A meta-analysis of the association between patients' early treatment outcome expectation and their posttreatment outcomes

Psychotherapy (Chic). 2018 Dec;55(4):473-485. doi: 10.1037/pst0000169.


Patients' outcome expectation (OE) represents their belief about the mental health consequences of participating in psychotherapy. A previous meta-analysis of 46 independent samples receiving the treatment of at least 3 sessions revealed a significant association between more optimistic baseline, or early treatment, OE and more adaptive posttreatment outcomes (weighted r = .12 or d = .24; Constantino, Glass, Arnkoff, Ametrano, & Smith, 2011). The present study represents an update to that meta-analysis. To be included, articles published through June 2017 had to (a) include a clinical sample, (b) include a therapist-delivered treatment of at least 3 sessions, (c) include a measure of patients' own OE, (d) include at least 1 posttreatment mental health outcome not explicitly referenced as a follow-up occasion, and (e) report a statistical test of the OE-outcome association. The updated meta-analysis was conducted on 81 independent samples (extracted from 72 references) with 12,722 patients. The overall weighted effect size was r = .18, p < .001, or d = .36, with high heterogeneity (I2 = 76%) and no evidence of publication bias. Several variables (patient age, measure type, and treatment manual used) moderated the OE-outcome association. These robust, replicated meta-analytic findings are discussed in light of methodological limitations and with regard to their practice implications. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Publication types

  • Meta-Analysis
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Attitude to Health*
  • Humans
  • Mental Disorders / psychology*
  • Mental Disorders / therapy*
  • Psychotherapy / methods*
  • Treatment Outcome