Available via license: CC BY 4.0
Content may be subject to copyright.
Peavyetal.
Addiction Science & Clinical Practice (2024) 19:8
https://doi.org/10.1186/s13722-024-00435-z
RESEARCH
Improving thequality ofcounseling
andclinical supervision inopioid treatment
programs: howcan technology help?
K. Michelle Peavy1, Angela Klipsch2, Christina S. Soma2* , Brian Pace2, Zac E. Imel2,3, Michael J. Tanana2,
Sean Soth4, Esther Ricardo‑Bulis4 and David C. Atkins2
Abstract
Background The opioid epidemic has resulted in expanded substance use treatment services and strained
the clinical workforce serving people with opioid use disorder. Focusing on evidence‑based counseling practices
like motivational interviewing may be of interest to counselors and their supervisors, but time‑intensive adherence
tasks like recording and feedback are aspirational in busy community‑based opioid treatment programs. The need
to improve and systematize clinical training and supervision might be addressed by the growing field of machine
learning and natural language‑based technology, which can promote counseling skill via self‑ and supervisor‑moni‑
toring of counseling session recordings.
Methods Counselors in an opioid treatment program were provided with an opportunity to use an artificial intel‑
ligence based, HIPAA compliant recording and supervision platform (Lyssn.io) to record counseling sessions. We
then conducted four focus groups—two with counselors and two with supervisors—to understand the integration
of technology with practice and supervision. Questions centered on the acceptability of the clinical supervision soft‑
ware and its potential in an OTP setting; we conducted a thematic coding of the responses.
Results The clinical supervision software was experienced by counselors and clinical supervisors as beneficial
to counselor training, professional development, and clinical supervision. Focus group participants reported
that the clinical supervision software could help counselors learn and improve motivational interviewing skills.
Counselors said that using the technology highlights the value of counseling encounters (versus paperwork). Clini‑
cal supervisors noted that the clinical supervision software could help meet national clinical supervision guidelines
and local requirements. Counselors and clinical supervisors alike talked about some of the potential challenges
of requiring session recording.
Conclusions Implementing evidence‑based counseling practices can help the population served in OTPs; another
benefit of focusing on clinical skills is to emphasize and hold up counselors’ roles as worthy. Machine learning tech‑
nology can have a positive impact on clinical practices among counselors and clinical supervisors in opioid treatment
programs, settings whose clinical workforce continues to be challenged by the opioid epidemic. Using technology
to focus on clinical skill building may enhance counselors’ and clinical supervisors’ overall experiences in their places
of work.
Open Access
© The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco
mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Addiction Science &
Clinical Practice
*Correspondence:
Christina S. Soma
tina@lyssn.io
Full list of author information is available at the end of the article
Page 2 of 11
Peavyetal. Addiction Science & Clinical Practice (2024) 19:8
e opioid epidemic has expanded substance use disor-
der (SUD) treatment services and forced rapid change in
already stretched systems of care [1, 2]. In this climate,
labor intensive implementation projects may be depri-
oritized, an unfortunate outcome if the implementation
target is an evidence-based (EB) counseling practice that
will ultimately help both patients and clinicians. Record-
ing and reviewing counseling sessions is one notable
barrier to implementation of EB counseling practices in
community-based substance use disorder treatment [3],
even though feedback and coaching are becoming the
norm in Motivational Interviewing training research tri-
als [4]. Concerns over the cost, time, and effort it takes to
systematize procedures for recording counseling sessions
are valid [5]. However, direct observation of clinical prac-
tice is not only imperative to arrive at EB fidelity [6, 7],
it also fits the recommendation of SAMHSA’s Consensus
Panel assembled to develop clinical supervision guide-
lines for SUD counselors [8].
Adequate clinical supervision is critical for EB imple-
mentation; it has also been demonstrated as a protec-
tive factor against burnout and prevention of turnover
in SUD treatment settings [9–11]. Maintaining a healthy
SUD workforce and the organizations in which they work
is increasingly important as treatment providers hasten
to accommodate a changing environment. From scal-
ing up telehealth efforts in the context of COVID [12] to
adapting treatment to an ever-evolving landscape of sub-
stances being used by SUD patients (e.g., methampheta-
mine and synthetic opioids; [13]), today’s SUD counselors
face increasing pressures while they treat a more affected
and abundant population than any other time in history.
Such challenges highlight the need to support SUD coun-
seling staff, both with hands-on supervision, as well as
with tools that allow them to practice and feel confident
about their clinical skills.
While technological innovation has been integrated
into SUD treatment for the past several years, these inno-
vations have not yet directly impacted clinical super-
vision and clinical skill building, including fidelity to
evidence-based counseling practices. us far, technol-
ogy efforts in the SUD treatment space have been limited
to patient self-monitoring [14]; technology-based assess-
ment, interventions and aftercare [15–17]; and mobile
apps facilitating video to directly observe buprenorphine
dosing for opioid use disorder [18]. Aside from web-
based training for SUD counselors [19, 20], technology as
a tool in SUD treatment has not yet been applied to the
SUD workforce.
