Use of Telemedicine for Opioid Use Disorder Treatment –Perceptions and Experiences of Opioid Use Disorder Clinicians

Abstract

Objective:

To understand clinician use of and opinions about telemedicine for opioid use disorder (tele-OUD) during the COVID-19 pandemic.

Methods:

An electronic national survey was administered in fall 2020 to 602 OUD clinicians recruited from WebMD/Medscape’s online panel. The survey completion rate was 97.3%.

Results:

On average, clinicians reported that 56.9% of their visits in the last month were via telemedicine (20.3% via audio-only and 36.6% via video). Most respondents (N=376, 62.5%) agreed that telemedicine has been as effective as in-person care. The majority (N=535, 88.9%) were comfortable using video for clinically stable patients, while half (N=297, 49.3%) were comfortable using video for patients who are not clinically stable. After the pandemic, most respondents (N=422, 70.1%) preferred to return to in-person care for the majority of visits; however, 95.3% thought telemedicine should be offered in some form. Most (N=481, 79.9%) would continue to offer telemedicine if reimbursement were the same as in-person, while 242 (40.2%) would continue if reimbursement were 25% lower. Clinicians with more OUD patients used more telemedicine and reported higher comfort levels treating clinically unstable patients, and clinicians with more Medicaid/uninsured patients used more audio-only and preferred to continue using telemedicine post-pandemic.

Conclusions:

Telemedicine made up the majority of OUD visits provided by surveyed clinicians, and the vast majority of clinicians would like the option to offer telemedicine to at least some of their patients in the future if there is adequate reimbursement. These findings can help inform telemedicine’s future role in the treatment of OUD.

1. Introduction

In 2019, an estimated 1.6 million people were diagnosed with opioid use disorder (OUD) in the United States;¹ however, only 18% received treatment,¹ and there were nearly 50,000 opioid overdose deaths.² The novel coronavirus disease 2019 (COVID-19) pandemic has exacerbated opioid misuse, with a resulting increase in overdoses.^3–5

COVID-19 has also driven a dramatic shift to telemedicine. Public health guidelines, stay-at-home orders, and policy waivers relaxing telemedicine restrictions have led to an unprecedented increase in telemedicine use,⁶ including telemedicine for treatment of OUD (“tele-OUD”). Prior to COVID-19, use of telemedicine for substance use disorders had been increasing but was relatively infrequent.⁷ Low use rates were due in part to regulatory and reimbursement barriers, and have been seen as a missed opportunity due to telemedicine’s potential to expand access to OUD treatment.⁷ The federal telemedicine waivers enacted due to COVID-19, including waiving the Ryan Haight Act requirement that clinicians conduct an in-person visit prior to prescribing OUD medications,⁸ helped to reduce barriers to and facilitate a rise in tele-OUD.⁹

Despite this shift, little is known about how clinicians are approaching tele-OUD or their opinions on its use. OUD clinicians’ experiences using telemedicine can inform clinical and policy guidance for tele-OUD treatment beyond the pandemic; for example, there is ongoing policy debate about the role of audio-only visits and whether these visits should be reimbursed after the pandemic.¹⁰ To understand clinician experiences and perspectives, we conducted an exploratory national survey of 602 OUD clinicians about their use of telemedicine and their opinions on using telemedicine. The specific goals were to assess frequency of and barriers to telemedicine use for OUD visits, frequency of audio-only OUD visits and reasons for conducting these visits, opinions on effectiveness of and comfort level with using tele-OUD, and perspectives on reimbursement level and ideal future role for tele-OUD. We also aimed to assess whether use of telemedicine and opinions about it differed by clinician characteristics and patient mix.

