Abstract
Maladaptive exercise (MalE) includes excessive, compulsive, or compensatory exercise and is a common eating-disorder (ED) symptom associated with increased severity, slower rates-of-recovery, and faster rates-of-relapse. Affect-regulation theories posit that MalE functions to reduce high negative affect (NA), although support for the affect-regulation model is mixed. Previous studies have not integrated ecological momentary assessment (EMA) with accelerometry or examined the affect-regulation model in individuals with EDs who frequently engage in MalE. The objective of this study was to examine trajectories of NA and positive affect (PA) through a 7-day EMA study combined with wrist-worn accelerometry in women with EDs (N=84). Piecewise generalized mixed-effects regression models evaluated the trajectories of PA and NA in the hours leading up to and following self-reported exercise and exercise identified via accelerometry. NA was increasing before self-reported exercise, although NA did not meaningfully change relative to objectively measured exercise. PA was increasing prior to exercise and decreasing after exercise, and this pattern was consistent for both self-reported and objectively measured exercise. Rates of rising PA were steeper in the hours leading up to higher intensity exercise episodes. Though inconsistent with affect-regulation models, the current study offers preliminary evidence that exercise is associated with disrupted affective responses among women with EDs who regularly engage in MalE. Results suggest that planning or anticipating high-intensity exercise may be rewarding for a considerable proportion of people with EDs. If replicated, treatments may consider decreasing the reward value placed on intense exercise and increasing value placed on low-intensity or non-exercise activities.