Flavio Iovoli, Saskia Rieger, Juan Martín Gómez Penedo, Jessica Fritz, Martin Cordes, Ruben Lauterbach, Henning Schöttke, Julian A Rubel
Models demonstrated low predictive performance overall. Resampling strategies influenced outcomes, with SMOTE consistently improving F1-scores compared with no resampling. The combination of random forest with SMOTE achieved the highest F1-score (0.303). Five key predictors of dropout emerged: younger age, unemployment, a higher number of comorbid Axis I disorders, absence of a partner, and moderate depression severity. All models showed reduced performance during external validation.
BACKGROUND: Therapy goals are central to psychotherapeutic interventions. While symptom-specific goals are commonly emphasized, patients often set diverse goals that may not align with standardized symptom measures. This study examines how patient characteristics are associated with the selection of specific therapy goal content categories and whether attainment in these domains shows associations with different psychotherapy outcomes.
METHOD: Therapy goals from 645 patients undergoing cognitive-behavioral therapy in an outpatient clinic were categorized using the Bern Inventory of Treatment Goals and analyzed. Baseline predictors of therapy goal selection and diversity were examined using multilevel logistic and ordinal regression models. The association between goal selection and their goal attainment with residualized post-treatment scores in symptoms, global functioning, and end-of-treatment patient satisfaction were assessed using Bayesian multilevel regression models.
RESULTS: Baseline patient characteristics were associated with the selection of specific therapy goal categories, alongside substantial therapist-level effects in several categories. Goal-category attainment showed differential associations with psychotherapy outcomes, with the strongest and most consistent effects observed for symptom-related outcomes.
CONCLUSION: These findings indicate that the attainment of specific therapy goal categories is meaningfully reflected in standardized outcome measures, but in a category-sensitive rather than uniform way. Combining therapy goal-based and standardized assessments may improve the alignment between patient priorities and outcome evaluation in routine psychotherapy.