Therapist effects and deliberate practice

Therapist effects account for approximately 5–8% of outcome variance; treatment differences account for roughly 1%. The best therapists in a service reliably achieve substantially better outcomes than the worst, across comparable caseloads, and the difference is not explained by which model they were trained in.

The uncomfortable findings

• Years of experience shows little or no relationship with outcome; some studies find small declines.
• Therapist self-assessment of effectiveness is essentially unrelated to actual outcomes, and clinicians place themselves on average well above the median.
• Confidence is not calibrated: clinicians identify only a minority of deteriorating cases without data.
• Some therapists are reliably worse than others with specific populations, and cannot tell which.

What does predict better outcomes

1. 1Facilitative interpersonal skills, measured by response to standardised difficult client vignettes.
2. 2Professional self-doubt — an appropriate uncertainty about one’s own effectiveness, which predicts better outcomes rather than worse.
3. 3Deliberate practice: solitary, effortful, feedback-driven rehearsal of specific skills at the edge of current ability.
4. 4Use of outcome data, and acting on it.