Physical activity and depression: A multiple mediation analysis

Abstract

ObjectivesPhysical activity is associated with reduced symptoms among people with depression, but the factors that may mediate this relationship are poorly understood. We conducted multiple mediation analyses to assess whether positive affect (PA), negative affect (NA), physical activity self-efficacy, coping self-efficacy and exercise-induced feelings cross-sectionally mediated the association and the relative importance of each of these. We also examined whether leisure-time, non-leisure time or total physical activity were more strongly associated with depression. MethodParticipants (N = 164) experiencing depression or low mood completed a one-off postal questionnaire containing measures of physical activity, depression, the potential mediators and covariate variables. Data were analysed using correlations and multiple mediation analyses, controlling for the covariates. ResultsHigher levels of leisure-time and total, but not non-leisure time, physical activity were significantly associated with lower depression. Improvement in PA, pleasant feeling states, NA and levels of physical exhaustion significantly mediated the association between leisure-time and total, but not non-leisure time, physical activity and depression. Post-hoc analyses showed that improvements in physical activity self-efficacy mediated the leisure-time physical activity and depression relationship through improved PA. Coping self-efficacy was not a statistically significant mediator. ConclusionsLeisure-time physical activity may be more beneficial for depression than non-leisure time physical activity, as it increases PA and pleasant feelings and reduces NA and physical exhaustion. PA responses may be partly dependent on improvement in physical activity self-efficacy. People's psychosocial experiences of physical activity may be more important predictors of their depression response than total energy expenditure.