Recently, advances in machine learning and speech sig-
nal processing have been integrated into psychotherapy
science, facilitating the development of technology to
automatically evaluate use of specific evidence-based
practices in recordings of substance use counseling and
psychotherapy generally [21]. ese machine-learning
based approaches to fidelity monitoring, are typically
trained by human labeled data (i.e., transcripts that have
been coded with a gold standard fidelity measure like the
Motivational Interviewing Skills Code [27]) and can be
competitive with human to human reliability of the same
sessions (see [22]). More recently, work has focused on
initial testing of technologically based tools for fidelity
monitoring with clinicians [23, 24]. However, there has
been no formal study of barriers or facilitators to imple-
menting technology supported supervision in real world
SUD specific clinical milieus. Because clinical supervi-
sion can serve as a protective factor against burnout and
turnover, it is important to examine innovative meas-
ures that promote quality and easy access to clinical
supervision.
e purpose of this report is to first describe the uptake
of a novel technology designed to increase SUD counse-
lor skill in motivational interviewing and improve clinical
supervision. Second, we aim to summarize results from
focus groups targeting counselors and clinical supervi-
sors in a large opioid treatment program, who discussed
proposed implementation of a clinical supervision plat-
form. Lyssn (or Lyssn.io) is a web-based platform that
supports evidence-based supervision with machine
learning based evaluation of motivational interviewing
(MI; see [25]) skills in counseling sessions. is clinical
supervision platform was designed with the intention
of providing mental health providers session recording
organization, immediate machine learning-based infor-
mation about content, MI metrics, and a platform to
facilitate supervision practices such as asynchronous dis-
cussion about the content of their sessions. e platform
was developed with the use of machine learning to cir-
cumvent the use of laborious human coding techniques
to provide session feedback (see [23]).
e current study examined implementation of a
cloud-based recording and feedback platform in an
Opioid Treatment Program (OTP), representative of
treatment programs in the Pacific Northwest, as well as
acceptability of this clinical supervision platform among
counselors and clinical supervisors. We conducted two
sets of focus groups (four total)—two with counselors
and two with supervisors—from an opioid treatment
Keywords Opioid treatment, Addiction counseling, Motivational interviewing, Machine learning
Page 3 of 11
Peavyetal. Addiction Science & Clinical Practice (2024) 19:8
program in Washington state. One set of counselor and
supervisor focus groups were asked questions about cur-
rent training and supervision practices. e second set
were asked questions surrounding how this particular
clinical supervision platform would help with clinical
practice.
Method
Study site
e study site was a non-profit opioid treatment program
(OTP) serving individuals with opioid use disorder across
three geographically diverse locations in Washington
State. Recruitment for the current study took place at the
largest of the three sites, which serves Seattle’s urban core
and contains upwards of 1400 patients. Many of these
OTP patients (up to 60% in certain zip codes) are home-
less, struggle with polysubstance use, as well as several
psychosocial instabilities. As is consistent with SAMHSA
guidelines for OTPs [26], counseling is mandatory at the
study OTP; the treatment setup is such that counselors
serve as the primary contact for OTP patients. Histori-
cally, the study OTP did not integrate direct observa-
tion (i.e., recording of counseling sessions) into clinical
training or supervision. Unrelated to the study, the
OTP was concurrently overhauling supervision proto-
cols to include direct observation, and these changes are
described in more detail in the focus group results.
e gold standard of treatment for opioid use disorder
(OUD) is medication treatment (MOUD), involving one
of the evidence-based medications targeting the disor-
der [46, 47]. However, there are several reasons to imple-
ment additional EBPs such as motivational interviewing
in the context of MOUD [48, 49]. Adjunctive treatments
may help increase adherence to treatments of infec-
tious diseases that are common among MOUD patients
(e.g., HIV, AIDS, hepatitis C), increase self-efficacy and
engagement with MOUD, and assess for treatment readi-
ness [50]. e site where the study took place provided
MOUD via both methadone and buprenorphine to treat
OUD. In addition, MI was utilized to aid with patient
retention, MOUD treatment adherence, and targeting
engagement with other health promoting behaviors (e.g.,
attending doctor’s appointments).
Participants
Study site counselors (n = 27) and clinical supervisors
(n = 2) were approached for inclusion in the current
study at a counseling staff meeting, which served as the
initial recruitment effort. New hires that occurred after
initial recruitment were emailed about the opportunity
to participate. While counselor participants represented
a variety of educational and professional backgrounds,
all participants had appropriate credentialing to practice
SUD counseling in Washington State, entitled Substance
Use Disorder Professional (SUDP) or a SUDP-Trainee
designation for individuals on the pathway to licensure.
Before this project, the majority of the counselors had no
prior experience of recording and reviewing their clinical
work.
Eleven participants and two clinical supervisors were
recruited into the study and agreed to use the clinical
supervision platform as a platform for recording coun-
seling sessions with patients who provided consent for
recording. Counselors and clinical supervisors then had
the opportunity to participate in focus groups (germane
to the current study), which were incentivized with $50
gift cards. Recordings of counselor and clinical supervi-
sor focus groups were de-identified during the research
process.
Clinical supervision platform
Prior to the focus groups, all participants were introduced
to, and had the opportunity to use, the clinical supervi-
sion platform which provided recordings of counseling
sessions, and machine learning based transcription,
global MI metrics, and prevalent conversation topics.
e focus group that was asked hypothetical questions
about how the software could be used were additionally
shown a report of the comprehensive machine-gener-
ated MI metrics (Fig.3). Counselors and clinical super-
visors accessed the clinical supervision platform via a
web-browser, where users can easily record new coun-
seling sessions or review previously recorded sessions
(see Fig. 1). e session review interface supports two
kinds of annotations: comments about the session as a
whole in a simple chat box, as well as time-linked com-
ments directly in the video playback (see Fig.2). e lat-
ter allows clinicians and their supervisors to immediately
queue up a portion of the session to review.
e cloud-based software included both a speech
processing pipeline and a machine learning engine. e
speech processing pipeline used state-of-the-art speech
recognition algorithms, specifically designed for and
trained on behavioral health conversations, to both gen-
erate a transcription, as well using speech features as
inputs to predict the full suite of MI fidelity codes based
on the Motivational Interviewing Skills Code system
(MISC; [27]; see [23, 23, 41, 42, 44, 45]). In addition to
continuing to improve the machine learning models from
thousands of hours of human coded sessions, research
has also been conducted to understand provider experi-
ences in using the tool and making user-design informed
changes (see [24]).
