Blog

July 2021

• Matching Patients to Therapists’ Strengths

  Constantino, M.J., Boswell, J.F., Coyne, A.E., Swales, T.P., & Kraus, D.R. (2021). Effect of matching therapists to patients vs assignment as usual on adult psychotherapy outcomes: A randomized clinical trial. JAMA Psychiatry, doi: 10:1001/jamapsychiatry.2021.1221.

  We know that about 60% of patients do not benefit fully from the psychotherapy they receive and that therapists differ in their effectiveness. In one study, above average therapists were twice as effective as below average therapists. In fact, most therapists are reliably effective for some presenting problems, but not for all. What if we could match a patient to a therapist based on what the therapist is good at? In this unique study in a large mental health treatment network, Constantino and colleagues evaluated the past effectiveness of 48 therapists who treated at least 5 patients each across a variety of patient problem domains. The authors identified those patient problem domains for which each therapist was reliably effective or ineffective using valid psychometric measures. Then the researchers randomly assigned 218 patients in the mental health treatment network to receive care as usual (patients were assigned to therapists based on which therapist was available at the time of referral), or to receive treatment from a therapist who was reliably effective for the patient’s problems. Out of 9 possible patient problem domains, therapists had an average of 1.56 (SD = 1.66) patient domains for which they were reliably effective and an average of 0.96 (SD = 1.65) patient problem domains for which they were reliably ineffective. Over 87% of therapists had at least one strength on which they could be matched to a patient, and 10.4% were reliably ineffective for all patient problem domains. To a moderate degree, patients treated by a therapist matched to their problem domain experienced greater weekly reductions in their general impairment compared with patients in the care as usual group (d = 0.75). Relative to their own average outcomes, a therapist achieved better patient outcomes when treating a matched patient than when treating a care as usual patient.

  Practice Implications

  This is the first study of its kind and so it needs to be replicated. Most therapists had a few patient domains for which they were effective. About 10% of therapists were ineffective across all patient domains. The results of the trial suggested that therapists are likely more effective with some patients than others, depending on the patient’s problems. Therapists should assess their outcomes with a range of patient problems and evaluate their effectiveness using valid psychometric instruments. If a therapist finds that they are less effective with some patients, then the therapist should receive more training and supervision for treating that patient problem.