Relational depth as predictor of mental health outcomes in psychotherapy: random intercept cross-lagged analyses

Abstract

Background: Relational Depth (RD) in psychotherapy patients has been associated with positive outcomes using qualitative and correlational designs. No research has assessed whether the frequency of RD over the course of therapy can predict outcomes using a more robust longitudinal design in a clinical setting. Methods: In a UK psychotherapy clinic, 1,086 clients (501 females, 234 males, 347 “prefer not to say,” and 4 other) completed the Relational Depth Frequency Scale (RDFS), the Agnew Relationship Measure-5 (ARM-5, therapeutic alliance), the CORE-10 (psychological distress, primary outcome), the Patient Health Questionnaire-9 (PHQ-9) (depression), and the Generalized Anxiety Disorder-7 (GAD-7) (anxiety) at intake and then every four sessions over the course of long-term psychotherapy. Random Intercept Cross-Lagged Panel Models (RI-CLPM) were used to assess the associations between variables within participant. Results: Frequency of RD was a significant predictor of change in depression (t = −2.56, p =.010; r = −.060 medium effect size), and anxiety symptoms (t = −2.081, p =.037; r = −.050 medium effect size), but did not predict change in general psychological distress (t = −1.545, p =.122; r = −.038 small effect size). None of the outcome variables predicted future change in RDFS. The alliance was not a significant predictor of (nor predicted by) change on the three outcome variables. RDFS was a predictor of alliance (t = 3.510, p