Mental Health Counseling From Conversational Content With Transformer-Based Machine Learning.

Importance: Use of asynchronous text-based counseling is rapidly growing as an easy-to-access approach to behavioral health care. Similar to in-person treatment, it is challenging to reliably assess as measures of process and content do not scale. Objective: To use machine learning to evaluate clinical content and client-reported outcomes in a large sample of text-based counseling episodes of care. Design, Setting, and Participants: In this quality improvement study, participants received text-based counseling between 2014 and 2019; data analysis was conducted from September 22, 2022, to November 28, 2023. The deidentified content of messages was retained as a part of ongoing quality assurance. Treatment was asynchronous text-based counseling via an online and mobile therapy app (Talkspace). Therapists were licensed to provide mental health treatment and were either independent contractors or employees of the product company. Participants were self-referred via online sign-up and received services via their insurance or self-pay and were assigned a diagnosis from their health care professional. Exposure: All clients received counseling services from a licensed mental health clinician. Main Outcomes and Measures: The primary outcomes were client engagement in counseling (number of weeks), treatment satisfaction, and changes in client symptoms, measured via the 8-item version of Patient Health Questionnaire (PHQ-8). A previously trained, transformer-based, deep learning model automatically categorized messages into types of therapist interventions and summaries of clinical content. Results: The total sample included 166 644 clients treated by 4973 therapists (20 600 274 messages). Participating clients were predominantly female (75.23%), aged 26 to 35 years (55.4%), single (37.88%), earned a bachelor's degree (59.13%), and were White (61.8%). There was substantial variability in intervention use and treatment content across therapists. A series of mixed-effects regressions indicated that collectively, interventions and clinical content were associated with key outcomes: engagement (multiple R = 0.43), satisfaction (multiple R = 0.46), and change in PHQ-8 score (multiple R = 0.13). Conclusions and Relevance: This quality improvement study found associations between therapist interventions, clinical content, and client-reported outcomes. Consistent with traditional forms of counseling, higher amounts of supportive counseling were associated with improved outcomes. These findings suggest that machine learning-based evaluations of content may increase the scale and specificity of psychotherapy research.

Effectiveness and feasibility of a motivational interviewing intake (MII) intervention for increasing client engagement in outpatient addiction treatment: an effectiveness-implementation hybrid design protocol.

BACKGROUND: Client discontinuation from outpatient addiction treatment programs is common, and the initial intake is the service delivery point with the highest attrition rate. Replacing the comprehensive intake assessment with a person-centered Motivational Interviewing (MI) intervention is a potential solution to address provider and client concerns about the disengaging, time-intensive nature of the typical initial intake. It remains unclear whether the use of an alternative to the standard intake at the initial visit can fit within typical organizational reporting requirements, whether it decreases attrition, and whether implementation of person-centered intake procedures within outpatient addiction treatment programs is feasible, acceptable, and can be sustained. PURPOSE: To describe the methods and design of an effectiveness-implementation hybrid type 1 trial of a Motivational Interviewing at Intake (MII) intervention using the Consolidated Framework for Implementation Research (CFIR). METHODS: The study will determine the effectiveness of two intake conditions: (1) standard comprehensive intake assessment (intake-as-usual [IAU]), and (2) MII consisting of a person-centered discussion between provider and client about the client's desire and intent to enter treatment. Although both interventions are focused on understanding client presenting complaints and needs for treatment, the delivery differs as the IAU uses a semi-structured assessment guide, while MII applies the theory of MI to have a conversation about treatment engagement. Adults seeking outpatient addiction treatment services will be randomly assigned to the MII condition (n = 75) or the IAU condition (n = 75). Primary outcomes will be client engagement (i.e., treatment entry, attendance, and completion) obtained from the electronic medical record. Secondary outcomes (client motivation and therapeutic alliance) will be putative mechanisms of client engagement assessed immediately before and after the intake. The trial also will explore determinants of effective, sustainable implementation using assessments of organizational readiness and capacity to change, as well as interviews on MII implementation feasibility. CONCLUSION: This trial of an MII intervention will investigate the feasibility of a motivational intervention as an initial contact with substance use treatment-seeking clients as well as indicators of intervention effectiveness within the systems where it is employed. Trial registration Clinicaltrials.gov identifier: NCT05489068.