During the second focus group, concentrated primar-
ily on hypothetical questions about how the platform
could be used, counselors and clinical supervisors were
Page 4 of 11
Peavyetal. Addiction Science & Clinical Practice (2024) 19:8
introduced to an interactive, web-based report (Fig.3),
an additional feature. e report includes MI specific
information via the traditional six MI fidelity statistics:
empathy, MI spirit, reflection-to-question ratio, per-
cent complex reflections, percent open questions, and
percent MI adherent. In addition, there is a timeline of
the entire session, where each talk-turn is linked to the
automated speech recognition transcript of the session
to facilitate review and study of specific exchanges
within the session. Each talk-turn includes predicted
MISC codes based on the machine learning engine and
a visual representation of vocally encoded arousal of
the speaker. In the focus groups, counselors and super-
visors discussed how this feature could be integrated
into clinical supervision and ultimately improve deliv-
ery of counseling services.
Fig. 1 The Clincial Supervision Platform User Interface
Fig. 2 The Clinical Supervision Recorded Session Interface
Page 5 of 11
Peavyetal. Addiction Science & Clinical Practice (2024) 19:8
Measures
To understand the acceptability of the clinical supervi-
sion software among counselors and clinical supervisors,
four focus groups were conducted; two with counselors
and two with clinical supervisors. Counselors and clinical
supervisors were separated for each of the focus groups
to encourage both groups to speak more freely about
their experiences. e same questions were used for the
counselor and clinical supervisor focus groups. Focus
group #1 was used to: (1) gather general information
about the current state of counselor training, clinical
supervision practices, and monitoring; (2) briefly intro-
duce staff to the clinical supervision software; and (3)
discuss how the technology could be used in the current
setting, and what the barriers would be. e objectives
of Focus group #2 were to: (1) demonstrate the feedback
report; and (2) discuss acceptability of the clinical super-
vision software including the feedback report, as well as
motivators and barriers to implementation. Focus groups
were conducted by members of the research team; the
four focus groups were recorded and transcribed for
analysis. In addition to focus groups, the number of ses-
sions recorded through the clinical supervision software
was also collected as a measure of engagement with the
platform (see Table1).
Qualitative analysis
Participant responses during the focus groups were
recorded and transcribed, and thematic analysis was
employed to analyze transcripts. Focus groups with
counselors included questions regarding how to use
the computer-generated report, processes to review the
report, using the report in supervision, and additional
metrics that individuals would have liked to see in the
report. For the clinical supervisor, there was an addi-
tional question of how counselors would react to using
the software. For the analysis, two authors (KMP; BP)
utilized a thematic analysis to identify common themes
Fig. 3 The Clinical Supervision MI Report User Interface
Page 6 of 11
Peavyetal. Addiction Science & Clinical Practice (2024) 19:8
that emerged within the focus groups. e researchers
then met and discussed emergent themes to reach a con-
sensus, resulting in the conceptual categories illustrated
by participants’ comments. Researchers met regularly to
resolve disagreements on conflicting theme coding, dis-
cuss their rationale for codes, and share their personal
biases that may have influenced the coding process.
Results
Table 1 summarizes information about 11 counselors
who agreed to participate in the study, how many ses-
sions they recorded. A total of 426 sessions (M = 38.7;
SD = 41.7; range = 0–125) were recorded over a 9-month
period. Counselors conduct approximately 80 sessions
per month, so there were approximately 7,920 number of
sessions possible. Patients need to consent before being
recorded, which may have limited the overall number
of possible sessions. We did not collect the number of
patients who consented to record.
Focus group results
Theme 1: The clinical supervision software asamechanism
forfocusing oncounseling andclinical supervision
asopposed todocumentation
A perceived benefit of the clinical supervision software
was that it is a mechanism for focusing on counseling
skills, as opposed to administrative and documentation
requirements. One counselor said: “I would like to know
what I could do differently to elicit a different response, to
show more empathy. To really learn that. With chemical
dependency there’s a lot of case management all of that.
I’d like to actually have a counseling session.” Another
counselor indicated: “I know that ‘counselor’ is attached
to our names, but it’s like we’re paper pushers. And we do
a lot of paperwork, and we have to do a lot of paperwork
by a certain amount of time and a certain set of paper-
work. And that’s our focus. It’s not really like eliciting this
change talk within the patient. You know that’s kind of
gone way out here for me. I want to know how I could do
it differently.”
Theme 2: The clinical supervision software asanaid
tosupervision
Counselors were largely very positive about their experi-
ence in clinical supervision; they described their Clini-
cal Supervisor as supportive and empowering. “[Clinical
Supervisor] always makes time for you. When I leave
[Clinical Supervisor’s] office I feel like a superhero. Like
I could take on the world. When you leave the office to
go downstairs, it’s important to have 110%.” e process
for obtaining Clinical Supervision was described as both
informal (e.g., “I just knock on [Clinical Supervisor’s]
door”), as well as formal, scheduled individual and group
supervision. Supervision reportedly consists of clinical
consultation along with discussion of processes/policies
and documentation review. Clinical Supervisors indicate
that the current and primary mechanism for determining
counselor performance depends heavily on documenta-
tion review and has little to no progress reporting. “If
there’s a tool that’s going to make tracking and helping…
because now I’m going to have to be doing…it’s almost
like progress notes with on the people I supervise which
I’ve not had to do in the past unless it was like a correc-
tive action kind of plan. Uh so that’s going to be new and
if this kind of technology is going to help with that I’m
definitely all for it.”
Both counselors and clinical supervisors talked about
the value of ongoing formal training (i.e., didactic; Con-
tinuing Education), specifically citing motivational inter-
viewing as a primary interest. “Training is so important…
it’s so empowering to the whole idea of counseling. You
need it. It would be really nice if I felt like I had the time
to leave [for training]. is clinic is just a little busy.”
Counselors and clinical supervisors cite large case-
loads and increasing responsibilities as a barrier to not
only training attendance, but also as a barrier for tar-
geted skill development. Clinical supervisors noted that
“one-off” trainings were important, but such a format
had limitations in terms of producing substantive skill
improvement across the counseling staff. One supervisor
remarked that “…you get four, 5days or a week’s worth of
that training, but then there needs to be follow-up. How
do you implement it?”.
