Different people respond differently to therapy: A demonstration using patient profiling and risk stratification

Area of Science:

• Clinical Psychology
• Psychotherapy Research
• Health Informatics

Background:

• Identifying patient characteristics linked to adverse psychological therapy outcomes is crucial.
• Developing effective patient profiling methods can enhance treatment personalization.

Purpose of the Study:

• To identify patient factors predicting poor outcomes in psychological therapy.
• To create a patient profiling tool for personalized intervention strategies.

Main Methods:

• Analysis of clinical assessment data from 1347 outpatients.
• Logistic regression used to identify predictors of reliable and clinically significant improvement (RCSI).
• Development of the Leeds Risk Index (LRI) based on regression coefficients.

Main Results:

• Disability, employment status, age, functional impairment, baseline depression, and outcome expectancy predicted RCSI.
• The Leeds Risk Index (LRI) stratified patients into risk groups with differing RCSI and completion rates.
• LRI scores correlated with treatment session progress ('not on track' status).

Conclusions:

• The Leeds Risk Index (LRI) effectively identifies patients at risk of poor psychological therapy progress.
• LRI facilitates personalized treatment recommendations for both low and high-intensity interventions.
• This tool supports tailored mental health care strategies based on individual patient risk profiles.