Abstract
Purpose :
The prevalence of myopia has increased recently in the U.S. and other countries. An ensuing increase in the prevalence of pathological consequences of myopia (PCM) is a concern. We examined the National Health and Nutrition Examination Survey (NHANES) data from 2005-2008 to evaluate its potential to provide population prevalence estimates of possible PCM.
Methods :
The NHANES is a nationally representative sample of the U.S. civilian population. In NHANES 2005-2008, digital retinal images were obtained for participants aged 40+ (45°, Canon Non-Mydriatic Retinal Camera CR6-45NM) and graded at the University of Wisconsin for diabetic retinopathy, age-related macular degeneration (AMD), and other retinal findings. We created variables based on the detailed grading to attempt to capture possible PCM. The most specific was based on the code "myopic degeneration" for the AMD exclusion variable. The most broad definition also included retinal abnormalities plus high myopia (≤-6D) OR retinal detachment or choroidal abnormalities not attributed to AMD or other ocular conditions. NHANES weights were used to account for the probability sampling scheme of the NHANES. SAS proc Surveyfreq was used to obtain stratified estimates. Analyses are reported at the person level.
Results :
A total of 5359 participants aged 40+ had gradable retinal images for possible PCM (34.1% 40-49, 29.6% 50-59, 19.0% 60-69, and 17.3% 70+; 5.4% severe myopia (<-5D); 6.6% moderate myopia (≥-5, <-3); 27.6% low myopia (≥-3, <-0.5), 34.4% emmetropic (≥-0.5, ≤0.5), 23.3% moderately hyperopic (>0.5, <3.0), 2.6% moderate-to-high hyperopes (≥3.0D).
See Table below.
For the most inclusive definition, prevalence of possible PCM was highest in those with high myopia (3.64%) (moderate myopes, 0.62%; low myopes, 0.82%; emmetropes, 0.32%; low hyperopes, 0.24%; moderate-to-high hyperopes, 0.52%). Prevalence of possible PCM was higher in older age groups (age 40-49: 0.64%; age 50-59: 0.56%; age 60-69: 0.82%; age 70+: 0.81%).
Conclusions :
Prevalence of possible PCM in NHANES 2005-2008 varied from 0.07% to 0.68%, depending on the definition used. Our higher prevalence estimates are similar to other published rates (Wong et a, 2014 AJO), but the validity of our definitions is unknown. Because the purpose of the NHANES retinal image grading was not focused on assessing PCM, we advise caution in using NHANES 2005-2008 data for this purpose.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.