2. Methods

We surveyed a national sample of OUD clinicians between November 30^th and December 11^th, 2020. Surveyed clinicians included primary care physicians (PCPs), psychiatrists, nurse practitioners or certified nurse specialists, and physician assistants. Participants were recruited using WebMD/Medscape’s online panel of 2.5 million clinicians who have joined the platform to access clinical content, continuing medical education activities, and clinical tools.¹¹ Clinicians can also opt to participate in research, and the platform is commonly used for clinician recruitment in federally-funded studies.^12,13 For this survey, WebMD/Medscape emailed a study invitation (with the subject line: “Paid Research-Harvard University Funded study on Opioid Use Disorder”) to all psychiatrists on the panel, as well as a random selection of PCPs, nurse practitioners or certified nurse specialists, and physician assistants. Clinicians who expressed interest (n=3,816) were screened for eligibility via an online instrument and those eligible were invited to participate. Recruitment continued until the target sample size (n=600) was reached. The survey completion rate (the proportion of those who completed the survey among all respondents who passed screening criteria) was 97.3%. Clinicians who completed the survey received a gift card for between $20-$30 for their time. The survey was determined to be exempt by the Harvard Faculty of Medicine Institutional Review Board. Informed consent was obtained electronically.

Survey questions focused on clinician experiences using telemedicine (defined as synchronous visits that occur via audio-only or videoconferencing technology) to treat patients with OUD in the last month. The survey included questions about the clinician’s training, his or her primary practice setting and patient population, proportion of OUD visits conducted via in-person vs. audio-only vs. video, reasons for conducting audio-only visits, barriers to implementing video visits including licensing regulations, opinions on the effectiveness of telemedicine, comfort level with using video visits to treat patients with OUD, perspectives on the ideal role for telemedicine for OUD care post-pandemic, and the role of reimbursement in influencing a clinician’s expected future use (See Appendix for survey questions). The survey instrument was informed by qualitative interviews with waivered clinicians in the spring of 2020¹⁴ and cognitive testing with 5 clinicians who represented the target population.

2.1. Inclusion/Exclusion Criteria

Primary care physicians (PCPs), psychiatrists, nurse practitioners or certified nurse specialists, and physician assistants were eligible if they held a current buprenorphine waiver, spent at least eight hours per week in an outpatient setting, prescribed OUD medications to two or more patients in a typical month, and had at least one telemedicine visit with an OUD patient since the start of the pandemic. To focus on outpatient clinicians who provided in-person care prior to the pandemic, clinicians were excluded if they used telemedicine for 75% or more of OUD visits in 2019 or mainly practiced in an inpatient setting such as emergency department or residential treatment program. We excluded clinicians whose main practice setting was the military health system, Veterans Affairs (VA) or Indian Health Service (IHS) because the decision to use telemedicine in these settings is primarily made at the system-level vs. individual clinician-level and these clinicians are subject to a different regulatory and payment system related to telemedicine.

2.2. Data analyses

We used Stata (version 16) to analyze responses. Statistical significance was tested using chi-square and ANOVA tests. We hypothesized that there would be differences in responses among clinicians with higher caseloads of OUD patients and Medicaid and uninsured patients compared to clinicians with lower proportions of these patients. We divided clinicians into tertiles based on proportion of OUD patients and proportion of the sum of Medicaid and uninsured patients to investigate these differences.

3. Results

Respondents consisted of PCPs (N=241, 40.0%), psychiatrists (N=141, 23.4%) nurse practitioners or certified nurse specialists (N=139, 23.1%), and physician assistants (PAs) (N=81, 13.5%) (Table 1). Practice settings included single specialty group practice (N=236, 39.2%), multi-specialty group practice (N=205, 34.1%), solo practice (N=98, 16.3%), opioid treatment program (N=31, 5.2%), or other practice setting (N=32, 5.3%). Two-hundred and thirty clinicians (38.2%) practiced in a large city, while 57 (9.5%) practiced in a rural area. A majority (N=351, 58.3%) did not use any telemedicine to treat their OUD patients in 2019. Respondents reported that the mean percent of OUD patients in their caseload was 27.8%. The majority of clinicians (N=403, 66.9%) indicated that 25% or less of patients in their caseload were OUD patients (the distribution is shown in Appendix A1). Respondent’s main practice settings were located across 49 states.

Table 1.

