Patterns of treatment for restless legs syndrome in primary care in the united kingdom

Background: The dopamine agonists pramipexole and ropinirole are licensed for the treatment of moderate to severe idiopathic restless legs syndrome (RLS) in Europe and the United States. In addition, various drugs that are not approved for this indication have been used for symptomatic treatment of RLS, including analgesics, quinine, and anxiolytics. Objective: The purpose of this analysis was to describe patterns of treatment of newly diagnosed RLS, including treatment effectiveness and resource utilization, in primary care in the United Kingdom. Methods: This was a cohort study that employed the UK General Practice Research Database. Two cohorts were assembled, one consisting of patients with an initial diagnosis of RLS between 1990 and 2004 and the other consisting of patients without RLS matched to cases in a 10:1 ratio by general practice, year of birth, sex, and registration with the practice on the case index date (date of the RLS diagnosis). The frequency of RLS-specific symptoms was estimated based on records of prescriptions for sleep medications and antidepressants, and reported cramps and leg problems other than RLS. Rates of resource use in the 2 years before and after the index date were estimated for both cohorts based on the numbers of total prescriptions, referrals to secondary care, and laboratory tests. Results: The RLS cohort comprised 8621 patients and the matched cohort 85,087 patients. Age and sex distributions were comparable between groups. The annual frequency of prescriptions for sleep medications in the RLS cohort increased significantly from 19.8% and 21.6% in the 2 years before the diagnosis of RLS to 27.4% in the first year after the diagnosis and 25.2% in the second year (all comparisons, P < 0.001). A similar pattern was observed for antidepressants in the RLS cohort (23.5%, 26.7%, 31.0%, and 29.6%, respectively; P < 0.001). Prescription rates were nearly constant in the matched cohort. The frequency of cramps and other leg problems was highest in the year before the diagnosis of RLS, declined in the first year after the diagnosis, and increased thereafter. Compared with the matched cohort, rates of prescriptions, referrals, and laboratory tests ranged from 49.7% to 59.0% higher in the RLS cohort in the 2 years before the diagnosis of RLS and from 63.4% to 91.4% higher in the 2 years after the diagnosis (all comparisons, P < 0.001). Conclusion: In patients with newly diagnosed RLS in this UK primary care cohort, use of medications that are not approved for the treatment of RLS was not associated with a reduction in clinical symptoms or health care resource utilization between 1990 and 2004.