Background Initiation of disease-modifying therapy early in the disease course of relapsing-remitting multiple sclerosis (RRMS) has demonstrated beneficial effects on clinical outcomes, but socioeconomic outcomes remain largely unexplored.Objective To investigate the association between the delay from disease onset to first treatment and the hazard of disability pension.Methods We performed a population-based cohort study with data from the nationwide Danish Multiple Sclerosis Registry and Danish nationwide registries. Patients with a disease onset between 1 January 1996 to 5 April 2016 were followed until disability pension or a competing risk/censoring event. 7859 patients were assessed for eligibility of which 5208 were included in the final cohort. Key inclusion criteria were: a diagnosis of multiple sclerosis, relapsing-remitting phenotype, treatment in history, age 18–65 years and an Expanded Disability Status Scale≤4. Patients were categorised according to time from onset to first treatment: within 1 year (early), between 1 and 4 years (intermediate) and from 4 to 8 years (late).Results Of the 5208 patients, 1922 were early, 2126 were intermediate and 1160 were late. Baseline clinical and socioeconomic variables were well balanced. The hazard of receiving disability pension increased with increasing delay of treatment initiation compared with the early group. Cox regression estimates adjusted for clinical and socioeconomic confounders: intermediate (HR, 1.37; 95% CI, 1.12 to 1.68) and late (HR, 1.97; 95% CI, 1.55 to 2.51).Conclusion Early treatment initiation is associated with a reduced risk of disability pension in patients with RRMS. This finding underlines the importance of early diagnosis and treatment on a patient-centred, socioeconomic disability milestone.