Clinical Supervisors voiced excitement about how the
clinical supervision software will help implement the
new supervision protocol. To maintain consistency with
SAMHSA’s Clinical Supervision Treatment Improvement
Table 1 Number of Recorded Sessions and Employment
Turnover of OTP Counselors
Participant Number of sessions
recorded Left
employment
(Y/N)
1 5 Y
2 8 N
3 23 Y
4 43 Y
5 9 N
6 91 N
7 71 N
8 0 Y
9 2 N
10 49 Y
11 125 Y
Page 7 of 11
Peavyetal. Addiction Science & Clinical Practice (2024) 19:8
Protocol TIP 52 [8], supervisors reported that the clini-
cal supervision software’s recording platform could help
them meet this objective more efficiently than sitting in
sessions with counselors. “… [Observation] is probably
part of the new managed care, value-based care, kind of
uh model that we’re all uh getting prepared to deal with.”
ey reported that the clinical supervision software feed-
back forms could help supervisors formulate their own
evaluative commentary more quickly, noting that simul-
taneously observing counseling and writing feedback
is “time consuming.” Another piece of the new supervi-
sion protocol that the clinical supervision software could
help with is the supervision documentation that will be
an expectation with the new protocol. One supervisor
reported: “It’s almost like [we need to keep] progress
notes for supervision—that’s new. If this kind of technol-
ogy would help with that, I’m all for it.”
Supervisors noted that the clinical supervision soft-
ware’s recording platform and feedback form could help
shape the content of supervision sessions themselves:
“I’m a little bit nervous about…when it comes time to
doing direct observation…kind of knowing okay, what
should I be focusing in on? So I can give feedback that’s
going to be useful.” Supervisors see how the clinical
supervision software feedback report could help struc-
ture counselor feedback and help them tailor supervision
based on skills being displayed, or not displayed, accord-
ing to the report.
Supervisors reported that session recordings could
ensure good clinical care. “[Recording sessions] is how
you know you’re providing the good service…You look at
the note, you know, and that tells you they can document.
Doesn’t tell you what kind of counselor they are per se.”
Supervisors noted the feedback is easier to give to coun-
selors who are performing well, and more difficult to give
then counselors who are not performing well. ey make
the point that the clinical supervision software’s feedback
form could pull out objective information, making the
process easier to give feedback to individuals who are not
performing well.
Theme 3: Session transcription could improve counselor
workow
Counselors were excited by the transcription function of
the clinical supervision software, seeing this feature as
both a way to help them chart more efficiently, and as a
means for connecting more genuinely with patients. One
counselor noted: “When you’re doing counseling you’re
either sitting at a computer and typing or you’re tak-
ing handwritten notes, or you commit what they say to
memory. I’d rather not be doing any three of those things
when I’m working with a patient. So ultimately I would
use the transcript to help complete my notes.”
Theme 4: Feedback asatool toimprove counselor MI skills
Counselors expressed an interest in how the clinical
supervision software’s feedback could help them improve
their clinical skills. Said one counselor: “Just having a
reflective practice about how I am doing as a counselor.
My whole generation of friends, we’re all gamers right?
So having a score to compare myself to. It’s something to
challenge myself to. To improve on.” Counselors also con-
sidered how the immediacy of the feedback forms could
transfer to skill uptake. For example: “I don’t want to wait
three times a year for a supervisor to come in and tell me
how I’m doing on my OARS skills, I want to know more
frequently than that. is is why I signed up for this pro-
ject in the first place.”
Theme 5: Concerns aboutrecording
Counselors made comments throughout the focus groups
about work capacity, citing high caseloads and limited
time to take on new tasks. is issue was top of mind for
counseling staff, and one concern with the clinical super-
vision software’s implementation was that it would be
adding to an existing full plate of responsibilities. ey
noted adding in a process to their routines can be diffi-
cult to integrate and remember (one counselor noted that
they had a hard time remembering to start recording).
Another person remarked that adding in another consent
for patients may seem overwhelming to them, given that
patients have an existing set of consents they sign at the
outset of treatment.
Counselors expressed concerns about how record-
ing might affect clinical interactions and patient expe-
riences. Specifically, there was a mixture of responses
about counselors’ experiences in getting patient consent
to record: “My population is not very trusting of systems.
ey‘re like ‘Wait wait, you want to record this for what?’
And it takes a lot of conversation and some of them are
still just resistant.” Another counselor also expressed ini-
tial concerns about the patients’ willingness to record
but observed patients’ overall desire to be helpful to the
counselor, their growth, and in turn helpful to fellow
patients: “For the patients, for the most part, when I talk
to them [about recording counseling] they seem to be
open. ‘If it helps somebody behind me I will do it’. at
sense of goodwill. I wasn’t expecting that.” Consensus
was that consent to record depended on both the patient
variables (e.g., whether patients present with paranoid
ideation), as well as how the opportunity is presented
(i.e., emphasis of recording is on counselor, and as a way
to help counselor skills, as opposed to focus on patient
content).
Counselor’s experiences with recording varied based
on prior recording experiences. One counselor who
had recorded sessions states: “I really hesitated to get
Page 8 of 11
Peavyetal. Addiction Science & Clinical Practice (2024) 19:8
[involved with] this project because I wasn’t too sure…
at what part…how genuine the patient would be in the
session. Because you have the recording right there. And
how genuine is the patient? But as it progressed I didn’t
think that was an issue. And that was…it was a really big
relief. If anything else it put me in a place it made it a lot
easier for me to focus. Into the session.” Finally, coun-
selors noted a concern about security, both in terms of
outsiders accessing recordings, as well as the possibility
of recordings being subpoenaed by law enforcement or
Child Welfare organizations.
