© Society of Behavioral Medicine 2019. All rights reserved. For permissions, please e-mail: email@example.com. BACKGROUND: People with multiple sclerosis (MS) often experience fatigue, which is aggravated by inactivity. Identifying mediators of changes in physical activity (PA) and fatigue self-management (FSM) behaviors could optimize future interventions that reduce the impact of MS fatigue. PURPOSE TO: examine the effects of telephone-delivered interventions on Social Cognitive Theory constructs and test whether these constructs mediated secondary outcomes of PA and FSM behaviors. METHODS: Participants with MS (n = 208; Mean age = 52.1; Female = 84.6%) were randomized into contact-control intervention (CC), PA-only intervention, and PA+FSM intervention. Step count (Actigraphy) and FSM behaviors as well as self-efficacy, outcome expectations, and goal setting for PA and FSM were measured at baseline, post-test (12 weeks), and follow-up (24 weeks). Path analyses using bias-corrected bootstrapped 95% confidence intervals (CI) determined whether constructs at post-test mediated behaviors at follow-up when adjusting for baseline measures. RESULTS: Path analysis indicated that PA-only (β = 0.50, p < .001) and PA+FSM interventions (β = 0.42, p < .010) had an effect on goal setting for PA, and that PA + FSM intervention had an effect on self-efficacy for FSM (β = 0.48, p = .011) and outcome expectations for FSM (β = 0.42, p = .029). Goal setting for PA at post-test mediated the effects of PA-only (β = 159.45, CI = 5.399, 371.996) and PA + FSM interventions (β = 133.17, CI = 3.104, 355.349) on step count at follow-up. Outcome expectations for FSM at post-test mediated the effects of PA + FSM intervention on FSM behaviors at follow-up (β = 0.02, CI = 0.001, 0.058). CONCLUSIONS: Goal setting for PA and outcome expectations for FSM may be important constructs to target in telephone-delivered interventions designed to reduce the impact of MS fatigue. TRIAL REGISTRATION: Clinicaltrials.gov (NCT01572714).