Predicting patients' responses to psychotherapy: are some more predictable than others?

Patient profiling predicts improvement across psychotherapy sessions on the basis of intake clinical characteristics. Patients completed questionnaires before their 1st session and at other points during treatment. After using hierarchical linear modeling to calculate expected courses of improvement, 1/2 of the sample was divided into 1 of 2 groups: those whose treatment responses across sessions matched or exceeded their expected courses (75% of the sample) and those whose responses failed to match expectations. A discriminant function analysis indicated that the groups could be differentiated on the basis of intake characteristics; that is, patients with higher discriminant scores were more likely than those with low discriminant scores to have responded to psychotherapy as predicted. Successful cross-validation of the predictor weights was performed with the other 1/2 of the sample.