Studies Estimate Prevalence of Vision Disorders in Children

Yael Waknine

October 10, 2011

October 10, 2011 — About 4% of preschoolers have myopia, 21% have hyperopia, and 10% have astigmatism, according to data from 2 studies supported by the National Institutes of Health and published in the October issue of Ophthalmology. Risk factors such as age, exposure to smoking, ethnicity, and access to health insurance were also identified.

Refractive errors such as myopia, hyperopia, and astigmatism are the most common vision problems in children and are correctable with eyeglasses, according to the articles, the first of which was led by Mark S. Borchert, MD, from the Doheny Eye Institute and the Department of Ophthalmology, Keck School of Medicine, University of Southern California, Los Angeles, and the Division of Ophthalmology, Children's Hospital Los Angeles, and the second of which was led by Roberta McKean-Cowdin, PhD, also from the Doheny Eye Institute. Identification and appropriate treatment of these conditions can prevent vision loss and the development of amblyopia (prevalence, 5%) and strabismus (prevalence, 2%).

"This is the first population-based study of preschool children in the United States to assess the burden of eye disease in this vulnerable population. It is also the largest study of its kind ever conducted worldwide," Rohit Varma, MD, MPH, a principal investigator, told Medscape Medical News in an interview. "Pediatricians and pediatric eye care providers should consider these data when identifying which children to follow more carefully and which children should be treated."

Dr. Varma is professor of ophthalmology and preventive medicine and director of the Glaucoma Service and Ocular Epidemiology Center at the University of Southern California's Keck School of Medicine.

The cross-sectional studies enrolled a total of 9970 children aged 6 months to 6 years from Los Angeles County in the Multi-Ethnic Pediatric Eye Disease Study (MEPEDS), and from Baltimore, Maryland, in the Baltimore Pediatric Eye Disease Study (BPEDS).

"These are monumental studies, very expensive, and carefully done," Robert W. Arnold, MD, told Medscape Medical News in an interview, noting the increased population size and better specific levels for black and Hispanic children relative to earlier MEPEDS and BPEDS trials.

Dr. Arnold is a pediatric ophthalmologist in private practice in Anchorage, Alaska, and also serves as a spokesperson for the American Association for Pediatric Ophthalmology and Strabismus. He did not participate in the studies.

Investigators found that astigmatism and myopia, but not hyperopia, may resolve with age. Infants aged 6 months to 1 year were about 3 times as likely to have astigmatism compared with children aged 5 to 6 years (23% vs 9%; 95% confidence interval [CI], 2.28 - 3.73). Likewise, those aged from 6 to 35 months were almost twice as likely to be myopic (spherical refractive error ≤ −1.00 diopters; 6% vs 3%; odds ratio [OR] ≥ 1.7).

In contrast, hyperopia (spherical refractive error ≥ 2 diopters) was more common in older children (those aged 5 to 6 years) compared with those aged 2 to 3 years (23% vs 18%; OR, 1.71; 95% CI, 1.19 - 2.47).

Myopic children were more than 4 times as likely to have astigmatism (cylindrical refractive error ≥ 1.5 diopters; OR, 4.6; 95% CI, 3.56 - 5.96) than those without refractive error, an effect that was less pronounced in children who were hyperopic (OR, 1.6; 95% CI, 1.39 - 1.94). Gestational exposure to tobacco significantly increased the risk for hyperopia and astigmatism (OR, 1.4 and 1.46, respectively), as well as strabismus.

Ethnicity also proved to be linked to the prevalence of refractive errors: myopia was more common among black children compared with Hispanic and non-Hispanic white children (6% vs 3% and 1%, respectively; OR, 6.0 and 3.2, respectively).

However, black children were the least likely to have hyperopia (17% vs 23% and 25%, respectively). Lack of health insurance was found to be a stronger factor, tripling the risk for amblyopia (OR, 1.5; 95% CI, 1.12 - 1.69). According to the study authors, the independent association between having hyperopia and health insurance is unlikely to be related directly but may represent a surrogate measure for other risk factors such as diet or prenatal care.

"MEPEDS and BPEDS show that the amblyopia risk factors are similar between ethnicities and over the preschool range, and that Medicaid patients have more risk," Dr. Arnold said. "Maternal smoking, or answering positive to a question about maternal smoking, increases risk as well."

Although these data provide some insights into factors associated with refractive errors, caution must be exercised because of the cross-sectional study design.

"These data...need to be validated in longitudinal studies," Dr. Varma said. "Also, treatment trials need to be conducted to identify which treatments will be effective."

Another perspective involves applicability of the data in a practice that involves children with different ethnicities, insurance, and variable prenatal exposure to smoking. Newly available billing for pediatric photoscreening (code 99174) requires appropriate selection of referral criteria based on amblyopia-related factors to ensure diagnosis validity and device cost-effectiveness, Dr. Arnold explained.

Dr. Arnold notes that the studies are also limited, in that hyperopia and astigmatism are not defined according to the American Association for Pediatric Ophthalmology and Strabismus guidelines that were established in 2003 to make reporting and development of vision screening uniform. In addition, the former National Institutes of Health–supported Vision in Preschoolers Study used different cutoffs still.

"What is missing is reporting with previously published cutoffs. I am concerned that if MEPEDS and BPEDS really wanted to help industry improve vision screening technology, they would report using American Association for Pediatric Ophthalmology and Strabismus guideline cutoffs in addition to their defined cutoffs," Dr. Arnold said.

Ophthalmology. 2011;118:1966-1973, 1974-1981. Borchert abstract, McKean-Cowdin abstract