Visualizing the dramatic variability in psychotherapy outcomes
Patients seeing therapists in the top 25% experienced effect sizes more than twice as large as those seeing bottom quartile therapists.
Source: Wampold & Brown (2005), N=6,146 patients
Clients of the most effective therapists were twice as likely to recover and 50% less likely to deteriorate compared to the least effective.
Source: Okiishi et al. (2003), N=1,841 patients, 91 therapists
Alarmingly, between 33-65% of therapists were classified as ineffective or harmful across different outcome domains.
Source: Kraus et al. (2011), N=6,960 patients, 696 therapists
This shows the actual patient outcomes depending on which type of therapist they see.
What the scale means: This 100-point scale represents change in symptoms (depression, anxiety, functioning)
from start to end of treatment. To put it in context: on a typical depression questionnaire (Beck Depression Inventory,
range 0-63), harmful therapists' patients get ~8 points worse, while effective therapists' patients improve ~16 points.
That's the difference between sliding into severe depression vs. achieving remission.
How common are these therapists? ~17% are harmful, ~40% are effective (see pie chart above for full breakdown).
Source: Kraus et al. (2011) - Effect sizes (d = -1.2 to +1.26) converted to illustrative scale
This shows both HOW MANY therapists fall into each category AND what that means for actual patient recovery rates. One standard deviation isn't trivial—it's the difference between 40% vs 60% of patients recovering.
Synthesized from Wampold & Brown (2005), Kraus et al. (2011), Okiishi et al. (2003)