What can clients tell us about whether to use motivational interviewing? An analysis of early-session ambivalent language.

BACKGROUND: Although motivational interviewing (MI) is an effective method for promoting change in problematic alcohol and other drug use, it does not benefit all clients. Clinicians have little empirical guidance on who is likely to benefit from MI and who is not. We hypothesized that differences in clients' spontaneously offered language early in the session would predict their responsiveness to MI during the remainder of the session. METHOD: The study obtained coding data from 125 counseling sessions from a large randomized controlled trial of clinician training. A cluster analysis created one group of clients whose language reflected ambivalence, and one group whose language reflected readiness to change. We conducted a univariate analysis of variance to compare the mean change in percent change talk across the session between groups. RESULTS: Clients whose language reflected ambivalence early in the session had a greater change in their percent change talk during the remainder of the session, compared to those whose language reflected greater readiness to change (F (1,90) = 63.02, t = 7.94, p < .001). Surprisingly, the group whose language reflected readiness had a decrease in their percent change talk during the remainder of the session (M = -10.9%, SD = 16.3%). Adjusting the results for regression to the mean effects did not eliminate these differences. CONCLUSION: Clients' language early in the session may offer clinicians some guidance on whether MI is likely to be useful or counterproductive in the treatment of substance use disorder.

Should Substance Use Counselors Choose a Direction For Their Clients? Motivational Interviewing Trainers May Be Ambivalent.

Motivational Interviewing (MI) is comprised of a client-centered relationship and a clear intention on the part of the practitioner to influence behavior change. This study explores MI trainers' decisions about their use of directionality in MI as they instruct others in the method. 111 MI trainers were asked to select content they would include in a hypothetical MI training. Almost half of trainers chose to teach trainees to "always maintain an attitude of equipoise", a strategy that is contradicted by MI theory and empirical data. This finding suggests a theoretical rift within the MI community with implications for substance use counseling.

Can a pure motivational interviewing intervention be manualized and still efficacious? A test of feasibility and initial efficacy.

The importance of attending to the therapeutic process despite the challenges in manualizing it is demonstrated in the empirical evolution of motivational interviewing (MI). Whereas manuals exist for adaptations of MI, no manual has been developed and tested for MI in its pure form (pure MI). This study evaluated the feasibility and initial efficacy of a pure MI intervention manual - MI for risky social drinking (MI-RSD) - designed to target risky social drinking behaviors in college students with social anxiety. A pilot sample of 42 college students completed measures of alcohol use and mental health symptoms and the MI-RSD intervention. We developed a manual for the 2-session MI-RSD intervention, trained 4 clinical doctoral students, and used observer-, therapist- and participant-completed measures to evaluate fidelity. Therapists met beginner proficiency in MI fidelity and participant gave high ratings of therapist adherence to MI and working alliance, demonstrating intervention feasibility. Also, participants reported significant reductions in hazardous drinking and evaluation fears, but not in social interaction anxiety. We offer preliminary evidence that pure MI can be manualized and effective. Specifically, MI-RSD represents an alternative to MI adaptations in mitigating alcohol-related harm for young adults. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Mechanisms of Motivational Interviewing: a Conceptual Framework to Guide Practice and Research.

