Algorithms for Empathy: Using Machine Learning to Categorize Common Empathetic Traits Across Professional and Peer-Based Conversations

Introduction In this article, we describe our creation of a machine-learning model that uses a combination of rule-based and natural language processing (NLP) algorithms. We show how this "Empathy Algorithm" was developed and how its results compare to three datasets of professional counseling and peer-led conversations.  Methods These conversation datasets were rated by people with varying degrees of empathetic expertise (from counselors to student volunteers) and labeled as either low- or high-quality empathy. Our methodology involved running both these "low-empathy" and "high-empathy" conversations through our algorithm and then looking for a correlation between conversations labeled "high empathy" and an increased presence of six empathy skills flagged by our algorithm.  Results We found positive correlations between four of the six skills that our algorithm measures (i.e., four empathizing skills showed up the same or more in each of the "high-empathy" conversations within the three datasets). This suggests that certain empathizing skills are not only consistently present in effective conversations but also quantifiable enough to be measured by today's machine-learning models. Conclusion While limitations of language, binary classifications, and non-verbal cues remain as opportunities for further development, using algorithms to objectively assess empathic skills represents an important step to improve client outcomes and refine communication practices for today's healthcare professionals.

Percentage of Heavy Drinking Days Following Psilocybin-Assisted Psychotherapy vs Placebo in the Treatment of Adult Patients With Alcohol Use Disorder: A Randomized Clinical Trial

IMPORTANCE

Although classic psychedelic medications have shown promise in the treatment of alcohol use disorder (AUD), the efficacy of psilocybin remains unknown.

OBJECTIVE

To evaluate whether 2 administrations of high-dose psilocybin improve the percentage of heavy drinking days in patients with AUD undergoing psychotherapy relative to outcomes observed with active placebo medication and psychotherapy.

DESIGN, SETTING, AND PARTICIPANTS

In this double-blind randomized clinical trial, participants were offered 12 weeks of manualized psychotherapy and were randomly assigned to receive psilocybin vs diphenhydramine during 2 day-long medication sessions at weeks 4 and 8. Outcomes were assessed over the 32-week double-blind period following the first dose of study medication. The study was conducted at 2 academic centers in the US. Participants were recruited from the community between March 12, 2014, and March 19, 2020. Adults aged 25 to 65 years with a DSM-IV diagnosis of alcohol dependence and at least 4 heavy drinking days during the 30 days prior to screening were included. Exclusion criteria included major psychiatric and drug use disorders, hallucinogen use, medical conditions that contraindicated the study medications, use of exclusionary medications, and current treatment for AUD.

INTERVENTIONS

Study medications were psilocybin, 25 mg/70 kg, vs diphenhydramine, 50 mg (first session), and psilocybin, 25-40 mg/70 kg, vs diphenhydramine, 50-100 mg (second session). Psychotherapy included motivational enhancement therapy and cognitive behavioral therapy.

MAIN OUTCOMES AND MEASURES

The primary outcome was percentage of heavy drinking days, assessed using a timeline followback interview, contrasted between groups over the 32-week period following the first administration of study medication using multivariate repeated-measures analysis of variance.

RESULTS

A total of 95 participants (mean [SD] age, 46 [12] years; 42 [44.2%] female) were randomized (49 to psilocybin and 46 to diphenhydramine). One participant (1.1%) was American Indian/Alaska Native, 3 (3.2%) were Asian, 4 (4.2%) were Black, 14 (14.7%) were Hispanic, and 75 (78.9%) were non-Hispanic White. Of the 95 randomized participants, 93 received at least 1 dose of study medication and were included in the primary outcome analysis. Percentage of heavy drinking days during the 32-week double-blind period was 9.7% for the psilocybin group and 23.6% for the diphenhydramine group, a mean difference of 13.9%; (95% CI, 3.0-24.7; F1,86 = 6.43; P = .01). Mean daily alcohol consumption (number of standard drinks per day) was also lower in the psilocybin group. There were no serious adverse events among participants who received psilocybin.

CONCLUSIONS AND RELEVANCE

Psilocybin administered in combination with psychotherapy produced robust decreases in percentage of heavy drinking days over and above those produced by active placebo and psychotherapy. These results provide support for further study of psilocybin-assisted treatment for AUD.

TRIAL REGISTRATION

ClinicalTrials.gov Identifier: NCT02061293.

