OBJECTIVES: Medication possession ratio (MPR) and persistence are typically used as measures of patients’ drug use patterns.  However, these measures may not capture all aspects related to real world dosing.  We aim to explore additional approaches to enhance meaningful interpretation of patients' drug use patterns using persistence, adherence and average daily dose (ADD) data of dimethyl fumarate (DF) in a real world dataset. METHODS: A retrospective cohort study using the MarketScan Commercial and Medicare Databases (March 2013 – January 2014) was conducted in adult multiple sclerosis patients.  Patients with DF claims who were continuously enrolled for at least 9 months before and after starting therapy were included (n=2,879). Outcomes included time to treatment non-persistence (switch to another disease modifying therapy (DMT) or drug discontinuation of ≥ 30 days), adherence (MPR and percent of days covered (PDC)), and estimated ADD.  Data were interpreted in the context of the FDA approved daily dosing (240 mg b.i.d.), and an earlier trial (1) reporting dose-related MRI findings indicating non-significant effects on brain lesions for 360 mg/day and 120 mg/day doses. RESULTS: The mean (SD) DF treatment duration was 96 (66) days (median 83) and 24% of patients became non-persistent. Mean (SD) MPR and PDC were 0.83 (0.26) and 0.75 (0.29), respectively. ADD (SD) was 417 mg/day (221) with 15.3% treated at < 240 mg/day, 5.6% at 240-359 mg/day, 47.8% at 360-479 mg/day, and 31.3% at ≥ 480 mg/day (i.e., 20.9% treated at <360 mg/day and 68.7% treated at less than the labeled dose). CONCLUSIONS: The finding of a large proportion of patients receiving potentially sub-optimal treatment, as determined by ADD on DF, indicates significant potential clinical consequences for these patients. The approach used in this study could be used in other similar studies examining treatment adherence to enhance meaningful interpretation of adherence data. 1. Lancet 2008; 372(9618):1463-72.