Moderators, Mediators, and Nonspecific Predictors of Treatment Outcome in an Intervention for Everyday Task Improvement in Persons With Executive Deficits After Brain Injury

Objective: To identify moderators, mediators, and predictors of everyday task performance after an experimental combination of errorless learning and goal management training. Design: Predictor analysis of a randomized controlled intervention trial. Setting: Outpatient rehabilitation centers. Participants: Patients (N=60) with acquired brain injury of nonprogressive nature with a minimal postonset time of 3 months. Interventions: Participants were randomly allocated to 8 sessions of errorless or conventional goal management training. Main Outcome Measure: Everyday task performance, assessed at baseline and after treatment by evaluating correct, ineffective, and missing task steps. Results: Demographic variables, neuropsychological test performance, subjective cognitive function, and quality of life were selected as candidate predictors. The results showed that age (P=.03) and estimated intelligence quotient (IQ) (P=.02) emerged as moderators. Higher age was associated with better everyday task performance after conventional goal management training, whereas higher IQ was associated with better performance after errorless goal management training. Higher executive function scores after training predicted improved everyday task performance across the 2 treatment conditions (P=.04). Conclusions: The identified predictors may contribute to a more tailored cognitive rehabilitation approach in which treatments and patients are better matched when clinicians decide to train everyday tasks. (C) 2016 by the American Congress of Rehabilitation Medicine