Patient persistency with ocular prostaglandin therapy: a population-based, retrospective study

Open-angle glaucoma affects an estimated 33 million individuals worldwide. An intraocular pressure >21 mm Hg in individuals with no evidence of optic nerve damage is termed ocular hypertension, a risk factor for glaucoma that has been estimated to affect as many as 10% of individuals 40 years of age or older.

The purpose of this research was to assess persistency (time on therapy) with prostaglandin analogues in the treatment of glaucoma or ocular hypertension.

This population-based, retrospective cohort study used the Protocare Sciences managed care database, which includes prescription and medical claims data from multiple managed care organizations. Patients 20 years of age or older who initiated therapy with latanoprost, bimatoprost, or travoprost (index drugs) between April 2001 and June 2002 were included. Patients were required to be continuously enrolled in the same plan for the 180 days preceding index prescription fill. Follow-up continued through June 30, 2002. Two outcome measures were analyzed: (1) discontinuation of the index prostaglandin and (2) either discontinuation or change in the index prostaglandin regimen. Changing therapy was defined as switching to or adding another ocular hypotensive agent. Cox regression models were used to compare rate ratios of discontinuation and discontinuation/change. Patient data were censored on termination of insurance coverage or at the end of the study period.

Overall, 7527 patients were prescribed a prostaglandin analogue; 4356 patients met the inclusion criteria (n = 2376, 993, and 987 for latanoprost, bimatoprost, and travoprost, respectively). A total of 58% of patients were women, and 74% were 65 years of age or older. Compared with latanoprost, those treated with bimatoprost were 38% more likely to discontinue and 31% more likely to discontinue/change therapy, and patients treated with travoprost were 36% more likely to discontinue and 29% more likely to discontinue/change therapy (P < 0.001 for each comparison).

Latanoprost-treated patients demonstrated significantly (P < 0.001) greater persistency than did those treated with either bimatoprost or travoprost.