Because it is unclear whether prevalence rates of POAG from studies performed outside the United States are applicable to the United States, we selected, when possible, studies that were performed in the United States only and chose not to use the pooled prevalence rates from the previously published meta-analysis. The primary criterion for study selection was that the data be from the most recent prevalence studies in the United States. Studies were included if POAG was defined on the basis of an assessment of the optic nerve, visual fields, or both. Intraocular pressure could not be used to define POAG. The detailed information on different diagnostic criteria and reasons to include or exclude each U.S. population-based study are available in our recent report.
3 The studies used in this estimate include the Baltimore Eye Survey for non-Hispanic whites and African Americans, the Los Angeles Latino Eye Study (LALES) for Hispanics, and the Singapore Eye Study for Chinese Americans.
6 –8 The reasons to exclude other population-based studies in the United States were as follows: (1) The method for screening individuals did not include visual fields and so may have led to an underestimation of the prevalence of POAG (Framingham Eye Study)
9 ; (2) IOP was included in the diagnostic criteria for POAG (Beaver Dam Eye Study)
10 ; (3) the sample was restricted to those aged 73 years and older, and detailed criteria for optic nerve assessment such as vertical cup-to-disc ratio and symmetry for the diagnosis of POAG were not available (Salisbury Eye Evaluation)
11 ; and (4) the use of statistically based criteria to define glaucomatous optic nerve damage may not provide an accurate estimate of the prevalence of POAG in Hispanics (Proyecto VER).
12 Because there were no population-based data for Asian Americans in the United States, we included a recent population survey of Chinese residents in Singapore.
8 This study was selected for two reasons: First, most Asians in the United States are Chinese.
13 Second, we chose the Singapore study rather than recent prevalence studies in China because Americans and Singaporeans have similar socioeconomic characteristics, including similar income and life expectancy and have comparable governmental subsidized medical insurance.
14 For other minorities, including American Indians, Native Alaskans, and Native Hawaiians and other Pacific Islanders, we used an unweighted pooled prevalence rate, modeled by a meta-analysis conducted by the Eye Disease Prevalence Research Group (EDPRG).
2