Objective: This propensity score–matched analysis from MSBase compared the effectiveness of cladribine with interferon β, fingolimod or natalizumab. Methods: We identified all patients with relapse-onset multiple sclerosis, exposure to the study therapies and ⩾1-year on-treatment follow-up from MSBase. Three pairwise propensity score–matched analyses compared treatment outcomes over 1 year. The outcomes were hazards of first relapse, disability accumulation and disability improvement events. Sensitivity analyses were completed. Results: The cohorts consisted of 37 (cladribine), 1940 (interferon), 1892 (fingolimod) and 1410 patients (natalizumab). The probability of experiencing a relapse on cladribine was lower than on interferon (p = 0.05), similar to fingolimod (p = 0.31) and higher than on natalizumab (p = 0.042). The probability of disability accumulation on cladribine was similar to interferon (p = 0.37) and fingolimod (p = 0.089) but greater than natalizumab (p = 0.021). The probability of disability improvement was higher on cladribine than interferon (p = 0.00017), fingolimod (p = 0.0025) or natalizumab (p = 0.00099). Sensitivity analyses largely confirmed the above results. Conclusion: Cladribine is an effective therapy for relapse-onset multiple sclerosis. Its effect on relapses is comparable to fingolimod and its effect on disability accrual is comparable to interferon β and fingolimod. Cladribine may potentially associate with superior recovery from disability relative to interferon, fingolimod and natalizumab.