In patients with chronic disorders, control self-efficacy is the confidence with managing symptoms and coping with the demands of illness. Can do treatment (CDT) is an intensive, 3-day, social cognitive theory-based, multidisciplinary treatment that focuses on identification of stressors, goal setting, exploration of boundaries, and establishment of new boundaries. An uncontrolled study showed that patients with relapsing remitting multiple sclerosis (RRMS) and low-disability had improved control self-efficacy six months after CDT. Hence, in a 6-month, single-centre, randomized (1:1), unmasked, controlled trial in RRMS patients with Expanded Disability Status Scale (EDSS) score ≤4.0, we compared CDT with no intervention and the option to receive CDT after completion of study participation. Follow-up assessments were at one, three and six months. Primary endpoint was control self-efficacy (Multiple Sclerosis Self-Efficacy Scale Control [MSSES-C] (minimum 90, maximum 900) at six months. Secondary endpoints were functional self-efficacy (MSSES-F), participation and autonomy (Impact on Participation and Autonomy questionnaire [IPA]), health-related quality of life (MS Quality of Life-54 Items questionnaire [MSQoL-54]), anxiety, depression (Hospital Anxiety and Depression Scale [HADS]) and coping skills (Utrecht Coping List [UCL]) at six months. Tertiary endpoint was care-related strain on support partners (Caregiver Strain Index) at six months. Of the 158 patients that were included, 79 were assigned to CDT and 79 to the control group. Two CDT patients discontinued treatment prematurely. Sixty-one (77%) control patients chose to receive CDT after study participation. Intention-to-treat ANCOVA analyses were performed with follow-up values as dependent, and condition, baseline values, disease duration and gender as independent variables. The mean (standard deviation [SD]) MSSES-C score in the CDT group vs. control group at baseline was 468 (162) vs. 477 (136), and at six months 578 (166) vs. 540 (135) (p = 0.100). Secondary and tertiary endpoints did not differ between groups, except for the UCL palliative reaction score being slightly higher in the CDT group (p = 0.039). On post hoc analyses the MSSES-C score at one and three months was higher in the CDT vs. control group: 597 (114) vs. 491 (131) (p<0.0001) and 561 (160) vs. 514 (143) (p = 0.018), respectively; and at one month the MSSES-F, IPA Limitations, HADS Anxiety and Depression, and MSQoL-54 Mental and Physical scores were also in favour of the CDT group. We conclude that in low-disability RRMS patients, the intensive 3-day social cognitive theory-based CDT did not improve control self-efficacy at six months follow-up compared to waitlist controls. The absence of a between-group difference at six months relates to a gradual improvement in the control group. In all, this social cognitive theory-based approach for improving self-efficacy needs further investigation before being broadly applied in RRMS patients.