For multiple sclerosis (MS) patients taking disease-modifying therapies (DMTs), adherence to treatment is a key component of achieving beneficial outcomes, such as delayed disease progression and the reduction and prevention of symptoms and relapses. The aim of this study was to assess the impact of a claims-based measure of MS disease severity on DMT adherence in a one-year study period. Patients were identified from Humana Medicare Advantage claims data from January 1, 2013 to December 31, 2015. Patients over the age of 18 with at least 12 months of continuous enrollment and > 1 outpatient MS visit with DMT use prior to the index date were included. Patients who switched DMT type (oral, platform, IV) during the study period were excluded. Medication possession ratios (MPR) for DMTs were calculated from pharmacy and medical claims over 12 months of claims data, and a previously developed claims algorithm was used to determine MS disease severity. Patients with MPRs of 0.8 or higher were considered adherent to DMT treatment. Multivariable logistic regression was used to evaluate the association of MS disease severity, gender, DMT type, and age category with DMT adherence. The study population of 3,347 patients had an average MPR of 84.7 (75% were classified as adherent). Multivariable logistic analysis demonstrated that compared to the 18-45 age group, the 46-64 and 65+ age groups were 1.33 (OR: 1.33 [95% CI 1.08-1.64]) and 1.55 (OR: 1.55, [95% CI 1.18-2.05]) times more likely to be adherent. Patients with a high level of MS disease severity were 53% (OR: 0.47, [95% CI 0.36-0.62]) less likely to be adherent compared to those with low MS disease severity. No significant difference was identified for gender or DMT type (oral, platform, or IV). Increased age and lower MS disease severity were associated with better DMT adherence. MS disease severity should be considered when assessing risk for low DMT adherence. No funding supported this project. The authors have nothing to disclose. Preliminary results were previously presented virtually at AMCP Annual 2020 in April 2020.