
The focus groups also highlighted financial and logistical barriers to providing MOUD treatment (Themes 2, 3, and 4 in Table 6). For example, participants raised concerns about the staff time and cost of acquiring necessary continuing education to provide MOUD, as well as the difficulties in ensuring a practice’s financial sustainability across the diverse MOUD billing codes and reimbursement rates. One provider stated that running an OUD program would lose money for their practice (quote 1a, Table 6). Providers also noted difficulties in establishing necessary workflows for providing MOUD, particularly in the context of multidisciplinary teams (quote 3c, Table 6). Finally, many focus group participants cited the lack of addiction treatment providers within their community as a significant barrier to patients (quote 2a, Table 6).
3.2.3 Provider Perceptions and Stigma
Focus group participants also emphasized the negative or uninformed perceptions associated with training for and treating patients with OUD and expressed a reluctance to treat what they perceived to be a potentially challenging population (Themes 5 and 6 in Table 6). Several providers expressed concern with their knowledge and the training demands to treat patients with OUD (quotes 4a and 4b, Table 6). One provider raised concerns about practices, particularly large ones, attracting a patient population dominated by persons with OUD (quote 5c, Table 6). Providers did emphasize the importance of psychosocial support as a component of OUD addiction treatment services, in addition to MOUD (quote 5a, Table 6). Another participant said that providers do not feel comfortable talking to patients who screen positive for OUD, often lack the knowledge to provide behavioral health support, and do not have access to on-site support from counselors or psychologists/psychiatrists (quote 4a, Table 6).