PURPOSE: To identify factors associated with undiagnosed open-angle glaucoma (OAG), primary open-angle glaucoma (POAG), and pseudoexfoliative glaucoma (PEXG) in an elderly population in Thessaloniki, Greece.DESIGN: Cross-sectional population-based study.METHODS: Randomly selected subjects > or =60 years (n = 2,554) participated in the study. Subjects were classified as having POAG or PEXG according to specific criteria and using a two-scale definition of glaucoma. Undiagnosed glaucoma was defined as absence of either prior diagnosis of glaucoma or ocular hypertension or prior medical treatment for glaucoma or prior glaucoma surgery. Logistic regression analyses were performed with age, gender, family history of glaucoma, history of cataract surgery, visual acuity, vertical cup-to-disk (C/D) ratio, intraocular pressure, Advanced Glaucoma Intervention Study visual field score, time since last eye doctor visit, and type of OAG as covariates.RESULTS: The prevalence of undiagnosed glaucoma was 57.1% (56/98) for POAG, significantly higher than the prevalence of 34.9% (15/43) for PEXG (P = .017). POAG patients presented three to four times increased risk to be undiagnosed compared with PEXG patients (P = .02 and P = .04, respectively). Patients with OAG who had not visited an ophthalmologist during the last year had six times an increased risk to be undiagnosed (P = .003). In POAG, smaller vertical C/D ratio was statistically significantly associated with an increased risk to be undiagnosed (P = .008).CONCLUSIONS: Lack of regular visits to an ophthalmologist was a major factor for undiagnosed OAG. POAG was more likely to be undiagnosed than PEXG. Since C/D ratio was associated with undiagnosed POAG, standardized protocols involving thorough evaluation of the optic disk are recommended for ophthalmologists.