Reviews of the motivational interviewing (MI) training literature demonstrate MI is a nuanced skill set that takes carefully planned didactic training, application of skills in context-specific practice settings, and ongoing support to promote reflective practice and sustained proficiency. Despite the robust knowledge base related to training and how MI works to achieve favorable outcomes, these two literature bases are not well integrated. In an effort to inform and guide future research, we propose the mechanisms of motivational interviewing (MMI) conceptual framework, which expands upon previous work. Specifically, the framework adds training as an ongoing process consistent with Bennett-Levy's (Behav Cogn Psychother 34:57-78, 2006) model of skill development and acquisition to the existing two-path framework that helps us to understand how MI works to achieve its desired effects (Magill et al., J Consult Clin Psychol 82:973-983, 2014). Herein, we describe measures used to evaluate the mechanisms within the four MMI framework links: initial training to competency, competency to proficiency, proficiency to talk about change, and talk about change to behavior change. Next, we synthesize the literature associated with each of the mechanisms of the MMI. We conclude by discussing implications for practice and research. This framework offers a more complete path structure to understand the mechanisms of change associated with MI that could improve our understanding of inconsistent effect sizes observed across prior trials evaluating MI effectiveness.

Is fidelity to motivational interviewing associated with alcohol outcomes in treatment-seeking 60+ year-old citizens?

BACKGROUND: Part of the variability in treatment outcomes for Motivational Interviewing (MI) may be explained by differences in the fidelity to MI. The Motivational Interviewing Treatment Integrity manual version 4 (MITI 4) is an improved measure of fidelity to elements of MI. It is not known whether the fidelity to MI, as measured by the MITI 4, is related to treatment outcome. OBJECTIVES: To examine whether fidelity to MI is associated with alcohol use outcomes - predictive validity of the MITI 4. METHOD: Twenty percent of the recorded sessions at the Danish sites of the Elderly Study were randomly drawn and coded for fidelity to MI with the MITI 4. The Elderly Study was an international, randomized controlled trial, in which people 60 years or older with Alcohol Use Disorders received either four weeks of Motivational Enhancement Therapy (MET) or four weeks of MET combined with up to eight additional sessions of the Community Reinforcement Approach- Senior (MET+CRA-S). Elements of MI and summary scores of the MITI 4 were used as predictors in a mixed effects regression analysis. Treatment outcomes were use of alcohol and consequences of drinking at 26-weeks follow-up. RESULTS: In total, 423 sessions representing 238 participants were randomly drawn and coded for fidelity to MI. Mean values of the treatment elements indicated high fidelity to MI, with higher fidelity to MI in the MET sessions, as compared to CRA-S sessions. None of the predictors in the multilevel model analyses were associated with outcome at follow-up. Exploratory analysis indicated reverse associations between one measure of MI-fidelity and drinking outcomes in the combined treatment (CRAS). CONCLUSION: The fidelity of the MI intervention, received by participants in this study, did not predict better treatment outcomes. MI may be less effective in populations which are already committed to change behavior. As expected and validating for the MITI 4, fidelity to MI-elements was lower in the combination of MI with other treatment approaches. Additionally, the timing of MI in these combined settings might be important for effectiveness.

With odds of a single session, motivational interviewing is a good bet.

First sessions of psychotherapy present a rare and potent opportunity for therapists and clients alike. Motivational interviewing is established as a stand-alone method for promoting behavior change as well as a useful prelude to other therapies. This article provides a rationale and empirical support for the use of motivational interviewing in first psychotherapy sessions. A case illustration and practical guidelines for therapists are included. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

Examining the influence of active ingredients of motivational interviewing on client change talk.

Motivational Interviewing (MI) is an evidenced-based intervention designed to help clients explore and resolve ambivalence around substance use. MI combines a humanistic tradition with behavioral components to facilitate client decisions concerning behavior change. As such, there is marked interest in the relationship between the two active ingredients of MI - the relational, or person-centered, components and the technical, or directional, behavioral components - on client in-session language. Yet, few studies have examined how these active ingredients operate in concert. Therefore, the current study evaluated the constellation of relational skills associated with client language, as well as the influence of technical skills on the relationship between provider relational skills and client change language. Specifically, we tested a latent construct of relational skill and its direct association with the proportion of client change talk. We then explored the mediating role of reflections of change and sustain talk (RefCT and RefST) on this relationship. The data for this secondary analysis are from Project ELICIT (N = 131), a randomized control trial evaluating the effects of MI training on client change language. We found support for a latent construct of relational skill (i.e., empathy, acceptance, collaboration, and autonomy/support). However, the relational skill construct did not predict client change language. There was support for an indirect effect, such that relational skills predicted RefCT and RefST, and RefCT and RefST predicted client change language. These results suggest that the synergistic implementation of the relational and technical components of MI is critical to facilitating a higher percentage of change talk.