The Psychedelic Debriefing in Alcohol Dependence Treatment: Illustrating Key Change Phenomena through Qualitative Content Analysis of Clinical Sessions

Research on the clinical applications of psychedelic-assisted psychotherapy has demonstrated promising early results for treatment of alcohol dependence. Detailed description of the content and methods of psychedelic-assisted psychotherapy, as it is conducted in clinical settings, is scarce.

Methods: An open-label pilot (proof-of-concept) study of psilocybin-assisted treatment of alcohol dependence (NCT01534494) was conducted to generate data for a phase 2 RCT (NCT02061293) of a similar treatment in a larger population. The present paper presents a qualitative content analysis of the 17 debriefing sessions conducted in the pilot study, which occurred the day after corresponding psilocybin medication sessions.

Results: Participants articulated a series of key phenomena related to change in drinking outcomes and acute subjective effects of psilocybin.

Discussion: The data illuminate change processes in patients' own words during clinical sessions, shedding light on potential therapeutic mechanisms of change and how participants express effects of psilocybin. This study is unique in analyzing actual clinical sessions, as opposed to interviews of patients conducted separately from treatment.

How Much Is Too Much? Patterns of Drinking During Alcohol Treatment and Associations With Post-Treatment Outcomes Across Three Alcohol Clinical Trials

OBJECTIVE

This secondary data analysis examined patterns of drinking during alcohol treatment and associated drinking outcomes during the first year following treatment. The goal was to provide clinicians with guidance on which patients may be most at risk for negative long-term outcomes based on drinking patterns during treatment.

METHOD

This study was an analysis of existing data (N = 3,851) from three randomized clinical trials for alcohol use disorder: the COMBINE Study (n = 1,383), Project MATCH (n = 1,726), and the United Kingdom Alcohol Treatment Trial (n = 742). Indicators of abstinence, non-heavy drinking, and heavy drinking (defined as 4/5 or more drinks per day for women/men) were examined during each week of treatment using repeated-measures latent class analysis. Associations between drinking patterns during treatment and drinking intensity, drinking consequences, and physical and mental health 12 months following intake were examined.

RESULTS

Seven drinking patterns were identified. Patients who engaged in persistent heavy drinking throughout treatment and those who returned to persistent heavy drinking during treatment had the worst long-term outcomes. Patients who engaged in some heavy drinking during treatment had better long-term outcomes than persistent heavy drinkers. Patients who reported low-risk drinking or abstinence had the best long-term outcomes. There were no differences in outcomes between low-risk drinkers and abstainers.

CONCLUSIONS

Abstinence, low-risk drinking, or even some heavy drinking during treatment are associated with the best long-term outcomes. Patients who are engaging in persistent heavy drinking are likely to have the worst outcomes and may require a higher level of care.

Development of a Psychotherapeutic Model for Psilocybin-Assisted Treatment of Alcoholism

Research activity on the potential clinical value of classic hallucinogens and other psychedelics has increased markedly in the past two decades, and promises to continue to expand. Experimental study of hallucinogen-assisted treatment, and any future clinical use, requires the development of psychotherapeutic models that are appropriate to the disorder being treated and effectively integrated with the pharmacologic component of the treatment. To provide a framework for thinking about possible treatment models, we provide an overview of the history of psychedelic-assisted treatment, review what is known about the therapeutic mechanisms of these treatments, and consider the various purposes of psychotherapy in the context of both research and clinical use of psychedelic-assisted treatment. We then provide a description of a therapy model we have developed and are currently using in a trial of psilocybin-assisted treatment for alcoholism. Finally, we discuss advantages and disadvantages of a range of alternative models, emphasizing the need for research to determine the most effective treatment models for any indications for which efficacy becomes established.

Psilocybin-assisted treatment for alcohol dependence: a proof-of-concept study

Several lines of evidence suggest that classic (5HT2A agonist) hallucinogens have clinically relevant effects in alcohol and drug addiction. Although recent studies have investigated the effects of psilocybin in various populations, there have been no studies on the efficacy of psilocybin for alcohol dependence. We conducted a single-group proof-of-concept study to quantify acute effects of psilocybin in alcohol-dependent participants and to provide preliminary outcome and safety data. Ten volunteers with DSM-IV alcohol dependence received orally administered psilocybin in one or two supervised sessions in addition to Motivational Enhancement Therapy and therapy sessions devoted to preparation for and debriefing from the psilocybin sessions. Participants' responses to psilocybin were qualitatively similar to those described in other populations. Abstinence did not increase significantly in the first 4 weeks of treatment (when participants had not yet received psilocybin), but increased significantly following psilocybin administration (p < 0.05). Gains were largely maintained at follow-up to 36 weeks. The intensity of effects in the first psilocybin session (at week 4) strongly predicted change in drinking during weeks 5-8 (r = 0.76 to r = 0.89) and also predicted decreases in craving and increases in abstinence self-efficacy during week 5. There were no significant treatment-related adverse events. These preliminary findings provide a strong rationale for controlled trials with larger samples to investigate efficacy and mechanisms.

