Association of Open-Angle Glaucoma With Dementia in California Medicare Beneficiaries
Nikolaos Topouzis,Ken Kitayama,Allan Puran,Fei Yu,Victoria L Tseng,Anne L Coleman
DOI: https://doi.org/10.1016/j.ajo.2024.07.009
2024-07-18
Abstract:Purpose: To examine the associations between open-angle glaucoma (OAG) subtypes and dementia in 2019 California Medicare beneficiaries. Design: Retrospective cross-sectional study. Methods: OAG diagnosis was determined by the International Classification of Diseases, Tenth Revision (ICD-10), diagnosis codes in part B claims, including the following OAG subtypes: primary open-angle glaucoma (POAG), normal tension glaucoma (NTG), pseudoexfoliative glaucoma (PXG), and pigmentary glaucoma (PG). Diagnoses of any dementia, Alzheimer dementia (AD), frontotemporal dementia (FTD), Lewy body dementia (LBD), and vascular dementia (VD) were identified by ICD-10 diagnosis codes. Covariates included demographics, systemic diseases, depression, hearing loss, obesity, smoking and alcohol-related disorders, and long-term aspirin, anticoagulant, and antithrombotic or antiplatelet use. Univariate and multivariable logistic regression models were used to assess the associations between OAG and dementia, adjusting for all covariates. Age-stratified analysis was also performed for beneficiaries aged 65-74 years, 75-84 years, and ≥85 years. Results: Among the 2,431,150 California Medicare beneficiaries included in this study, 104,873 (4.3%) had POAG, 9199 (0.4%) had NTG, 4045 (0.2%) had PXG, and 1267 (0.05%) had PG. The overall prevalence of any dementia was 3.2% (n = 79,009). In adjusted analyses, the odds of any dementia were lower for beneficiaries with all OAG subtypes compared with beneficiaries without glaucoma (odds ratio [OR] = 0.74 for POAG, OR = 0.74 for PXG, OR = 0.60 for NTG, and OR = 0.38 for PG; P < .01). In age-stratified analyses, beneficiaries with PXG had greater odds of VD (OR 2.84, P = .006; aOR 2.18, P = .04) in the youngest age stratum (65-74 years) compared to patients with no glaucoma. The odds for any dementia were lower for beneficiaries with all OAG subtypes compared to beneficiaries without glaucoma in the oldest, but not in the youngest, age stratum. Conclusions: In the 2019 California Medicare population, PXG is associated with an increased likelihood of VD in beneficiaries 65-74 years old, whereas other subtypes of POAG are associated with a decreased likelihood of any dementia. These findings may suggest selection bias because older adults who continue to follow up with glaucoma care may be more cognitively intact. Further studies are needed to better understand the complex relationship between glaucoma, dementia, and their subtypes.