PND8 - Real-World Outcomes and Medication Adherence to Disease Modifying Therapies (DMTS) Among Patients with Multiple Sclerosis (MS)

Abstract

To assess the relationship between disease modifying therapy (DMT) adherence, clinical outcomes, and healthcare costs among patients with multiple sclerosis (MS). We conducted a retrospective cohort study using electronic health record data and an insurance claims database (2014-2016) from an integrated health system. We included patients with a diagnosis of MS who were >18 years of age, seen in a neurology clinic during a two-year period, and who had > 2 prescription insurance claims available. At each patient’s first and last neurology visit in the timeframe, DMT was defined by administration type (i.e. oral, injectable or infusion). Adherence to DMT was calculated using proportion of days covered (PDC) 12 months following the index prescription order, or until a 90-day discontinuation, whichever came first. MS relapses, hospital admissions, emergency room (ER) visits, and per member per month (PMPM) healthcare costs were compared among patients’ adherent (PDC>0.8) versus nonadherent (PDC<0.8). 197 MS patients met the inclusion criteria. Overall median PDC was 0.90 with 67% adherent and 33% non-adherent. Median PDC adherence level was highest for patients on injectables (0.92, N=122), orals (0.86, N=69) and infusions (0.84, N=6). Patients who were adherent had significantly lower risk of ER visits (13% vs. 27%, respectively, p=0.01), fewer hospitalizations (8% vs. 12%, p=0.30), fewer relapses (1% vs. 3%, p=0.26), and lower PMPM medical ($823 vs $1122, p=0.339), inpatient ($154 vs $288, p=0.491), and outpatient costs ($380 vs $456, p=0.451), but significantly higher pharmacy costs ($4828 vs $2850, p<0.001). Within a population of MS patients seen at a neurology clinic, adherence to DMTs was correlated with a decrease in ER visits and an increase in pharmacy costs. Hospital admissions, relapses, and non-pharmacy PMPM healthcare costs were lower in the adherence group, but did not reach statistical significance perhaps due to inadequate sample size.