Characteristics of Sample (N=602)

Training	N	%
Primary care physician	241	40.0%
Psychiatrist	141	23.4%
Nurse practitioner or certified nurse specialist	139	23.1%
Physician assistant	81	13.5%
Main practice setting
Single specialty group practice	236	39.2%
Multi-specialty group practice	205	34.1%
Solo practice	98	16.3%
Opioid treatment program (methadone program)	31	5.2%
Other	32	5.3%
Location of main practice setting
Large city	230	38.2%
Suburb near a large city	223	37.0%
Small city or town	92	15.3%
Rural area	57	9.5%
Years since completing medical training
Less than 5 years	100	16.6%
5–10 years	191	31.7%
11–15 years	95	15.8%
16 or more years	216	35.9%
Use of telemedicine in 2019 for patients with OUD
No telemedicine in 2019	351	58.3%
1–5% of visits	89	14.8%
6%–30% of visits	105	17.4%
31%–74% of visits	57	9.5%
Patient insurance types	Mean percent
Private/commercial insurance	34.4%
Medicaid	33.0%
Medicare	19.6%
No Insurance	9.0%
Other	4.1%
Mean percent of patients with OUD	27.8%
Mean percent of Black/African American patients	20.8%
Mean percent of Latino patients	17.9%
Mean percent of patients who prefer a language other than English	15.1%

3.1. Use of Telemedicine for OUD Visits

We asked participants to estimate the percent of their OUD visits in the last month that were delivered in-person vs. via audio-only vs. via video. On average, clinicians reported conducting 43.1% of OUD visits in-person and 56.9% via telemedicine (20.3% via audio-only and 36.6% via video) (Table 2). Overall use of telemedicine varied by clinician type and by proportion of OUD patients. Psychiatrists used telemedicine most frequently and PAs least frequently (mean percent of 73.8% and 44.2% of OUD visits, respectively) (P≤0.001). Clinicians in the highest tertile for proportion of OUD patients used telemedicine at higher rates, on average, than those in the lowest and middle tertiles (mean percent of 61.1% in highest tertile and 53.3% in the lowest (P=0.0497).

Table 2.

Mean Percent of OUD Visits Conducted In-Person vs. Via Audio-Only vs. Via Video, Overall Sample and by Subgroups

Clinician category	Mean % of OUD visits via in-person	P-Value	Mean % of OUD visits via audio-only	P-Value	Mean % of OUD visits via video	P-Value
Overall sample (602)	43.1%		20.3%		36.6%
Clinician type
Primary care physician (241)	49.5%	P≤0.001	19.9%	P=0.02	30.6%	P≤0.001
Psychiatrist (141)	26.2%		24.9%		48.9%
Nurse practitioner or certified nurse specialist (139)	41.7%		20.0%		38.3%
Physician assistant (81)	55.8%		14.1%		30.1%
Practice setting
Solo practice (98)	40.8%	P=0.38	16.3%	P=0.45	42.9%	P=0.03
Single specialty group practice (236)	41.4%		21.0%		37.5%
Multi-specialty group practice (205)	46.0%		21.7%		32.3%
Opioid treatment program (31)	49.0%		20.9%		30.2%
Other (32)	38.3%		18.1%		43.7%
Tertiles based on percent OUD patients
Tertile 1: 1–10% OUD patients (238 clinicians)	46.7%	P=0.0497	18.7%	P=0.36	34.6%	P=0.10
Tertile 2: 11–25% OUD patients (165 clinicians)	43.0%		22.3%		34.7%
Tertile 3: 30–100% OUD patients (199 clinicians)	38.9%		20.6%		40.5%
Tertiles based on percent Medicaid and uninsured patients
Tertile 1: 0–25% Medicaid/uninsured patients (227 clinicians)	44.8%	P=0.08	16.1%	P≤0.001	39.1%	P=0.32
Tertile 2: 27–55% Medicaid/uninsured patients (190 clinicians)	45.5%		19.1%		35.4%
Tertile 3: 60–100% Medicaid/uninsured patients (185 clinicians)	38.5%		26.7%		34.7%

When examining use of audio-only and video specifically, we observed differences by practice setting and by proportion of Medicaid/uninsured patients. Clinicians in solo practices and “other” practice types used video most frequently (mean percent of 42.9% and 43.7% of OUD visits, respectively), while those in opioid treatment programs used it the least (mean percent of 30.2%) (P=0.03). Clinicians with more Medicaid/uninsured patients used more audio-only visits than those with fewer Medicaid/uninsured patients (mean percent of 26.7% in highest tertile and 16.1% in the lowest (P≤0.001)).