Like counselors, supervisors also predicted time con-
straints to be a barrier to the clinical supervision software
implementation. One supervisor put it: “e thought of
adding anything new, even if it’s positive, can be a bar-
rier.” Supervisors indicated that direct observation via
recorded counseling sessions would be a culture shift
among counseling staff, and they wondered whether
counselors would be mistrustful of this level of oversight.
“Maybe there would be fear that I would judge them, or
that they would get in trouble.” ey noted that openness
to this level of supervision would likely vary based on
counselor variables; one Clinical Supervisor speculated
that newer counselors may be more open to it. “ey are
learning and absorbing.”
Discussion
e current study tested the acceptability of the clini-
cal supervision software, a recording platform for coun-
seling sessions that provides transcripts, as well as an
AI-backed report that provides counselors and supervi-
sors real time feedback about counselor performance and
MI skills. At the study site, a large number of counselors
opted to try out the new technology, albeit a number had
left the OTP during the clinical supervision software trial
period, indicative of high turnover found in OTP sites
[28]. Overall, after reviewing two sets of focus groups
(four sessions total), we found that the clinical supervi-
sion software was experienced by counselors and clinical
supervisors as beneficial to counselor training, profes-
sional development, clinical supervision, and importantly
to the provision of counseling to OTP patients.
Our data indicate that 11 counselors at a large OTP vol-
untarily tried the clinical supervision software to improve
their skills; these counselors recorded a total of 426 coun-
seling sessions over 9months demonstrating uptake of
a new technology. An organization needs early adopters
to help bring their peers along, and the counselors who
participated in the current study serve as a proxy for
“champions”, a key piece to successful implementation of
innovations in SUD treatment settings [29].
Supervisors and counselors alike expressed enthusiasm
about the clinical supervision software and its potential
impact on clinical supervision, albeit for slightly differ-
ent reasons. Counselors reported that they value clini-
cal supervision, describing it as supportive and vital to
their work. is sentiment is common amongst SUD
counselors [30], but supervision practices vary greatly
[31], one study indicating that up to a third of counse-
lors did not even receive any clinical supervision [32].
Other researchers have shown that supervisors in addic-
tion treatment settings tend to report more time pro-
vided to supervision, more interactions and feedback
than do their matched counselor counterparts [33]. A
clinical supervision software tool could have the effect of
prompting consistently scheduled supervision that has a
purpose and structure.
Study site supervisors noted that the clinical supervi-
sion software could aid with the increasing supervision
demands from state and national agencies, as well as
the added supervision documentation included in these
new requirements. Specifically, supervisors reported that
their organization was already moving towards requiring
a more codified and systematic clinical supervision pro-
tocol. Interviews with SUD treatment providers about
value-based care indicate concerns about staff not being
adequately trained in evidence-based practices, and that
the training burden would be challenging [34].
While counselors were not as closely attuned with
forthcoming supervision protocols and quality assur-
ance issues, they did express interest in getting immedi-
ate and more frequent feedback about their performance.
Feedback from counselors has since informed the devel-
opment of an automatically generated session summary
that is now provided by the clinical supervision software
as another way to aid with the clinical documentation
process. ough participants noted concern regarding
security and the potential for recordings to be subpoe-
naed, they posited that the use of a recording platform to
deepen their understanding of session content was highly
valuable. Future implementation efforts should continue
to focus on ensuring the privacy, security, and parame-
ters in which recordings may become part of a court pro-
ceeding, with the ultimate goal of protecting providers
and their patients.
Supervisors remarked that the clinical supervision
software’s machine generated feedback reports would
assist in focusing on skills and allow more accurate and
objective feedback about clinical skills. Supervisors
reported that the organization currently depends on
documentation review to gauge performance. Record-
ing and feedback reports would take some of the guess-
work out of performance evaluation. Counselors may
also appreciate more objectivity in measuring perfor-
mance, which could ultimately lower turnover. Clinical
supervisors also noted that one-time trainings, though
Page 9 of 11
Peavyetal. Addiction Science & Clinical Practice (2024) 19:8
important, lacked the necessary follow up that research
has demonstrated to be necessary for skill retention
[35]. If routinely used in clinical supervision, the soft-
ware may be perceived by counselors that feedback is
more objective, obviating quality of care to both coun-
selor and clinical supervisor, taking favoritism (or
perceptions of such) out of the running as a driver of
performance evaluation results.
Counselors and supervisors saw the clinical supervi-
sion software as a tool to help counselors improve their
Motivational Interviewing skills by regularly listening
to counseling sessions, reviewing the feedback form,
and using these tools to reflect on their own work.
Counselors also remarked that the clinical supervision
software commands a real focus on clinical skills, an
emphasis that is a change of pace given that they feel
their jobs are consumed with “paper pushing” activi-
ties. In other words, counselors spoke to a possible
conflict between their values (e.g., counseling; rap-
port building) and the values of the larger organiza-
tion and regulating bodies (e.g., documentation and
other requirements that detract from patient/counselor
interactions). ese types of value conflicts have been
shown to play a role in burnout and turnover [36]. At
any job there is a balance between the work employ-
ees must do and what they want to do. When asked in
a qualitative study, rural SUD counselors noted that
increased access to professional education and oppor-
tunities would enhance recruitment and retention in
the workplace [37]. An AI-based supervision platform
could aid counselors by auto summarizing chart notes
and keep counseling skill development top of mind, the
balance between “have tos” and “want tos” is weighted
in a way that counselors’ work can remain connected to
the reasons many of them were drawn to the field in the
first place: because they wanted to help.
Like many large and complicated OTPs, counselors and
supervisors stated that they had a great number of exist-
ing demands, making the implementation of anything
new more challenging. Supervisors also wondered about
the acceptance of the clinical supervision software more
broadly, understandably considering counselors who did
not volunteer to participate. ey acknowledged that
with a heterogeneous staff it makes sense that some will
be open to it and others not. Reluctance to record work
samples or have work directly observed is commonplace
among SUD providers [32]. Changing cultural organiza-
tion to promote consistent recording may indeed appear
as a sizable obstacle; however, direct observation is not
only what is recommended in SAMHSA TIP 54, but also
shown to be more accurate and comprehensive in terms
of understanding what counselors are doing with their
patients [38].