TRIAL REGISTRATION

NCT02061293.

Brief intervention for patients with problematic drug use presenting in emergency departments: a randomized clinical trial

IMPORTANCE

Medical treatment settings such as emergency departments (EDs) present important opportunities to address problematic substance use. Currently, EDs do not typically intervene beyond acute medical stabilization.

OBJECTIVE

To contrast the effects of a brief intervention with telephone boosters (BI-B) with those of screening, assessment, and referral to treatment (SAR) and minimal screening only (MSO) among drug-using ED patients.

DESIGN, SETTING, AND PARTICIPANTS

Between October 2010 and February 2012, 1285 adult ED patients from 6 US academic hospitals, who scored 3 or greater on the 10-item Drug Abuse Screening Test (indicating moderate to severe problems related to drug use) and who were currently using drugs, were randomized to MSO (n = 431), SAR (n = 427), or BI-B (n = 427). Follow-up assessments were conducted at 3, 6, and 12 months by blinded interviewers.

INTERVENTIONS

Following screening, MSO participants received only an informational pamphlet. The SAR participants received assessment plus referral to addiction treatment if indicated, and the BI-B participants received assessment and referral as in SAR, plus a manual-guided counseling session based on motivational interviewing principles and up to 2 "booster" sessions by telephone during the month following the ED visit.

MAIN OUTCOMES AND MEASURES

Outcomes evaluated at follow-up visits included self-reported days using the patient-defined primary problem drug, days using any drug, days of heavy drinking, and drug use based on analysis of hair samples. The primary outcome was self-reported days of use of the patient-defined primary problem drug during the 30-day period preceding the 3-month follow-up.

RESULTS

Follow-up rates were 89%, 86%, and 81% at 3, 6, and 12 months, respectively. For the primary outcome, estimated differences in number of days of use (95% CI) were as follows: MSO vs BI-B, 0.72 (-0.80 to 2.24), P (adjusted) = .57; SAR vs BI-B, 0.70 (-0.83 to 2.23), P (adjusted) = .57; SAR vs MSO, -0.02 (-1.53 to 1.50), P (adjusted) = .98. There were no significant differences between groups in self-reported days using the primary drug, days using any drug, or heavy drinking days at 3, 6, or 12 months. At the 3-month follow-up, participants in the SAR group had a higher rate of hair samples positive for their primary drug of abuse (265 of 280 [95%]) than did participants in the MSO group (253 of 287 [88%]) or the BI-B group (244 of 275 [89%]). Hair analysis differences between groups at other time points were not significant.

CONCLUSIONS AND RELEVANCE

In this sample of drug users seeking emergency medical treatment, a relatively robust brief intervention did not improve substance use outcomes. More work is needed to determine how drug use disorders may be addressed effectively in the ED.

TRIAL REGISTRATION

clinicaltrials.gov Identifier:NCT01207791.

Does treatment readiness enhance the response of African American substance users to Motivational Enhancement Therapy?

The development of effective treatments for African Americans and other ethnic minorities is essential for reducing health disparities in substance use. Despite research suggesting that Motivational Enhancement Therapy (MET) may reduce substance use among African Americans, the findings have been inconsistent. This research examined the extent to which readiness-to-change (RTC) affects response to MET among African American substance users. The study was a secondary analysis of the 194 African American substance users participating in a multisite randomized clinical trial evaluating MET originally conducted within the National Drug Abuse Treatment Clinical Trials Network. The participants were randomly assigned to receive either three sessions of MET or Counseling-As-Usual (CAU) followed by the ordinary treatment and other services offered at the five participating outpatient programs. Participants were categorized as either high or lower on RTC based on their scores on the University of Rhode Island Change Assessment. The participants reported their substance use at baseline and throughout the 16 weeks after randomization. Among the high RTC participants, those in MET tended to report fewer days of substance use per week over time than participants in CAU. However, among the lower RTC participants, the CAU group tended to report fewer days of substance use over time than MET participants. In contrast to previous thinking, the findings suggest that MET may be more effective for high than lower RTC African American participants.

