PurposeAlthough all disease-modifying therapies (DMTs) reduce risk of relapse in multiple sclerosis (MS), many factors, including route of administration, influence selection of first-line DMT. Knowledge of real-world treatment patterns and effectiveness in reducing relapses across DMTs is important to understanding factors influencing this choice. This study sought to describe treatment patterns and relapses among newly treated adults with MS and by DMT route of administration (oral, injectable, and infusion). MethodsIBM MarketScan research databases were used to identify MS adults newly initiating DMTs (index event) from January 1, 2011–April 1, 2016, who had 12 months of continuous preindex and postindex medical and pharmacy benefits. Newly treated patients were those with ≥2 nondiagnostic claims with an International Classification of Diseases, Ninth Revision, Clinical Modification (340) or Tenth Revision, Clinical Modification (G35) code and no DMT prescription claims in the 12 months’ preindex. Persistence and adherence were measured from index until the earliest of ≥60 days without DMT, switching DMTs, or end of follow-up. Relapses were defined using a validated claims-based algorithm and measured in the 12-month preindex and postindex periods. Regression analysis adjusting for patient characteristics and prior relapses was used to determine the association between DMT route of administration and odds of 12-month persistence, odds of postindex relapse, and number of postindex relapses. FindingsOf 9378 newly treated MS patients meeting inclusion criteria; average age was 46.7 years, and 73.3% were female. Most patients initiated an injectable (65.5%) or oral (26.1%) DMT. Relapses decreased markedly from preindex to postindex (32.9%–24.0%), which was highest among oral users (35.8%–21.6%). Patients with no (vs ≥3) relapses preindex were more likely to be relapse free postindex (81.6% vs 31.4%). Nonpersistence (39.1% overall) was lowest among oral users (33.4%) and higher among those with versus without a postindex relapse (50.6% vs 35.5%). Patients initiating oral versus injectable agents were more likely to be persistent at 12 months (odds ratio [OR], 1.45; p < 0.0001) and less likely to relapse (OR, 0.75; p < 0.0001) postindex. Switches were uncommon (~10%) across cohorts. Preindex relapses were associated with increased odds of postindex relapses (OR, 1.73; p < 0.0001) but not with odds of persistence at 12 months. ImplicationsThe 12-month nonpersistence rate was high among all MS patients but lower among oral users. Oral users were also less likely to relapse postindex. Despite the effectiveness of DMTs in reducing relapses, the low persistence, lack of switching to a new DMT, and continued relapses highlight an unmet need in the MS treatment landscape.