Limitations
ough presenting powerful results that represent
some of the realistic challenges for counselors and clini-
cal supervisors, there are some limitations to the study.
Counselors who volunteered for participation in the
study may have been more eager and enthusiastic to par-
ticipate, and therefore results may be difficult to general-
ize. e counselors that volunteered for the study likely
represent those that tend to show more openness to new
learning opportunities and perhaps more motivated to
improve their work.
In a post-hoc collection of counselor attrition data, of
the 11 counselors recruited into the study, more than
50% (n = 6) left employment during the pilot period. Due
to turnover in counseling staff, there was heterogeneity
in counselor experience and exposure to the technology.
Some focus group participants had substantial experi-
ence with the clinical supervision software, others had
not used it at the time of the focus group. SUD counse-
lors continue to face new and increasing issues that lead
to burnout and high turnover [39]. As with any stretched
system, SUD treatment organizations may at best prior-
itize immediate crises and other pressing tasks over EB
counseling implementation, at worst skimp on clinically
vital activities such as supervision. e COVID-19 pan-
demic continues to contribute to significant workflow
disruptions, implicating long standing changes to OUD
treatment [40], and ushering in widespread reliance on
technology (i.e., telehealth; [12]). New systems and tel-
ehealth practices may provide opportunities for patients
and providers to remain connected, however, these sys-
tems continue to add some burdens to counselor work-
flows and prohibit some counselors from consistently
using new technology.
Finally, the sample of participants who volunteered to
participate in focus groups was small, and pulled from
one treatment setting, and thus also difficult to generalize
to the broader network of OTP counselors and clinical
supervisors. Further research should be conducted with
OTP clinics to help generalize the benefits and continued
needs of these mental health providers.
Conclusion
Technology can play a positive role in supporting the
implementation of evidence-based counseling at sites
like an OTP, where workloads are stretched due to
the ongoing opioid epidemic. Counselors and clini-
cal supervisors interviewed in our focus groups were
enthusiastic about the clinical supervision platform’s
utility in improving motivational interviewing skills
and enhancing clinical supervision. To support the
SUD workforce, we need to find innovative ways to help
Page 10 of 11
Peavyetal. Addiction Science & Clinical Practice (2024) 19:8
clinicians feel connected to their work and confident
in the clinical supervision they receive. Treatment set-
tings and researchers should continue to consider how
technology can improve the services provided to peo-
ple with SUDs.
Acknowledgements
We would like to thank the counselors and clinical supervisors of Evergreen
Treatment Services for their participation in our focus groups. Their ongoing
commitment to people with opioid use disorder is inspiring. We would also
like to thank Grin Lord for facilitating the focus groups.
Author contributions
KMP helped conceive and design the focus groups, assisted in coordinating
focus group activities, supervised the data analysis, and wrote the manuscript
in collaboration with all other authors. AK is the technical implementation
officer at Lyssn.io. Ms. K facilitated usage of the platform, provided technical
support, supported the design of focus group questions and organization,
and organized and managed the focus group data. CSS is a postdoctoral
fellow at Lyssn.io. Dr. S assisted in finalizing the organization and content of
the manuscript, formatting, and the submission process. BP is the director
of clinical artificial intelligence at Lyssn.io. Dr. P contributed to the thematic
coding process and contributed to editing. ZEI is the chief science officer and
co‑founder of Lyssn.io. Dr. I contributed to initiating the grant writing process,
collaborating with our opioid treatment services partners, and compiling the
literature review. MJT is the chief technology officer and co‑founder of Lyssn.
io. Dr. Tanana was the primary contributor to the development of the machine
learning algorithms for Lyssn Supervision Platform and MI feedback platforms;
he contributed focus group questions and manuscript writing. SS oversees
clinical services at the study site. Mr. S provided consultation and supervision
on the project by helping coordinate study site staff and resources. ER‑B was
the Research Coordinator for the project; she collected data and coordinated
all study activities. DA is the CEO and co‑founder of Lyssn.io. Dr. A contributed
to initiating the grant writing process and leading the collaboration process
with our opioid treatment services partners.
Funding
This project was funded by the National Institute on Drug Abuse (NIDA), Small
Business Innovation Research (SBIR) grant (R44DA046243‑S1).
Availability of data and materials
Data sharing is not applicable to this article as no datasets were generated or
analyzed during the current study.
Declarations
Ethics approval and consent to participate
The study was approved through the Advarra Institutional Review Board
(CR00256234).
Consent for publication
Not applicable.
Competing interests
Conflict of Interest Disclosure: Zac E. Imel, David C. Atkins, and Michael J.
Tanana are co‑founders of a technology company, Lyssn.io (http:// www. lyssn.
io/), which is focused on technology to support training, supervision, and
quality assurance of behavioral health care.
Author details
1 PRISM, Department of Community and Behavioral Health, Elson S. Floyd
College of Medicine, Washington State University, Spokane, WA, USA. 2 Lyssn.
Io, Seattle, Washington, USA. 3 University of Utah, Salt Lake City, UT, USA. 4 Ever‑
green Treatment Services, Seattle, Washington, USA.
Received: 19 December 2022 Accepted: 5 January 2024
References
1. McBain RK, Dick A, Sorbero M, Stein BD. Growth and distribution of
buprenorphine‑waivered providers in the United States, 2007–2017. Ann
Intern Med. 2020;172(7):504–6.
2. Mojtabai R, Mauro C, Wall MM, Barry CL, Olfson M. Medication treatment
for opioid use disorders in substance use treatment facilities. Health Aff.
2019;38(1):14–23.
3. Doumas DM, Miller RM, Esp S. Continuing education in motivational
interviewing for addiction counselors: Reducing the research‑to‑practice
gap. J Addict Offender Couns. 2019;40(1):36–51.
