Personalized prediction of response to smartphone-delivered meditation training

A meditation app called the Healthy Minds Program (HMP) can reduce psychological distress, but its benefits vary by individual. Using data from a randomized controlled trial of 662 school system employees who used the app for four weeks or served as controls, researchers built an algorithm that predicts who will benefit most. The algorithm, called a Personalized Advantage Index, was based on baseline clinical and demographic traits and successfully identified individuals who showed greater distress reduction with the app versus no intervention. A simpler model using only repetitive negative thinking worked nearly as well. The approach could help people decide whether to try a meditation app.

Study at a glance

Abstract

Meditation apps are popular and may reduce psychological distress, including during the COVID-19 pandemic. However, it is not clear who is most likely to benefit. Using randomized controlled trial data comparing a 4-week meditation app (Healthy Minds Program; HMP) with an assessment-only control in school system employees (n=662), we developed an algorithm predicting who is most likely to benefit from HMP. Baseline clinical and demographic characteristics were submitted to a machine learning model to develop a “Personalized Advantage Index” (PAI) reflecting an individual’s expected reduction in distress (preregistered primary outcome) from HMP vs. control. Significant Group x PAI interactions emerged, indicating that PAI scores moderated group differences in outcome. A regression model including repetitive negative thinking as the sole predictor performed comparably well. Finally, we demonstrate the translation of predictive models to personalized recommendations of expected benefit, which could inform users’ decisions of whether to engage with a meditation app.