22 August 2024 in Administration and Policy in Mental Health and Mental Health Services Research
17 February 2024 in Psychotherapy Research
Given the importance of emotions in psychotherapy, valid measures are essential for research and practice. As emotions are expressed at different levels, multimodal measurements are needed for a nuanced assessment. Natural Language Processing (NLP) could augment the measurement of emotions. The study explores the validity of sentiment analysis in psychotherapy transcripts. We used a transformer-based NLP algorithm to analyze sentiments in 85 transcripts from 35 patients. Construct and criterion validity were evaluated using self- and therapist reports and process and outcome measures via correlational, multitrait-multimethod, and multilevel analyses. The results provide indications in support of the sentiments' validity. For example, sentiments were significantly related to self- and therapist reports of emotions in the same session. Sentiments correlated significantly with in-session processes (e.g., coping experiences), and an increase in positive sentiments throughout therapy predicted better outcomes after treatment termination. Sentiment analysis could serve as a valid approach to assessing the emotional tone of psychotherapy sessions and may contribute to the multimodal measurement of emotions. Future research could combine sentiment analysis with automatic emotion recognition in facial expressions and vocal cues via the Nonverbal Behavior Analyzer (NOVA). Limitations (e.g., exploratory study with numerous tests) and opportunities are discussed.
15 December 2023 in Administration and Policy in Mental Health and Mental Health Services Research
Outcome measurement including data-informed decision support for therapists in psychological therapy has developed impressively over the past two decades. New technological developments such as computerized data assessment, and feedback tools have facilitated advanced implementation in several seetings. Recent developments try to improve the clinical decision-making process by connecting clinical practice better with empirical data. For example, psychometric data can be used by clinicians to personalize the selection of therapeutic programs, strategies or modules and to monitor a patient’s response to therapy in real time. Furthermore, clinical support tools can be used to improve the treatment for patients at risk for a negative outcome. Therefore, measurement-based care can be seen as an important and integral part of clinical competence, practice, and training. This is comparable to many other areas in the healthcare system, where continuous monitoring of health indicators is common in day-to-day clinical practice (e.g., fever, blood pressure). In this paper, we present the basic concepts of a data-informed decision support system for tailoring individual psychological interventions to specific patient needs, and discuss the implications for implementing this form of precision mental health in clinical practice.
2 November 2023 in Psychotherapy Research
1 August 2020 in Journal of Consulting and Clinical Psychology
The study explored the presence of different patterns of change in a sample of patients who received cognitive therapy for depression sequentially in two different formats: group and individual. Our hypothesis was that patients' baseline characteristics (e.g., symptom severity, self-esteem) would discriminate patterns of response to group and individual therapy. 108 adults who met criteria for depression and completed the treatments included in a randomized controlled trial (RCT) were assessed with the Beck Depression Inventory-II (BDI-II), the Clinical Outcome in Routine Evaluation-Short Form B (CORE-SFB), the Global Assessment of Functioning (GAF), and the repertory grid technique. Growth mixture modeling was carried out to identify the patterns of change. Mixed linear models and repeated measures analysis of variance were performed to compare patients' characteristics in each pattern. Multinomial logistic regression was used to compute predictive models for the patterns from patients' characteristics. Finally, hierarchical linear regression was used to establish the power of each pattern to predict treatment outcome. A 3-class solution was obtained: group therapy improvers, individual therapy improvers, and nonimprovers. Group therapy improvers started therapy with less severe levels of depression and psychological distress, higher functioning and self-esteem, lower perceived social isolation, and lower dilemmatic construction of the self than the other groups. Individual therapy improvers and nonimprovers presented similar characteristics at baseline. However, a significant proportion of nonimprovers presented a concurrent diagnosis of fibromyalgia. The greater the impairment that patients present at baseline, the more likely they are to benefit from individual therapy after group therapy. A diagnosis of fibromyalgia can be considered a risk factor for therapy failure in the treatment of depression. (PsycInfo Database Record (c) 2020 APA, all rights reserved).
1 August 2024 in Journal of Consulting and Clinical Psychology
To date, many prediction studies in psychotherapy research have used cross-sectional data to predict treatment outcome. The present study used intensive longitudinal assessments and continuous time dynamic modeling (CTDM) to investigate the temporal dynamics of affective states and emotion regulation in the early phase of therapy and their ability to predict treatment outcome. Ninety-one patients undergoing psychological treatment at a university outpatient clinic took part in a 2-week ecological momentary assessment (EMA) period. Participants answered self-report measures on positive affect (PA), negative affect, and emotion regulation (ER) four times a day. Hierarchical Bayesian CTDM was conducted to identify temporal effects within (autoregressive) and between (cross-regressive) PA, negative affect, and ER. The resulting CTDM parameters, simple EMA parameters (e.g., mean), and cross-sectional predictors were entered into a LASSO model to be examined as predictors of treatment outcome at Session 15. Two significant predictors were identified: initial impairment and the continuous time cross-effect of PA on ER. The final model explained 40% of variance in treatment outcome, with the cross-effect (PA-ER) accounting for 4% of variance beyond initial impairment. The results demonstrate that temporal patterns of affective EMA data are valuable for the mapping of individual differences and the prediction of treatment outcome. This information can be used to provide therapists with feedback to personalize treatments. (PsycInfo Database Record (c) 2024 APA, all rights reserved).