ATS protocol useful in measuring decreased VA in children aged 48 to 71 months

Almost all children aged 48 to 71 months can be tested for decreased visual acuity (VA) using the Amblyopia Treatment Study (ATS) vision testing protocol, but the protocol is useful for this purpose in only two-thirds of those aged 30 to 47 months, said David S. Friedman, MD, PhD, MPH, in a session about pediatric eye diseases.

Dr. Friedman, of the ophthalmology and public health departments at Johns Hopkins University, Baltimore, reported some results of the Baltimore Pediatric Eye Disease Study in his presentation.

The rationale for this particular research, he said, was that no such previously published population-based studies of preschool children existed, there was a lack of an accurate estimate of the prevalence of VA loss, and the thought that the understanding of cases could help guide public health interventions.

African-American and white children aged 6 to 71 months from northeast Baltimore and east Baltimore County were eligible for the study. Researchers went door to door, visiting more than 59,000 homes to conduct short interviews; longer interviews and exams were done at clinics set up in the neighborhood from December 2003 to March 2007. Exams were performed at the homes of those who did not keep their appointments at the clinic; 1,268 (49.8%) of the children undergoing examination at home or in the clinic were African-American, and 1,030 (40.5%) were white. Participants were referred for further care as appropriate.

The exam included the ATS VA testing protocol for children aged 30 to 71 months. Children were tested with their glasses if they were wearing them, but this was a rare occurrence, Dr. Friedman said, adding that 1.8% of white children and 1.6% of African-American children had glasses.

For the purposes of their study, the researchers defined decreased VA as worse than 20/50 in children aged 30 to 47 months and worse than 20/40 in those aged 48 to 71 months.

"We defined decreased vision differently for different age ranges based on the fact that several previous reports had shown that you don't get perfect vision in younger children very often," he said.

The testing protocol included an initial test of presenting vision, a same-day retest after cycloplegia with refraction in subjects found to have decreased vision, and repeated testing on a separate day in children found to have decreased vision or found to be untestable.

Respondents and non-respondents were similar in terms of age, ethnicity, sex, parental report of eye health and general health, whether they ever had had an eye problem diagnosed, and whether they had had difficulty in seeing in the past year, Dr. Friedman said.

"However, there were differences, and those who attended the exam, either in the clinic or at a home visit, were more likely to have had some form of reported health problem at birth," he said. "Their caregivers had more education and were less likely to work outside the home."

Fixation preference testing was found to be poorly correlated with VA in the older children, Dr. Friedman said.

Overall, 97% of those aged 48 to 71 months and 67% of those aged 30 to 47 months were able to be tested using the ATS protocol, he said, adding that this finding mirrors that of some other studies of young children. The African-American children had a lower testability, adjusting for age and sex, as did boys aged 30 to 47 months compared with girls of the same age.

The only statistically significant racial difference in decreased vision, he said, was found to be same-day testing, which occurred in 0.5% of the white children and 1.6% of the African-American children.

Refractive error was the main cause of decreased vision and accounted for it in four of seven white children and six of 13 African-American children, but on retesting without glasses after cycloplegia, the vision of three of the African-American children was found to be normal.

Researchers also found that there was no difference between African-American children and white children in regards to rates of decreased VA on presentation and best measured vision, Dr. Friedman said. Also, half of the decreased VA was found to be due to refractive error.

"The prevalence of bilateral reduced vision on presentation was less than 2% in both whites and African-Americans," he said. "The prevalence of decreased vision may be higher in those untestable who had poor vision, so we may not be capturing this whole amount of decreased vision."

The overall prevalence of bilateral decreased vision in this age group, however, is low, Dr. Friedman concluded.