3.2. Use of Audio-Only

As noted above, a mean percent of 20.3% (median=10%, IQR=0–30%) of OUD visits in the last month were conducted via audio-only. Most clinicians (N=431, 71.6%) delivered at least one audio-only visit in that period.

We asked clinicians who had delivered at least one audio-only visit to select their top three reasons for using audio-only visits from a list of options (Fig. 1). The most common were: 1) some patients do not have devices for video visits (e.g.., smartphone) (N =305, 70.8 %); 2) some patients do not know how to do video visits (N =237, 55.0 %); and 3) there are technical challenges with video visits (e.g., connection problems) (N =217, 50.3 %).

Figure 1. Reasons for Using Audio-Only Visits for Patients with OUD, Among Clinicians Who Had Any Audio-Only Visits in the Last Month (N=431).

Note: Clinicians were asked to select their three main reasons for using audio-only from these options.

When asked about the statement, “My patients receive higher quality care when they participate in video visits compared to audio only-visits,” the majority (N=258, 59.9%) of the 431 clinicians with at least one audio-only visit in the last month agreed or strongly agreed, while 114 (26.5%) disagreed or strongly disagreed.

3.3. Barriers to Implementing Video Visits

We asked clinicians to indicate their three main barriers to implementing video visits for patients with OUD in the last month from a list of options (Figure 2). The most common were: 1) lack of patient readiness (e.g., patients do not have devices, patients do not have digital literacy, patients do not have broadband) (N=480, 79.7%); 2) technology problems at the clinic or provider level (e.g., problems with hardware, software, and/or broadband) (N=322, 53.5%); and 3) lack of infrastructure (e.g., lack of HIPAA compliant technology, equipment) (N=178, 29.6%). Clinicians were also asked about the extent to which licensing regulations have prevented telemedicine use. Approximately one third (N=201, 33.4%) agreed or strongly agreed with the statement: “licensing regulations have prevented me from using telemedicine with some of my patients with opioid use disorder.”

Figure 2. Barriers to Implementing Video Visits for Patients with OUD (N=602).

Note: Clinicians were asked to select their three main barriers to using video from these options.

3.4. Opinions about using Telemedicine - Effectiveness and Comfort Level

Most respondents (N=376, 62.5%) (Table 3) agreed or strongly agreed with the statement, “I think the care that I have provided via telemedicine has been as effective as in-person care,” while 196 (32.6%) disagreed or strongly disagreed.

Table 3.

Clinician Opinions on Effectiveness of Telemedicine, Overall Sample and by Tertile

	“I think the care that I have provided via telemedicine has been as effective as in-person care”								P-Value
	Disagree		Agree		Don’t know		N/A
Overall Sample	196	32.6%	376	62.5%	28	4.7%	2	0.3%
Tertiles based on proportion of OUD patients
Tertile 1 (1–10% OUD patients, 238 clinicians)	85	35.7%	141	59.2%	11	4.6%	1	0.4%	P=0.12
Tertile 2 (11–25% OUD patients, 165 clinicians)	64	38.8%	93	56.4%	8	4.8%	0	0.0%
Tertile 3 (30–100% OUD patients, 199 clinicians)	47	23.6%	142	71.4%	9	4.5%	1	0.5%
Tertiles based on proportion of Medicaid and uninsured patients
Tertile 1 (0–25% Medicaid/uninsured patients, 227 clinicians)	77	33.9%	142	62.6%	7	3.1%	1	0.4%	P=0.65
Tertile 2 (27–55% Medicaid/uninsured patients, 190 clinicians)	64	33.7%	116	61.1%	10	5.3%	0	0.0%
Tertile 3 (60–100% Medicaid/uninsured patients, 185 clinicians)	55	29.7%	118	63.8%	11	5.9%	1	0.5%

Note: “Disagree” column combines “disagree” and “strongly disagree” responses; “Agree” column combines “agree” and “strongly agree” responses.