Lessons learned from measuring fidelity with the Motivational Interviewing Treatment Integrity code (MITI 4).

BACKGROUND AND AIMS: The Motivational Interviewing Treatment Integrity code (MITI) measures fidelity to, and the quality of, Motivational Interviewing (MI), and can also be used when MI is combined with other treatment methods. The current study presents a fidelity measurement with the MITI 4.2.1, in both Motivational Enhancement Therapy sessions and the combined Community Reinforcement Approach-Senior (CRAS). METHOD: The MITI 4.2.1 was used to evaluate treatment sessions provided in the Elderly Study, a multi-national randomized trial evaluating treatment for alcohol use disorders in the elderly. Following expert recommendations, training was conducted at two international sites as well as at the Danish site. Twenty percent of the sessions at the Danish study site were rated. Twelve percent were multiply rated by all raters. Interrater reliability was assessed by the Intraclass Correlations Coefficient (ICC). RESULTS: Mean ICC of the 52 sessions rated by all raters was 0.78 (95% CI: 0.70; 0.86). The rare measures confront and emphasize autonomy, and the global measure softening sustain talk only reached fair levels of ICC, while the remaining measures were good or excellent. In the sessions of MI combined with other treatment approaches in the CRAS, the MITI 4.2.1 has a similar reliability as in MET sessions only, except for the measure persuade with permission. CONCLUSION: The MITI 4.2.1 is a reliable instrument for measuring fidelity to Motivational Interviewing elements, also in the context of Community Reinforcement Approach Senior. However, in softening sustain talk, the rare measures, and persuade with permission it has proved more difficult to reach high levels of interrater reliability.

Short- and Long-Term Effects of Within-Session Client Speech on Drinking Outcomes in the COMBINE Study.

OBJECTIVE: There is evidence of a causal chain in motivational interviewing (MI) involving counselor MI-consistent skills, client change language, and outcomes. MI was a key component of the combined behavioral intervention in the Combined Pharmacotherapies and Behavioral Interventions for Alcohol Dependence (COMBINE) Study. Participants in COMBINE were treatment-seeking and medication-seeking, and were required to maintain a period of abstinence before enrollment. Counselors were closely monitored and were temporarily removed from the study for remediation and supervision if their performance fell below threshold. The purpose of this study was to evaluate the MI causal chain among initial-session combined behavioral intervention audio recordings in this unique sample of highly motivated clients and highly monitored counselors. METHOD: Session 1 audio recordings (N = 254, 73.5% men) were evaluated using the Motivational Interviewing Skill Code Version 2.0 with good interrater reliability. Counselor speech was measured using a summary measure of MI-consistent behavior. Client change language was assessed, using change talk and sustain talk. The outcome measure of drinks per week was computed using the COMBINE data set. RESULTS: Higher sustain talk was significantly associated with increased drinking during treatment, whereas higher change talk was significantly associated with decreased drinking at the 1-year follow-up. In addition, there were significant indirect effects linking counselor behavior, client speech, and drinking both during treatment and at 1 year. CONCLUSIONS: Results supported the posited causal chain for MI. Despite somewhat lower variability of counselor behavior and use of a coding instrument that did not capture directional counselor behaviors, counselor behavior, client speech, and drinking outcomes were clearly linked in this unique sample.

Motivational interviewing and the clinical science of Carl Rogers.