4. Madson MB, Villarosa‑Hurlocker MC, Schumacher JA, Williams DC,
Gauthier JM. Motivational interviewing training of substance use treat‑
ment professionals: a systematic review. Subst Abuse. 2018. https:// doi.
org/ 10. 1080/ 08897 077. 2018. 14753 19.
5. Martino S, Paris M Jr, Añez L, Nich C, Canning‑Ball M, Hunkele K, Olmstead
TA, Carroll KM. The effectiveness and cost of clinical supervision for moti‑
vational interviewing: a randomized controlled trial. J Subst Abuse Treat.
2016;1(68):11–23.
6. Proctor E, Silmere H, Raghavan R, Hovmand P, Aarons G, Bunger A, Griffey
R, Hensley M. Outcomes for implementation research: conceptual distinc‑
tions, measurement challenges, and research agenda. Adm Policy Mental
Health Mental Health Serv Res. 2011;38(2):65–76.
7. Bearman SK, Schneiderman RL, Zoloth E. Building an evidence base for
effective supervision practices: An analogue experiment of supervision to
increase EBT fidelity. Adm Policy Mental Health Mental Health Serv Resh.
2017;44(2):293–307.
8. Center for Substance Abuse Treatment. Competencies for substance
abuse treatment clinical supervision (Technical Assistance Publication
(TAP) Series 21–1, DHHS Publication No. (SMA) 07–4243). Rockville: Sub‑
stance Abuse and Mental Health Services Administration; 2007.
9. Beitel M, Oberleitner L, Muthulingam D, Oberleitner D, Madden LM,
Marcus R, Eller A, Bono MH, Barry DT. Experiences of burnout among drug
counselors in a large opioid treatment program: a qualitative investiga‑
tion. Subst Abuse. 2018;39(2):211–7.
10. Knudsen HK, Ducharme LJ, Roman PM. Clinical supervision, emotional
exhaustion, and turnover intention: a study of substance abuse treat‑
ment counselors in the clinical trials network of the National Institute on
Drug Abuse. J Subst Abuse Treat. 2008;35(4):387–95.
11. Oser CB, Biebel EP, Pullen E, Harp KL. Causes, consequences, and preven‑
tion of burnout among substance abuse treatment counselors: a rural
versus urban comparison. J Psychoact Drugs. 2013;45(1):17–27.
12. Uscher‑Pines L, Sousa J, Raja P, Mehrotra A, Barnett M, Huskamp HA. Treat‑
ment of opioid use disorder during COVID‑19: experiences of clinicians
transitioning to telemedicine. J Subst Abuse Treat. 2020;1(118):108124.
13. Mattson CL, Tanz LJ, Quinn K, Kariisa M, Patel P, Davis NL. Trends and geo‑
graphic patterns in drug and synthetic opioid overdose deaths—United
States, 2013–2019. Morb Mortal Wkly Rep. 2021;70(6):202.
14. Serre F, Fatseas M, Swendsen J, Auriacombe M. Ecological momentary
assessment in the investigation of craving and substance use in daily life:
a systematic review. Drug Alcohol Depend. 2015;1(148):1–20.
15. Marsch LA, Dallery J. Advances in the psychosocial treatment of addic‑
tion: the role of technology in the delivery of evidence‑based psychoso‑
cial treatment. Psychiatric Clinics. 2012;35(2):481–93.
16. McPherson SM, Burduli E, Smith CL, Herron J, Oluwoye O, Hirchak K, Orr
MF, McDonell MG, Roll JM. A review of contingency management for the
treatment of substance‑use disorders: adaptation for underserved popu‑
lations, use of experimental technologies, and personalized optimization
strategies. Subst Abuse Rehabilit. 2018;9:43.
17. Moore BA, Fazzino T, Garnet B, Cutter CJ, Barry DT. Computer‑based inter‑
ventions for drug use disorders: a systematic review. J Subst Abuse Treat.
2011;40(3):215–23.
18. Godersky ME, Klein JW, Merrill JO, Blalock KL, Saxon AJ, Samet JH, Tsui
JI. Acceptability and feasibility of a mobile health application for video
directly observed therapy of buprenorphine for opioid use disorders in
an office‑based setting. J Addict Med. 2020;14(4):319.
19. Weingardt KR, Villafranca SW, Levin C. Technology‑based training in cog‑
nitive behavioral therapy for substance abuse counselors. Subst Abuse.
2006;27(3):19–25.
20. Weingardt KR, Cucciare MA, Bellotti C, Lai WP. A randomized trial compar‑
ing two models of web‑based training in cognitive–behavioral therapy
for substance abuse counselors. J Subst Abuse Treat. 2009;37(3):219–27.
Page 11 of 11
Peavyetal. Addiction Science & Clinical Practice (2024) 19:8
21. Aaes‑van Doorn K, Kamsteeg C, Bate J, Aaes M. A scoping review
of machine learning in psychotherapy research. Psychother Res.
2021;31(1):92–116.
22. Cao J, Tanana M, Imel ZE, Poitras E, Atkins DC, Srikumar V. Observing
dialogue in therapy: categorizing and forecasting behavioral codes. arXiv
preprint arXiv: 1907. 00326. 2019.
23. Imel ZE, Pace BT, Soma CS, Tanana M, Hirsch T, Gibson J, Georgiou P,
Narayanan S, Atkins DC. Design feasibility of an automated, machine‑
learning based feedback system for motivational interviewing. Psycho‑
therapy. 2019;56(2):318.
24. Hirsch T, Soma C, Merced K, Kuo P, Dembe A, Caperton DD, Atkins DC,
Imel ZE. “ It’s hard to argue with a computer" Investigating Psychothera‑
pists’ Attitudes towards Automated Evaluation. In Proceedings of the
2018 Designing Interactive Systems Conference. 2018. pp. 559–571.
25. Miller WR, Rollnick S. Motivational interviewing: Helping people change.
New York: Guilford press; 2012.