Clinicians were asked about their comfort level with using video visits for two types of OUD patients: established patients who are clinically stable in their recovery and established patients who are not clinically stable. The vast majority (N=535, 88.9%) reported being somewhat or very comfortable using video for clinically stable patients, while 297 (49.3%) (Table 4) reported being somewhat or very comfortable using video for clinically unstable patients. Comfort level with using video for patients who are not clinically stable varied according to proportion of OUD patients. Clinicians with more OUD patients were more likely to indicate that they were somewhat or very comfortable caring for clinically unstable patients via video (61.8% in highest, 46.7% in middle, and 40.8% in lowest tertile, P≤0.001).

Table 4.

Clinician Opinions on Comfort Level with Using Video Visits for Clinically Stable and Unstable Patients, Overall Sample and by Tertile

	“How comfortable were you with caring for established patients who are clinically stable via video visits?”						P-Value	“How comfortable were you with caring for established patients who are not clinically stable via video visits?”						P-Value
	Comfortable		Uncomfortable		N/A			Comfortable		Uncomfortable		N/A
Overall Sample	535	88.9%	49	8.1%	18	3.0%		297	49.3%	276	45.8%	29	4.8%
Tertiles based on proportion of OUD patients
Tertile 1 (1–10% OUD patients, 238 clinicians)	209	87.8%	20	8.4%	9	3.8%	P=0.52	97	40.8%	121	50.8%	20	8.4%	P≤0.001
Tertile 2 (11–25% OUD patients, 165 clinicians)	145	87.9%	16	9.7%	4	2.4%		77	46.7%	83	50.3%	5	3.0%
Tertile 3 (30–100% OUD patients, 199 clinicians)	181	91.0%	13	6.5%	5	2.5%		123	61.8%	72	36.2%	4	2.0%
Tertiles based on proportion of Medicaid and uninsured patients
Tertile 1 (0–25% Medicaid/uninsured patients, 227 clinicians)	207	91.2%	16	7.0%	4	1.8%	P=0.88	109	48.0%	111	48.9%	7	3.1%	P=0.16
Tertile 2 (27–55% Medicaid/uninsured patients, 190 clinicians)	164	86.3%	18	9.5%	8	4.2%		99	52.1%	79	41.6%	12	6.3%
Tertile 3 (60–100% Medicaid/uninsured patients, 185 clinicians)	164	88.6%	15	8.1%	6	3.2%		89	48.1%	86	46.5%	10	5.4%

Note: “Comfortable” column combines “somewhat comfortable” and “very comfortable” responses; “Uncomfortable” column combines “somewhat uncomfortable” and “very uncomfortable” responses.

3.5. Future Role of Telemedicine

Most respondents (N=422, 70.1%) indicated that they prefer to return to in-person care for the majority of their visits after the pandemic. However, those with more Medicaid/uninsured patients were less likely to report a desire to return to in-person care for the majority of visits (61.1% in highest tertile and 75.8% in lowest tertile, P=0.02).

We asked clinicians to select their ideal role for telemedicine for their OUD patients after the pandemic from a list of options (Table 5): telemedicine should not be offered at all (N=19, 3.2%); telemedicine should be offered, but only in special circumstances (e.g., for patients with mobility challenges, for patients who are traveling) (N=169, 28.1%); telemedicine should be offered frequently, but only to established patients (N=232, 38.5%), and telemedicine should be offered frequently to both new and established patients (N=173, 28.7%). There were differences in ideal role by tertile based on Medicaid/uninsured patients (P=0.01); for example, those with more Medicaid/uninsured patients were more likely to report that telemedicine should be offered frequently to both new and established patients.

Table 5.

