0721 TWO TIMES GREATER RISK FOR NEW ONSET MENTAL DISORDERS AMONG PATIENTS WITH PRIMARY RESTLESS LEGS SYNDROME RECEIVING DE NOVO DOPAMINE AGONISTS: A LARGE-SCALE RETROSPECTIVE CLAIMS MATCHED COHORT ANALYSIS

Abstract Introduction: FDA-approved treatments for primary Restless Legs Syndrome (PRLS), a sensorimotor neurological disorder, include a calcium channel ɑ2-δ ligand (gabapentin enacarbil) and dopamine agonists (DAs; pramipexole, ropinirole, and rotigotine). Increased risk for mental disorders among patients receiving DAs for Parkinson’s disease is well documented. However, risk among patients with P-RLS, treated at lower indicated DA doses, remains unknown. We compared likelihood for new onset mental disorders between matched cohorts newly diagnosed with PRLS, naïve to DAs, without history of mental disorders, initiating versus not initiating DAs. Methods: Selected from 6.5-year (7/1/2008–12/31/2014) MarketScan Commercial and Medicare Supplemental claims databases were patients aged ≥18 years with ≥1 RLS claim and ≥2 years of data before and after initial (index) RLS diagnosis. Excluded were patients with ≥1 mental disorder diagnosis or ≥1 pharmacy fill for antidepressants, antipsychotics or DAs in the 2-year pre-index period and patients ever diagnosed with Parkinson’s, kidney disease, iron deficiency, or pregnancy. Identified patients were classified into 2 groups: Those receiving (DA+) versus not receiving (DA-) ≥1 DA fill in the 2-year post-index period. Each DA+ patient was matched to a DA- patient on sex, age at index diagnosis, region, employment and illness burden. Parallel follow-up periods were determined for each matched pair. McNemar’s test examined group differences in percent of patients receiving mental disorder diagnoses by validated mental disorder severity classifications. If significant (p≤0.05), logistic regression examined odds ratios (OR) and 95% confidence intervals (95%CI). Results: From 5,419 patients, 1,080 matched pairs were identified. Compared to DA- matched controls, DA+ patients were 2 times more likely (OR 2.0, 95%CI 1.54–2.59, p