Individual patterns of activity predict the response to physical exercise as an intervention in mild to moderate depression

Abstract

AbstractPhysical exercise (PE) as antidepressive intervention is a promising alternative, as shown by multiple meta-analyses. However, there is no consensus regarding optimal intensity and duration of exercise, and there are no objective criteria available for personalized indication of treatment. The aims of this study were (1) to evaluate whether individual activity patterns before intervention can predict the response to treatment; and (2) to evaluate whether the patient outcome can be improved by using prior information on treatment efficacy at individual level. The study included subjects with mild to moderate depression randomized to three levels of exercise intensity in the Regassa study. Using a previously developed pipeline for data analysis, we have generated linear regression ensembles to predict the response to treatment using features extracted from actigraphy recordings. To understand the contribution of individual features, we performed a Bayesian analysis of coefficients, and found that different levels of PE intensity yield distinct signatures in enriched feature subsets. Next, we used the trained ensembles for a counterfactual analysis of response and remission rate provided prior knowledge of response to treatment outcome. The response to either PE regime was estimated for all patients, irrespective of original treatment assignment. Each patient was then virtually allocated to the PE regime predicted to yield best outcome, and the response and remission rates were compared against simulated random assignment to treatment. The counterfactual analysis showed that assignment to best individual PE regime yields significantly higher (increase by 28%) remission rates as compared to random assignment to treatment, which is accounted for by improved response in about 32% of the patients as compared to observed treatment outcome. While it is not possible to claim individual protocol optimization, our data suggest it may be possible to identify a PE regime to yield the best results for a patient based on individual circadian patterns of activity.