Opinions on Ideal Role of Telemedicine for OUD Patients After the Pandemic, By Overall Sample and Tertiles based on Proportion OUD Patients and Proportion Medicaid/Uninsured Patients

Ideal role	Telemedicine should not be offered at all	Telemedicine should be offered, but only in special circumstances (e.g., for patients with mobility challenges, for patients who are traveling)	Telemedicine should be offered frequently, but only to established patients	Telemedicine should be offered frequently to both new and established patients	Other	P-value
Overall Sample	19 (3.2%)	169 (28.1%)	232 (38.5%)	173 (28.7%)	9 (1.5%)
Tertiles based on proportion of OUD patients						P=0.42
Tertile 1 (1–10% OUD patients, 238 clinicians)	7 (2.9%)	64 (26.9%)	99 (41.6%)	63 (26.5%)	5 (2.1%)
Tertile 2 (11–25% OUD patients, 165 clinicians)	8 (4.9%)	48 (29.1%)	65 (39.4%)	43 (26.1%)	1 (0.6%)
Tertile 3 (30–100% OUD patients, 199 clinicians)	4 (2.0%)	57 (28.6%)	68 (34.2%)	67 (33.7%)	3 (1.5%)
Tertiles based on proportion of Medicaid and uninsured patients						P=0.01
Tertile 1 (0–25% Medicaid/uninsured patients, 227 clinicians)	12 (5.3%)	65 (28.6%)	83 (36.6%)	63 (27.8%)	4 (1.8%)
Tertile 2 (27–55% Medicaid/uninsured patients, 190 clinicians)	5 (2.6%)	62 (32.6%)	78 (41.1%)	41 (21.6%)	4 (2.1%)
Tertile 3 (60–100% Medicaid/uninsured patients, 185 clinicians)	2 (1.1%)	42 (22.7%)	71 (38.4%)	69 (37.3%)	1 (0.5%)

We asked clinicians whether they would continue to offer telemedicine under two reimbursement scenarios: if reimbursement for telemedicine were the same as for in-person visits and if it were 25% lower than for in-person visits. Most clinicians (N=481, 79.9%) agreed or strongly agreed that they would continue to offer telemedicine if reimbursement were the same, while only 242 (40.2%) agreed or strongly agreed that they would continue to offer telemedicine if reimbursement were 25% lower than in-person visits.

4. Discussion

Clinicians in our survey used telemedicine for more than half of all OUD visits in late fall 2020, including a sizeable share (20%) via audio-only. Most felt that the telemedicine care they have provided has been as effective as in-person care and were comfortable using video with clinically stable patients; however, just under half were comfortable using video with clinically unstable patients.

More than two-thirds of clinicians personally prefer to return to in-person care for the majority of visits post-pandemic. However, when asked about the ideal role of telemedicine for their patients with OUD and presented with options for different patient types, a similar percentage (67%) thought telemedicine should be offered frequently after the pandemic, with about 40% indicating it should be offered frequently to established patients and approximately 30% indicating it should be offered frequently to both new and established patients. An additional 28% thought telemedicine should be offered, but only in special circumstances such as for patients with mobility challenges or who are traveling. Together these survey results indicate that telemedicine will play an important role in OUD treatment post-pandemic and will have an expanded role as one of several options (in-person, video, audio-only) for clinicians, but will likely represent a relative minority of visits and use may vary substantially across clinicians.

The clinician’s patient population will likely be an important driver of both how much telemedicine is used and modality (video vs. audio-only). Clinicians who care for more OUD patients were more likely to use telemedicine overall and were also more likely to be comfortable treating clinically unstable patients via video. These differences may be driven by greater familiarity with treating OUD and therefore greater comfort with treating unstable patients despite the limitations of the modality. Further, OUD, similar to other behavioral health conditions, may be better suited for telemedicine compared to other conditions that require more frequent physical exams, and clinicians treating more OUD patients may invest in using more telemedicine. In addition, research has shown that telemedicine may increase retention in OUD treatment, its effectiveness is similar to that of face-to-face visits,^15,16 and it is increasingly seen as a low-barrier method to accessing OUD care.¹⁷ Clinicians with more OUD patients may recognize these advantages. Guidelines that outline best practices as well as training programs may be important for increasing comfort with telemedicine use for clinicians who are less experienced or have fewer OUD patients.

