Precision mental health care improves outcomes

Matching patients with therapists based on therapists’ performance strength can improve mental health care (MHC) outcomes, a recent randomized clinical trial has shown.

“Our study established a minimalist method of precision MHC. Results of the study showed that quality of MHC could be substantially improved by using therapist performance data to determine the specific patients to treat. This method provides stakeholders [ie, patients, therapists, and administrators] with shortlisted options of patients and therapists for optimizing MHC beyond chance levels,” said the authors. [JAMA Psychiatry 2021;doi:10.1001/jamapsychiatry.2021.1221]

In the study, 218 adult outpatients (mean age, 33.9 years; female, 67.4 percent; White, 88.5 percent) who had to make their own MHC decisions were recruited from six community MHC clinics in Cleveland, Ohio, US, between November 2017 and April 2019. Enrolled patients were randomly assigned to good-fitting therapists (matched group) or to therapists pragmatically (case assignment as usual [CAU] group).

Therapists’ pretrial performance was assessed from a mean of 28.48 historical patient cases based on patients’ pre- and post-therapy reporting across 12 problem domains of the routinely administered Treatment Outcome Package (TOP).

Overall, therapists had a mean of 1.56 strengths (domains classified as effective) and 0.96 weaknesses (domains classified as ineffective). A majority (87.5 percent) of therapists had ≥1 strength for which they could be matched or ≥1 weakness for which they could be avoided, with 12.5 percent of therapists classified as neutral (neither effective nor ineffective) across all domains, allowing them to be matched at level 5.

Patients in the matched group experienced a significant weekly reduction in general symptomatic and functional impairment across all TOP domains (γ₁₁₀, -0.03; 95 percent confidence interval [CI], -0.05 to -0.01; p=0.02; patient-level d=0.75) compared with those in the CAU group.

Similarly, significant weekly reductions in global distress (as measured by the Symptom Checklist-10) and domain-specific impairment were observed in patients in the matched group vs the CAU group (global distress: γ₁₁₀, -0.16; 95 percent CI, -0.30 to -0.02; patient-level d=0.50) (domain-specific impairment: γ₁₁₀, -0.01; 95 percent CI -0.01 to -0.006; patient-level d=0.60).

No adverse events were reported in both treatment groups.