Test-retest reliability of self-report measures in a dually diagnosed sample

Many individuals diagnosed with a substance use disorder are also diagnosed with another psychiatric disorder. Little is known regarding which treatments are efficacious for these dually diagnosed individuals (DDI). Characterizing the psychometric properties of assessments used with DDI samples is essential to efficacy studies with DDI. This study examined the internal consistency and test-retest reliability of self-report instruments among DDI. Most subscales demonstrated high test-retest reliability; one subscale demonstrated poor reliability. Internal consistency was similar to that of non-DDI samples. This exploratory study suggests that, while some instruments should be interpreted cautiously, DDI samples can be accurately assessed with self-report measures.

Design of NIDA CTN Protocol 0047: screening, motivational assessment, referral, and treatment in emergency departments (SMART-ED)

BACKGROUND

Medical settings such as emergency departments (EDs) present an opportunity to identify and provide services for individuals with substance use problems who might otherwise never receive any form of assessment, referral, or intervention. Although screening, brief intervention, and referral to treatment models have been extensively studied and are considered effective for individuals with alcohol problems presenting in EDs and other medical settings, the efficacy of such interventions has not been established for drug users presenting in EDs.

OBJECTIVES

This article describes the design of a NIDA Clinical Trials Network protocol testing the efficacy of an screening, brief intervention, and referral to treatment model in medical EDs, highlighting considerations that are pertinent to the design of other studies targeting substance use behaviors in medical treatment settings.

METHODS

The protocol is described, and critical design decisions are discussed.

RESULTS

Design challenges included defining treatment conditions, study population, and site characteristics; developing the screening process; choosing the primary outcome; balancing brevity and comprehensiveness of assessment; and selecting the strategy for statistical analysis.

CONCLUSION

Many of the issues arising in the design of this study will be relevant to future studies of interventions for addictions in medical settings.

SCIENTIFIC SIGNIFICANCE

Optimal trial design is critical to determining how best to integrate substance abuse interventions into medical care.

American Indian methamphetamine and other drug use in the Southwestern United States

To investigate the extent of methamphetamine and other drug use among American Indians (AIs) in the Four Corners region, we developed collaborations with Southwestern tribal entities and treatment programs in and around New Mexico. We held nine focus groups, mostly with Southwestern AI participants (N = 81) from three diverse New Mexico communities to understand community members, treatment providers, and clients/relatives views on methamphetamine. We conducted a telephone survey of staff (N = 100) from agencies across New Mexico to assess perceptions of methamphetamine use among people working with AI populations. We collected and analyzed self-reported drug use data from 300 AI clients/relatives who completed the Addiction Severity Index (ASI) in the context of treatment at three diverse addiction treatment programs. Each focus group offered a unique perspective about the effect of drugs and alcohol on each respective community. Though data from the phone surveys and ASIs suggested concerning rates of methamphetamine use, with women more adversely affected by substance use in general, alcohol was identified as the biggest substance use problem for AI populations in the Southwest. There appears to be agreement that methamphetamine use is a significant problem in these communities, but that alcohol is much more prevalent and problematic. There was less agreement about what should be done to prevent and treat methamphetamine use. Future research should attend to regional and tribal differences due to variability in drug use patterns, and should focus on identifying and improving dissemination of effective substance use interventions.

Motivational enhancement therapy for African American substance users: a randomized clinical trial

Limited empirical evidence concerning the efficacy of substance abuse treatments among African Americans reduces opportunities to evaluate and improve program efficacy. The current study, conducted as a secondary analysis of a randomized clinical trial conducted by the Clinical Trials Network of the National Institute of Drug Abuse, addressed this knowledge gap by examining the efficacy of motivational enhancement therapy (MET) compared with counseling as usual (CAU) among 194 African American adults seeking outpatient substance abuse treatment at 5 participating sites. The findings revealed higher retention rates among women in MET than in CAU during the initial 12 weeks of the 16-week study. Men in MET and CAU did not differ in retention. However, MET participants self-reported more drug-using days per week than participants in CAU. Implications for future substance abuse treatment research with African Americans are discussed.

