Regional variations and trends in the prevalence of diagnosed glaucoma in the Medicare population

Abstract

Purpose: To determine the prevalence of diagnosed glaucoma in the Medicare population and to assess regional variations and trends.

Design: Retrospective, cross-sectional study.

Participants: A 5% random sample of Medicare beneficiaries aged ≥ 65 years, excluding those in health maintenance organizations.

Methods: All claims with a glaucoma diagnosis code submitted by ophthalmologists, optometrists, or ambulatory surgery centers were used to estimate prevalence of the diagnosis of glaucoma for each year from 2002 to 2008. Regional variation in diagnosed glaucoma was examined in 9 large geographic regions and in 179 smaller subregions, controlling for patient characteristics and provider supply.

Main outcome measures: The prevalence of diagnosed open-angle glaucoma suspect (OAG-s), open-angle glaucoma (OAG), angle-closure glaucoma suspect (ACG-s), and angle-closure glaucoma (ACG), trends over time, and regional variations in prevalence.

Results: The overall prevalence increased from 10.4% in 2002 to 11.9% by 2008, largely owing to increase in diagnosed OAG-s (from 3.2% to 4.5%; P<0.001). The relative prevalence of diagnosed OAG compared with diagnosed ACG was 32:1. In 2008, multivariable models showed that the New England and Mid-Atlantic regions had 1.7 times more diagnosed OAG-s than the reference region (East South Central; New England: odds ratio [OR], 1.66; 95% confidence interval [CI], 1.58-1.75; Mid-Atlantic: OR, 1.66; 95% CI, 1.59-1.73). The odds of diagnosed OAG was 36% higher in New England (OR, 1.36; 95% CI, 1.30-1.42) and 31% higher in the Mid-Atlantic (OR, 1.31; 95% CI, 1.26-1.36) than in the reference region. The New England and Mid-Atlantic regions had the highest odds of diagnosed ACG-s and the Mid-Atlantic region had the highest odds of diagnosed ACG. Among 179 subregions, the New York area had high diagnosis rates of all glaucoma types.

Conclusions: The relative prevalence of diagnosed ACG compared with diagnosed OAG was lower than expected from population-based data, possibly owing to failure to perform gonioscopy. Substantial regional differences in diagnosed rates existed for all types of glaucoma, even after adjusting for patient characteristics and provider concentration, suggesting possible overdiagnosis in some areas and/or underdiagnosis in other areas. Regionally higher diagnosis rates in the New York area deserve further study.