The alliance in adult psychotherapy: A meta-analytic synthesis.
Level 2 - randomized trial
Meta-analysis of correlational/observational studies examining alliance and outcomes
PubMed 29792475 · doi:10.1037/pst0000172
What was done
A three-level random-effects meta-analysis was performed on 295 independent studies published between 1978 and 2017, encompassing more than 30,000 adult psychotherapy patients. The authors examined the relationship between therapeutic alliance and treatment outcomes across face-to-face and internet-based therapy formats.
What was found
Face-to-face psychotherapy showed an overall alliance-outcome correlation of r = .278 (95% CI [.256, .299], p < .0001; equivalent d = .579). Internet-based psychotherapy demonstrated a similar association (r = .275 across k = 23 studies). Heterogeneity was present among effect sizes, and 2% of the 295 effect sizes were negative. The positive association was reported to remain consistent across assessor perspectives, outcome measures, treatment approaches, patient characteristics, and countries.
Why it matters
This large meta-analytic synthesis demonstrates that the therapeutic alliance is a consistent, moderate predictor of psychotherapy outcome across diverse treatment modalities and delivery formats.
Limits
The included data are correlational and cannot establish whether strong alliance causes better outcomes or reflects early symptomatic improvement. Significant heterogeneity was present among effect sizes, 2% of effects were negative, and specific diagnostic breakdowns or quality assessments are not reported in the abstract.
Cited by
- supports Empirical data show that effective psychotherapy with a strong therapeutic alliance and actionable tools produces significant life and mental health benefits.