The clinical method of motivational interviewing (MI) evolved from the person-centered approach of Carl Rogers, maintaining his pioneering commitment to the scientific study of therapeutic processes and outcomes. The development of MI pertains to all 3 of the 125th anniversary themes explored in this special issue. Applications of MI have spread far beyond clinical psychology into fields including health care, rehabilitation, public health, social work, dentistry, corrections, coaching, and education, directly impacting the lives of many people. The public relevance and impact of clinical psychology are illustrated in the similarity of MI processes and outcomes across such diverse fields and the inseparability of human services from the person who provides them, in that both relational and technical elements of MI predict client outcomes. Within the history of clinical psychology MI is a clear product of clinical science, arising from the seminal work of Carl Rogers whose own research grounded clinical practice in empirical science. As with Rogers' work 70 years ago, MI began as an inductive empirical approach, observing clinical practice to develop and test hypotheses about what actually promotes change. Research on MI bridges the current divide between evidence-based practice and the well-established importance of therapeutic relationship. Research on training and learning of MI further questions the current model of continuing professional education through self-study and workshops as a way of improving practice behavior and client outcomes. (PsycINFO Database Record

A randomized controlled trial to influence client language in substance use disorder treatment.

BACKGROUND: Client language is hypothesized to be a mechanism of action in motivational interviewing (MI). Despite the association of change and sustain talk with substance treatment outcomes, it not known whether providers can intentionally influence this language as hypothesized. OBJECTIVE: This is a randomized controlled trial to investigate whether substance use providers can be trained to influence client language. METHODS: Treatment providers specializing in substance use disorders (n=190) were randomly assigned to standard training in MI (MI-AU) or training emphasizing an influence of client language (MI-LEAF). Treatment sessions with actual clients were evaluated 3, 6 and 12 months after training by masked raters. Frequencies of client change and sustain talk were the outcome variables. RESULTS: Sustain talk, but not change talk, was significantly lower in clients whose providers had received the specialized training (b=-0.175, SE=0.087, p=0.046, CI[-0.348 to 0.002], d=-0.325). Mediation analyses supported a causal chain between a) training, b) providers' attempts to minimize sustain talk in treatment sessions via directive reflective listening and c) client sustain talk in the treatment session (κ2=0.0833, bootstrap SE=0.0394, 95% CI [0.0148, 0.1691]). CONCLUSIONS: With specialized training, providers can reduce the amount of opposition language their clients offer when considering a change in their substance use. Demonstrating that client language is under partial control of the provider supports the feasibility of clinical trials to investigate the impact of shaping client language on treatment outcomes.

The Motivational Interviewing Treatment Integrity Code (MITI 4): Rationale, Preliminary Reliability and Validity.

UNLABELLED: The Motivational Interviewing Treatment Integrity code has been revised to address new evidence-based elements of motivational interviewing (MI). This new version (MITI 4) includes new global ratings to assess clinician's attention to client language, increased rigor in assessing autonomy support and client choice, and items to evaluate the use of persuasion when giving information and advice. METHOD: Four undergraduate, non-professional raters were trained in the MITI and used it to review 50 audiotapes of clinicians conducting MI in actual treatments sessions. Both kappa and intraclass correlation indices were calculated for all coders, for the best rater pair and for a 20% randomly selected sample from the best rater pair. RESULTS: Reliability across raters, with the exception of Emphasize Autonomy and % Complex Reflections, were in the good to excellent range. Reliability estimates decrease when smaller samples are used and when fewer raters contribute. CONCLUSION: The advantages and drawbacks of this revision are discussed including implications for research and clinical applications. The MITI 4.0 represents a reliable method for assessing the integrity of MI including both the technical and relational components of the method.

Therapist empathy, combined behavioral intervention, and alcohol outcomes in the COMBINE research project.

