Preschool vision study phase 1 complete

One goal of the Vision In Preschoolers Study is to identify the best screening tests for detecting childhood vision disorders.

A note from the editors: To help the general ophthalmologist prepare for getting students ready to return to school in the fall, Ocular Surgery News has compiled this new Back-to-School feature. In it you will find information on vision screening in preschool-aged children; information about managing conjunctivitis; and how to build a better dispensary. We hope to make this an annual feature in Ocular Surgery News.

Phase 1 of the three-part Vision In Preschoolers Study showed variations in performance among the most commonly used vision screening tests.

The study is being performed in part because of disagreements at the state and federal government levels over whether screening tests or comprehensive eye examinations are needed at the preschool level to identify eye disorders.

Currently, many children are screened in preschool by nurses or volunteers. Variations in screening methods and high levels of false positives have prompted pediatric health providers to conduct the Vision In Preschoolers (VIP) Study.

Many tests are used nationwide, said Graham E. Quinn, MD, of the VIP executive committee.

“Lots of different screening procedures are being used in different parts of the country. Some states mandated screenings but didn’t specify what type; others didn’t even mandate them. People just wanted to bring some rigor to it,” Dr. Quinn told Ocular Surgery News.

Phase 1 of the VIP Study evaluated commonly used screening tests, said Paulette P. Schmidt, OD, the lead investigator and chairwoman of the VIP Study. She began developing the concept for the study with an interdisciplinary group of collaborators in 1994.

“In phase 1, we looked at how test performance compared in a very controlled environment when optometrists and ophthalmologists experienced in caring for children conducted screening tests. We saw that even though the tests were on equal footing — the same kids, environment, testers – the tests varied.”

As the ophthalmic community digests the VIP part 1 results, debate continues over whether screening is effective enough to detect childhood eye problems. Many vision advocates believe that screenings should be replaced by comprehensive eye examinations. Others say the current screening process needs reform, but that comprehensive eye examinations would be a waste of available funds.

Organizations and experts are already scrutinizing the phase 1 results. The Vision Council of America referred to the study as evidence that all children should receive a comprehensive eye examination. Others have said it is a step in the right direction that fails in some areas.

Methods

The VIP Study evaluated 2,588 children age 3 to 5 years in Head Start programs in five states during two time periods, October 2001 to June 2002 (year 1) and October 2002 to June 2003 (year 2).

The study evaluated the screenings based on how well they identified four targeted conditions (amblyopia, strabismus, significant refractive error and unexplained reduced visual acuity) according to severity. Researchers compared sensitivity and specificity to detect target conditions in three levels of condition severity; group 1 conditions were considered to need early intervention. Group 2 conditions were important to detect early but were not as urgent as group 1. Group 3 conditions were considered the least urgent.

“Overall sensitivity was calculated as the proportion of children who failed the screening test who actually had a targeted condition. Specificity means that if the child did not have a condition, we called that a pass,” Dr. Quinn said.

Because some of the test machines had set sensitivity criteria of 94%, researchers used that as a standard for comparison. The researchers calculated sensitivity at 90% on screening tests that provided results that allowed such an adjustment, Dr. Quinn said.

“We thought 94% was high and that 90% was more reasonable. We used that as our yardstick for most of the comparisons. In one of the tables, we did everything at 94% so we could directly compare. But when you make specificity go up that high, sensitivity will go down. There are a lot more kids being examined when sensitivity goes down,” he said.

Of the 11 screening tests evaluated, noncycloplegic retinoscopy, the Retinomax Autorefractor, the SureSight Vision Screener and the Lea Symbols Distance Visual Acuity test were the most accurate for screening children at highest risk of developing the targeted eye conditions.

Overall, 90% of the children with group 1 conditions were accurately identified and referred for comprehensive eye examinations. Ten percent of children with no conditions were referred, a percentage that was to be expected considering the high specificity rating, Dr. Quinn said.

The study findings showed that some traditional screening tests, such as the MTI Photoscreener, did not perform as well as other screeners.

“The three tests that actually targeted refraction error – noncycloplegic retinoscopy, Retinomax Autorefractor and the SureSight Vision Screener — performed better than two photoscreeners, and those two photoscreeners performed very similarly,” Dr. Schmidt said.

