Abstract
Abstract: :
Purpose: Persistency with glaucoma medication is important as a long–term outcome, and rates typically are low. This study assessed restart rates for the prostaglandin analog (PG) class by determining the percentage of patients who discontinued and then restarted therapy. Methods: We obtained pharmacy claims for 3 PGs (bimatoprost, latanoprost, travoprost) that were filled between 1999 and 2001; claims were submitted to several managed care plans in the US that contribute to the Protocare Database. These claims supported a retrospective cohort study that included 1) all patients who initiated a PG, and 2) the subgroup of patients who subsequently discontinued PG therapy. Patients were excluded if they filled any topical ocular hypotensive therapy in the 180 days before initiating a PG. We estimated the outcome of PG use 180 days after therapy initiation by combining survival analysis results from the 2 cohorts. Results: Of the 4356 patients initiating PG therapies, 2503 (57%) were included in the database (after 180 days were still plan members and had not switched ocular hypotensive therapies). In all, 477/2503 patients (19% [95% CI, 18% to 21%]) discontinued and then restarted their initial PG by day 180. An additional 40 patients discontinued and then restarted a different PG (2% [95% CI, 1% to 2%]). The remaining 1986 patients either persisted with their initial PG through 180 days (n=879; 35% [95% CI, 33% to 37%]), restarted a different class of topical therapy (n=280; 11% [95% CI, 10% to 12%]), or failed to restart any topical therapy (n=827; 33% [95% CI, 31% to 35%]). Conclusion: Previous studies showing low persistency rates for glaucoma therapy fail to evaluate restarts. Restart analyses are crucial for assessing long–term treatment of a chronic disease such as glaucoma. In general, persistency remains a significant challenge as 33% of patients discontinued and failed to restart any topical therapy.
Keywords: clinical (human) or epidemiologic studies: health care delivery/economics/manpower • clinical (human) or epidemiologic studies: outcomes/complications • clinical (human) or epidemiologic studies: prevalence/incidence