OBJECTIVE: Common factors such as therapist empathy play an important role in treatment for addictive behaviors. The present study was a secondary analysis designed to evaluate the relation between therapist empathy and alcohol treatment outcomes in data from a large, multisite, randomized controlled trial. METHOD: Audio-recorded psychotherapy sessions for 38 therapists and 700 clients had been randomly selected for fidelity coding from the combined behavioral intervention condition of Project COMBINE. Sessions were evaluated by objective raters for both specific content (coping with craving, building social skills, and managing negative mood) and relational components (empathy level of the therapist). Multilevel modeling with clients nested within therapists evaluated drinks per week at the end of treatment. RESULTS: Approximately 11% of the variance in drinking was accounted for by therapists. A within-therapist effect of empathy was detected (B = -0.381, SE = 0.103, p < .001); more empathy than usual was associated with subsequent decreased drinking. The Social and Recreational Counseling module (B = -0.412, SE = 0.124, p < .001), Coping with Cravings and Urges module (B = -0.362, SE = 0.134, p < .01), and the Mood Management module (B = -0.403, SE = 0.138, p < .01) were also associated with decreased drinking. No between-therapist effect was detected, and the Empathy × Module Content interactions were not significant. CONCLUSIONS: The results of the study appear consistent with the hypothesis that skills building and therapist empathy are independent contributions to the overall benefit derived from the combined behavioral intervention.

Outpatient treatment of alcohol use disorders among subjects 60+ years: design of a randomized clinical trial conducted in three countries (Elderly Study).

BACKGROUND: The proportion of 60+ years with excessive alcohol intake varies in western countries between 6-16 % among men and 2-7 % among women. Specific events related to aging (e.g. loss of job, physical and mental capacity, or spouse) may contribute to onset or continuation of alcohol use disorders (AUD). We present the rationale and design of a multisite, multinational AUD treatment study for subjects aged 60+ years. METHODS/DESIGN: 1,000 subjects seeking treatment for AUD according to DSM-5 in outpatient clinics in Denmark, Germany, and New Mexico (USA) are invited to participate in a RCT. Participants are randomly assigned to four sessions of Motivational Enhancement Treatment (MET) or to MET plus an add-on with eight sessions based on the Community Reinforcement Approach (CRA), which include a new module targeting specific problems of older adults. A series of assessment instruments is applied, including the Form-90, Alcohol Dependence Scale, Penn Alcohol Craving Scale, Brief Symptom Inventory and WHO Quality of Life. Enrolment will be completed by April 2016 and data collection by April 2017. The primary outcome is the proportion in each group who are abstinent or have a controlled use of alcohol six months after treatment initiation. Controlled use is defined as maximum blood alcohol content not exceeding 0.05 % during the last month. Total abstinence is a secondary outcome, together with quality of life andcompliance with treatment. DISCUSSION: The study will provide new knowledge about brief treatment of AUD for older subjects. As the treatment is manualized and applied in routine treatment facilities, barriers for implementation in the health care system are relatively low. Finally, as the study is being conducted in three different countries it will also provide significant insight into the possible interaction of service system differences and related patient characteristics in predictionof treatment outcome. TRIAL REGISTRATION: Clinical Trials.gov NCT02084173 , March 7, 2014.

Within-session communication patterns predict alcohol treatment outcomes.

BACKGROUND: Within-session client speech is theorized to be a key mechanism of behavior change in motivational interviewing (MI), a directional, client-centered approach to behavior change. Client change talk (CT: speech indicating movement toward changing a problematic health behavior) and sustain talk (ST: speech supporting continuing a problematic health behavior) have each shown relationships with outcomes. However, it may be the case that patterns of within-session client speech, rather than counts of client speech, are important for producing change. METHODS: Recorded initial MI/MET psychotherapy sessions from Project MATCH had been previously rated using the Motivational Interviewing Sequential Code for Observing Process Exchange (MI-SCOPE), a mutually exclusive and exhaustive sequential coding system. From these existing data, session conditional probabilities for transitions of interest (the transition from CT to more CT, and the transition from reflections of CT to CT) were analyzed as empirical Bayes estimates of log-normalized odds ratios. RESULTS: CT frequencies and these log-normalized odds ratios were entered as independent variables into longitudinal generalized estimating equation (GEE) models predicting within-treatment and post-treatment drinking. While all variables were significant predictors of within-treatment drinking, only the CT-CT transition emerged as a significant predictor of decreased drinking after treatment. CONCLUSIONS: The momentum of a client's speech about change during an MI session may be a better predictor of outcome than is a simple frequency count of it. Attending not only to the mere occurrence of CT, but also recognizing the importance of consecutive client statements of CT, may improve treatment outcomes.