Pediatric Eye Exam

Timeline according to joint guidelines from the American Association for Pediatric Ophthalmology and Strabismus, the American Academy of Ophthalmology and the American Academy of Pediatrics. Birth to 1 year Evaluation of vision is recommended in the first 3 months of life and again between 6 months and 1 year. Exams at the pediatrician’s office should include these evaluations as age-appropriate: Visual fixation “fix and follow” assessment External eye and eyelid assessment Ocular motility assessment Pupil examination Red reflex examination Noncompliant children; children with a family history of strabismus, amblyopia or other eye disease; and those with suspected ocular abnormalities should be referred to an ophthalmologist for a comprehensive eye examination. 1 to 3 years Yearly exams in the pediatrician’s office should include ophthalmoscopy when possible. In preverbal children, Teller acuity cards and picture tests are useful. As the child becomes verbal, visual acuity assessment is necessary. Children over age 4 may read the Snellen chart. Noncompliant children; children with a family history of strabismus, amblyopia or other eye disease; and those with suspected ocular abnormalities should be referred to an ophthalmologist for a comprehensive exam. 3 to 5 years First comprehensive ophthalmic eye exam in the ophthalmologist’s office should be carried out between ages 3 and 5 years. Relevant family history regarding eye disease and spectacle use in parents or siblings is required. Parental observations are important. Questions to ask: Does your child seem to see well? Does your child hold objects close to his or her face when trying to focus? Do your child’s eyes appear straight, or do they seem to cross or drift or seem lazy? Do your child’s eyes appear unusual? Do your child’s eyelids droop, or does one eyelid tend to close? Have your child’s eyes ever been injured? Children in whom eye abnormalities or refractive errors are detected should be managed appropriately. Initial comprehensive eye exam to adulthood Depending on the child’s health and family history, he or she should undergo comprehensive eye exams every 1 to 2 years. Children may undergo additional photoscreening examinations while in the classroom. Children diagnosed with refractive errors during this time should undergo visual acuity testing every 6 to 12 months (depending on the severity of their refractive errors) after initial spectacle prescription.

Sensitivity

According to the study, the most accurate screeners were the Lea Symbols Visual Acuity test, the HOTV Visual Acuity test, noncycloplegic retinoscopy, Stereo Smile II, Power Refractor II, the SureSight Vision Screener and the Retinomax Autorefractor.

In year 1, noncycloplegic retinoscopy had 90% sensitivity, the highest for screening children in group 1. The Retinomax Autorefractor had 87% sensitivity, the Lea Symbols Visual Acuity test had 77% sensitivity and the HOTV VA had 72% sensitivity.

In year 2, the Retinomax Autorefractor had 88% sensitivity for group 1, the SureSight Vision Screener had 81% sensitivity, and the Stereo Smile II and Power Refractor II had 72% sensitivity.

When screeners were analyzed further, researchers determined that for detection of amblyopia and refractive error, noncycloplegic retinoscopy, the Retinomax and the SureSight had the best results. For strabismus, HOTV VA, the Retinomax and the Stereo Smile II tests were considered best.

Overall, the study failed to detect one-third of children with targeted eye conditions.

“That means that one out of every three kids with all of the targeted conditions were not picked up. However, the group 1 conditions were the most severe and really the ones you want to pick up and treat. We got into the 85% to 90% sensitivity, meaning very few kids are missed,” Dr. Quinn said.

Dr. Schmidt said the VIP study group is currently deliberating sensitivity level, the issue of an acceptable number of children that can be missed.

Researchers pointed out that the tests in phase 1 were performed by specially trained eye professionals, and the extent of their training may not be frequently replicated. The researchers also said there was an overrepresentation of eye disorders among the study population.

Mixed reviews

Because the study data could affect state and federal legislation concerning spending on screenings and examinations, many experts are following the results.

H. Jay Wisnicki, MD, a pediatric ophthalmologist in New York, said the study is a step in the right direction but falls short of meeting expectations. He said flaws in the study’s design skewed results. For example, he said, one of the more traditional and well-received screeners, the MTI Photoscreener, did not fare well in the study although it was shown to work well in previous studies.

“One of the concerns [researchers] had was with photoscreening, which can also be used in younger kids. Some of the other tests require verbal responses, which you can’t do in preverbal children, when it’s most important to screen them,” Dr. Wisnicki said.

Next phases

Dr. Quinn estimated that between 14% and 20% of 3 and 4 year olds have some kind of vision screening.

“Only about 10% or 12% of preschool children have an eye exam. So there’s a huge unmet need of about 3 or 4 million children coming through the system every year, with only a small percentage getting eye evaluations. … We’re trying to figure out how to care for these kids,” he said.

Phase 2 of the VIP study is further evaluating the Lea Visual Acuity test, Retinomax Autorefractor, the Stereo Smile II and the SureSight Vision Screener when conducted by pediatric nurses and lay people who most commonly conduct screening tests, Dr. Schmidt said. Phase 3 will evaluate test and tester performance in a broader population of preschool children and include analysis of the cost of combinations of tests and testers that performed well in the earlier phases, she said.

For Your Information:

• Graham E. Quinn, MD, can be reached at 1st Floor, Wood Center, The Children’s Hospital of Philadelphia, Philadelphia, PA 19104; 215-590-2791; fax: 267-426-5015.
• Paulette P. Schmidt, OD, MS, can be reached at The Ohio State University College of Optometry, 320 W. 10th Ave., Columbus, OH 43210; 614-292-3189; fax: 614-247-6907; e-mail: pschmidt@optometry.osu.edu.
• H. Jay Wisnicki, MD, can be reached at Beth Israel Medical Center, Ophthalmology Department, 10 Union Square East, Suite 3B, New York, NY 10003; 212-844-2020; fax: 212-844-8221.

Reference:

• The Vision In Preschoolers Study Group. Comparison of Preschool Vision Screening Tests as Administered by Licensed Eye Care Professionals in the Vision in Preschoolers Study. Ophthalmology.2004;111:637-650.