Psychology and Psychotherapy: Theory, Research and Practice

Skip to Search

Skip to Navigation

Volume 89 Issue 4 (December 2016), Pages 369-482

Early symptom change in adult outpatients: Relationship with patient characteristics and therapeutic alliance (pages 402-417)


This study has three aims: (1) to identify early change trajectories, (2) to examine patient characteristics predicting trajectory classes, and (3) to investigate the moderating effect of trajectory class membership on the working alliance.


In a sample of 402 outpatients (M age = 38.27, 64.9% women), outcome scores (OQ‐45) on three time points (sessions 1, 3, and 5) were analysed by means of Latent Class Growth Curve Analysis. A multinomial logistic regression was used to investigate whether patient's age, gender, initial distress (OQ‐45), and personality traits (Dimensional Assessment of Personality Pathology–Short Form) predicted trajectory class membership. A repeated‐measures ANOVA examined whether differences in trajectory classes moderated the alliance quality (Working Alliance Inventory–Short version) over time.


Four trajectory classes were identified: High distress–no change; moderate/high distress–small improvement, low distress–moderate improvement, and moderate/low distress–strong improvement. Gender, initial distress, and emotion dysregulation significantly predicted patients’ trajectory class membership. The Task/Goal alliance component increased linearly over time, whereas the Bond component followed a reversed U‐shaped pattern. The alliance was not moderated by trajectory class membership.


Four clinically relevant subgroups could be identified on the basis of the early symptom change trajectories. Gender, initial psychological distress, and emotion dysregulation predicted patient's trajectory class membership. Trajectory class membership did not influence the development of the early working alliance.

Practitioner points

  • As early change in highly distressed patients (Axis I and II) is not characterized by significant symptom change, clinicians should support and facilitate emotion regulation and social skills.
  • Investing in a strong alliance is recommended, although it does not differentiate early responders from non‐responders.

Add This link

Bookmark and Share>