Category: Health Care System / Public Policy
Keywords: Psychotherapy Outcome | Service Delivery | Community-Based Assessment / Intervention
Presentation Type: Symposium
Aims: Research demonstrates significant differences among therapists in both their skill and client outcomes. Therapist effectiveness data are rarely used in treatment decision making, and research on therapist effects rarely applies risk-adjustment for key client variables. The aims of this paper are to (a) describe results from a study that examined the stability and predictive validity of therapist effectiveness across outcome domains using risk-adjusted outcomes; and (b) discuss implications of therapist effects for psychotherapy practice, training, and mental health care policy.
Methods: Initial and follow-up multidimensional outcome data on the Treatment Outcome Package (TOP; Kraus et al., 2005) were collected on N = 3,540 clients treated in naturalistic settings by a sample of N = 59 therapists. After risk-adjusting outcomes using random forest models, outcome data from the first 30 clients of each therapist were used to classify each therapist’s effectiveness on 12 outcome domains. These results were then compared with outcome data from the therapist’s next 30 clients using multiple metrics, including hierarchical linear modeling-derived correlation coefficients.
Findings: After accounting for diverse client factors in a random forest model, therapist level outcome variance was as high as 18.28% for substance abuse (depression = 11.82%). Therapist effectiveness was relatively stable (r = .94 for substance abuse; r= .81 for depression), with some domain-specificity. Therapists classified as “exceptional” in a particular outcome domain were more likely to remain above average with future cases, suggesting that a therapist’s past performance is an important predictor of their future performance.
Conclusions: Mental health care systems should consider integrating therapist performance data in their decision making. Routine data can be used to derive algorithms that identify therapists with whom a particular client is more likely to experience treatment benefit, or is less likely to experience a negative treatment effect. Such therapist performance information has implications for psychotherapist training, case referral and assignments, and payment models.
Assistant Professor of Psychology
University at Albany, SUNY
Friday, November 17
8:30 AM – 10:00 AM
The asset you are trying to access is locked. Please enter your access key to unlock.