EFFECT OF JOB SKILLS TRAINING ON EMPLOYMENT AND JOB SEEKING BEHAVIORS IN AN AMERICAN INDIAN SUBSTANCE ABUSE TREATMENT SAMPLE

Employment difficulties are common among American Indian individuals in substance abuse treatment. To address this problem, the Southwest Node of NIDA's Clinical Trials Network conducted a single-site adaptation of its national Job Seekers Workshop study in an American Indian treatment program, Na'Nizhoozhi Center (NCI). 102 (80% men, 100% American Indian) participants who were in residential treatment and currently unemployed were randomized to (1) a three session, manualized program (Job seekers workshop: JSW) or (2) a 40-minute Job Interviewing Video: JIV). Outcomes were assessed at 3-month follow up: 1) number of days to a new taxed job or enrollment in a job-training program, and 2) total hours working or enrolled in a job-training program. No significant differences were found between the two groups for time to a new taxed job or enrollment in a job-training program. There were no significant differences between groups in substance use frequency at 3-month follow-up. These results do not support the use of the costly and time-consuming JSW intervention in this population and setting. Despite of the lack of a demonstrable treatment effect, this study established the feasibility of including a rural American Indian site in a rigorous CTN trial through a community-based participatory research approach.

Motivational Interviewing with Underage College Drinkers: A Preliminary Look at the Role of Empathy and Alliance

This study evaluated the impact of a motivational interview (MI) on alcohol use in underage college drinkers, and examined the specific role of empathy and alliance in MI. Fifty-five underage heavy drinkers were randomized to a one-session MI or no-treatment control. Empathy and alliance were evaluated through the MITI, participant, and therapist ratings. At two-month follow-up, multivariate tests reveal no significant interaction effects. Means comparisons indicated that both groups showed reductions in alcohol-related problems, however, only the MI sample evidenced significant reductions in binge drinking. In addition, despite the reductions of both groups, effect sizes indicated that the MI group outperformed the control in terms of binge-drinking and alcohol-related problems. Contrary to predictions, empathy and alliance showed no relationships with outcomes.

Integrating spirituality into alcohol treatment

Spirituality is presumed by millions of Americans to be directly relevant to problems of alcohol abuse. We summarize findings regarding the role of religion and spirituality in the prevention and treatment of substance abuse and present a case illustration. We also consider mechanisms responsible for these effects. We offer advice about why, by whom, and how religion and spirituality should be discussed with clients with substance use disorders. In a recent clinical trial, therapists trained in a client-centered approach to facilitate exploration of spirituality fostered clients' use of spiritual practices. We suggest that the therapist's ability to skillfully engage clients in a discussion of spirituality is largely determined by how the therapist balances the dual roles of authoritative expert and evocative facilitator.

Psychometrics of the Drinker Inventory of Consequences (DrInC)

The Drinker Inventory of Consequences (DrInC) was developed for Project MATCH to assess the consequences of drinking in five domains. The present study, using baseline data from the 1382 participants involved in the COMBINE Study, analyzed the psychometrics of the 50-item DrInC and its shorter form, the 15-item SIP. Findings indicate the DrInC is reliable, valid, and clinically useful and that DrInC subscales are internally consistent and non-redundant. In an examination of the shorter version of the DrInC, findings suggest that the SIP is suitable when assessing the overall level of drinking-related consequences.

Course of recovery from alcoholism

This article represents the proceedings of a symposium at the 2005 Research Society on Alcoholism meeting in Santa Barbara, California, organized and chaired by Kamilla L. Venner. This symposium integrated current empirical research on the course of recovery from alcoholism from multiple perspectives, an aim that is consistent with NIAAA's new focus on the process of recovery. The presentations and presenters were as follows: (1) The Role of Community Services and Informal Support on 7-Year Drinking Outcomes in Treated and Untreated Drinkers, by Helen Matzger; (2) The Sequence of Recovery Events in a Native American Sample, by Kamilla L. Venner; (3) Transformational Change in Recovery, by Alyssa A. Forcehimes; (4) Social Settings and Substance Use: Contextual Factors in Recovery, by Rudolf H. Moos; and (5) A Broader View of Change in Drinking Behavior, by discussant Mark L. Willenbring. A theme connecting the presentations was that treatment is but one discrete aspect to recovery and that sustained recovery is often influenced by an individual interaction with others within a social context. Collectively, presentations underscored the need to think more broadly about factors contributing to the remission of alcohol dependence.

De profundis: spiritual transformations in Alcoholics Anonymous

The mechanism of change in Alcoholics Anonymous is described as "spiritual transformation." A.A. acknowledges that such transformation can occur gradually; however, nearly all of the examples presented in the Big Book of A.A. involve discrete and sudden experiences that resemble the phenomenon of quantum change. The sequence offered describes how spiritual transformations transpire. The sequence begins with hitting bottom, recognition of inability to control the problem. A feeling of contrition follows, describing not only sorrow for the present state, but also desire for a new way. The final step is the act of surrendering one's will to a higher power. The de profundis sequence sets the process of spiritual transformation in motion, offering stabilization to sobriety.