BACKGROUND: Multiple sclerosis (MS) frequently gives rise to depressive and anxiety symptoms, but these are often undertreated. This study investigated the effect of mindfulness-based cognitive therapy (MBCT) and cognitive rehabilitation therapy (CRT) on psychological outcomes and quality of life (QoL), and whether they mediate treatment effects on MS-related cognitive problems. METHODS: This randomized controlled trial included MS patients with cognitive complaints (n = 99) and compared MBCT (n = 32) and CRT (n = 32) to enhanced treatment as usual (n = 35). Baseline, post-treatment and 6-months follow-up assessments included patient-reported outcome measures (PROMS) and cognitive outcomes (self-reported and neuropsychological assessment). PROMS concerned psychological symptoms, well-being, QoL, and daily life function. Linear mixed models indicated intervention effects on PROMS and mediation effects of PROMS on cognitive outcomes. RESULTS: MBCT positively affected depressive symptoms (Cohen's d (d) = -0.46), fatigue (d = -0.39), brooding (d = -0.34), mindfulness skills (d = 0.49), and mental QoL (d = -0.73) at post-treatment. Effects on mindfulness skills remained significant 6 months later (d = 0.42). CRT positively affected depressive symptoms (d = -0.46), mindfulness skills (d = 0.37), and mental QoL (d = -0.45) at post-treatment, but not at 6-month follow-up. No effects on anxiety, well-being, self-compassion, physical QoL, and daily life function were found. Treatment effects on self-reported, but not objective, cognition were mediated by psychological symptoms and mindfulness skills. CONCLUSIONS: MBCT and CRT reduced a wide array of psychological symptoms and improved mental QoL. These improvements seemed to impact self-reported cognitive problems after both treatments, whereas objective cognitive improvements after MBCT seemed independent of improvement in psychological symptoms. Future studies should investigate long-term sustainability of these beneficial effects. TRIAL REGISTRATION: The trial was prospectively registered in the Dutch Trial registry on 31 May 2017 (NL6285; https://trialsearch.who.int/Trial2.aspx?TrialID=NTR6459 ).