Clinicians with more Medicaid/uninsured patients were more likely to use audio-only visits and to prefer to continue telemedicine use post-pandemic. This may be driven by the fact that these clinicians recognize certain barriers their patients face, such as transportation barriers to in-person visits and obstacles to using video visits. Overall, clinicians in our study used audio-only visits because their patients did not have the necessary technology to use video visits. There is ongoing policy debate about the role of audio-only visits after the pandemic and whether they should be reimbursed.¹⁸ There are concerns that audio-only visits could result in fraud and increased spending,^10,18–20 and could contribute to the development of a two-tiered system where higher-income patients receive care via video visits and disadvantaged patients receive audio-only visits.²¹ This could exacerbate disparities of care given that the majority of surveyed clinicians who provided audio-only visits felt audio-only visits were lower quality than video visits. However, others view audio-only as key to maintaining access, particularly for disadvantaged populations who may not have broadband internet or smartphones.^18,20 For example, a recent study using data from 41 Federally Qualified Health Centers (FQHCs) in California found that visit volume declined modestly for primary care visits in March-August 2020 but remained stable for behavioral health visits because telemedicine visits (particularly by telephone) replaced in-person visits.²⁰ The study found that telephone visits were highest in April 2020, making up 65.4% of all primary care visits and 71.6% of behavioral health visits.²⁰ In July 2021, as part of the annual Physician Fee Schedule (PFS) proposed rule, CMS proposed expanding access to audio-only care by allowing payment to eligible practitioners when they provide certain behavioral health services to patients via audio-only telephone calls from their homes.²² Given clinicians’ perceptions of patient barriers to video visits and the increased use of audio-only among clinicians with more Medicaid/uninsured patients, our results highlight that, if reimbursed, audio visits will play a key role in OUD care. Our results also emphasize the need for supportive grants and practices that could improve accessibility of video care to patients in low-income and rural communities.

There is also a debate about telemedicine reimbursement after the pandemic. Some have pushed for reimbursement to remain the same as for in-person care,²³ while others have suggested that payment parity may increase spending and that telemedicine reimbursement should be lower than for in-person care to reflect the potentially lower costs of providing it.¹⁹ The policy debate has largely focused on what it costs to deliver telemedicine and less on the effect that reduced reimbursement will have on provider behavior and, thus, patient access to telemedicine. Our results suggest that the majority of OUD clinicians would continue offering telemedicine to at least some patients if reimbursement were the same as in-person care, but substantially fewer would continue to offer it if reimbursement were 25% lower than in-person care despite the fact that many surveyed clinicians view it as an effective substitute for in-person care and over 95% thought telemedicine should be offered in some form after the pandemic. These positive opinions are consistent with other COVID-19 era studies of behavioral health clinicians; for example, a study of mental health care providers found that respondents were positive about telepsychiatry and most would like to continue using it for at least 25% of their caseload after the pandemic.²⁴

There are several limitations of this study. First, the survey was administered to a non-probability-based sample, so results may not be representative of the population of waivered clinicians. However, our sample of respondents was generally similar to prior national surveys of waivered clinicians on available characteristics, including clinician type, years since training, and rural/urban location²⁵ (see Appendix). This survey focused on clinicians who can prescribe OUD medications; thus, we are not able to examine views and practices among other types of OUD clinicians who do not prescribe. Results were based on survey responses in November/December 2020 and may not be representative of views throughout the pandemic. Finally, we did not control for testing of multiple outcomes due to the exploratory nature of this study.

5. Conclusion

While many clinicians indicated a desire to return to in-person care for the majority of their visits after the pandemic, nearly all indicated that telemedicine should continue to be offered to their patients in some form. Those who saw more Medicaid/uninsured patients were more likely to think that telemedicine should be offered frequently. There was substantial variation across clinicians in their comfort level with using telemedicine for less stable patients, with greater comfort levels among clinicians with more experience in OUD treatment. Two key areas of uncertainty that will drive future telemedicine use for OUD treatment are the role of audio-only visits, particularly for individuals from underserved populations, and the level of reimbursement.

Highlights:

• Surveyed clinicians used telemedicine for over half of visits for patients with opioid use disorder in the last month at the time of the survey (fall 2020), including a sizeable share via audio-only.

• Most clinicians thought telemedicine has been as effective as in-person care, and clinicians were more comfortable treating clinically stable patients via video than clinically unstable patients.

• After the pandemic, most clinicians prefer to return to in-person care for the majority of their visits but nearly all would like the option to offer telemedicine to their patients in some form.