26. Abuse S, Administration MH. Federal guidelines for opioid treatment
programs. HHS publication no (SMA) PEP15‑FEDGUIDEOTP. 2015.
27. Miller WR, Moyers TB, Ernst D, Amrhein P. Manual for the motivational
interviewing skill code (MISC). Unpublished manuscript. Albuquerque:
Center on Alcoholism, Substance Abuse and Addictions, University of
New Mexico. 2003.
28. Hatch‑Maillette MA, Harwick R, Baer JS, Masters T, Cloud K, Peavy M, Wiest
K, Wright L, Beadnell B, Wells EA. Counselor turnover in substance use
disorder treatment research: observations from one multisite trial. Subst
Abuse. 2019;40(2):214–20.
29. Wood K, Giannopoulos V, Louie E, Baillie A, Uribe G, Lee KS, Haber
PS, Morley KC. The role of clinical champions in facilitating the use of
evidence‑based practice in drug and alcohol and mental health settings:
a systematic review. Implement Res Pract. 2020;1:2633489520959072.
30. Schmidt EA. Clinical supervision in the substance abuse profession: a
review of the literature. Alcohol Treat Q. 2012;30(4):487–504.
31. Watkins CE Jr. What do clinical supervision research reviews tell us?
Surveying the last 25 years. Couns Psychother Res. 2020;20(2):190–208.
32. Culbreth JR. Clinical supervision of substance abuse counselors: current
and preferred practices. J Addict Offender Couns. 1999;20(1):15–25.
33. Laschober TC, de Tormes Eby LT, Sauer JB. Clinical supervisor and counse‑
lor perceptions of clinical supervision in addiction treatment. J Addict Dis.
2012;31(4):382–8.
34. O’Grady MA, Lincourt P, Gilmer E, Kwan M, Burke C, Lisio C, Neighbors CJ.
How are substance use disorder treatment programs adjusting to value‑
based payment? A statewide qualitative study. Subst Abuse Res Treat.
2020;14:1178221820924026.
35. Schwalbe CS, Oh HY, Zweben A. Sustaining motivational interviewing: a
meta‑analysis of training studies. Addiction. 2014;109(8):1287–94.
36. Maslach C, Leiter MP. Early predictors of job burnout and engagement. J
Appl Psychol. 2008;93(3):498.
37. Murphy J. Improving the recruitment and retention of counselors
in rural substance use disorder treatment programs. J Drug Issues.
2022;52(3):434–56.
38. Martino S, Gallon S, Ball SA, Carroll KM. A step forward in teaching addic‑
tion counselors how to supervise motivational interviewing using a
clinical trials training approach. J Teach Addict. 2008;6(2):39–67.
39. Volker R, Bernhard B, Anna K, Fabrizio S, Robin R, Jessica P, Rudolf S, Lucia
D, Jürgon R, Franz H, Christine S. Burnout, coping and job satisfaction in
service staff treating opioid addicts—from Athens to Zurich Stress and
Health. J Int Soc Investig Stress. 2010;26(2):149–59.
40. Peavy KM, Darnton J, Grekin P, Russo M, Green CJ, Merrill JO, Fotinos C,
Woolworth S, Soth S, Tsui JI. Rapid implementation of service delivery
changes to mitigate COVID‑19 and maintain access to methadone
among persons with and at high‑risk for HIV in an opioid treatment
program. AIDS Behav. 2020;24(9):2469–72.
41. Imel ZE, Steyvers M, Atkins DC. Computational psychotherapy research:
Scaling up the evaluation of patient–provider interactions. Psychother‑
apy. 2015;52(1):19.
42. Xiao B, Imel ZE, Georgiou PG, Atkins DC, Narayanan SS. “ Rate my thera‑
pist”: automated detection of empathy in drug and alcohol counseling
via speech and language processing. PLoS ONE. 2015;10(12):e0143055.
43. Tanana M, Hallgren K, Imel Z, Atkins D, Smyth P, Srikumar V. Recursive neu‑
ral networks for coding therapist and patient behavior in motivational
interviewing. In: Proceedings of the 2nd workshop on computational
linguistics and clinical psychology: from linguistic signal to clinical reality.
2015. pp. 71–79.
44. Can D, Marín RA, Georgiou PG, Imel ZE, Atkins DC, Narayanan SS. “It
sounds like…”: A natural language processing approach to detecting
counselor reflections in motivational interviewing. J Couns Psychol.
2016;63(3):343.
45. Gaut G, Steyvers M, Imel ZE, Atkins DC, Smyth P. Content coding of psy‑
chotherapy transcripts using labeled topic models. IEEE J Biomed Health
Inform. 2015;21(2):476–87.
46. Mattick RP, Breen C, Kimber J, Davoli M. Buprenorphine maintenance
versus placebo or methadone maintenance for opioid dependence.
Cochrane Database Syst rev. 2014. https:// doi. org/ 10. 1002/ 14651 858.
CD002 207. pub4.
47. National Academies of Sciences, Engineering, and Medicine. Medications
for opioid use disorder save lives. Washington: National Academies Press;
2019.
48. Oesterle TS, Thusius NJ, Rummans TA, Gold MS. Medication‑assisted treat‑
ment for opioid‑use disorder. In Mayo Clinic Proceedings. Elsevier, 2019
Vol. 94, No. 10, pp. 2072–2086.
49. Sayegh CS, Huey SJ Jr, Zara EJ, Jhaveri K. Follow‑up treatment effects of
contingency management and motivational interviewing on substance
use: a meta‑analysis. Psychol Addict Behav. 2017;31(4):403.
50. Mumba MN, Findlay LJ, Snow DE. Treatment options for opioid use disor‑
ders: a review of the relevant literature. J Addict Nurs. 2018;29(3):221–5.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in pub‑
lished maps and institutional affiliations.
