Vision screening in preschool children: benefits and harms still unclear
Evidence uncertain / Widespread implementation of further vision screening tests unjustified
Amblyopia is a decrease in vision in one or both eyes that is not correctable by vision aids or refractive surgery. It is currently unclear whether the frequency or severity of amblyopia decreases in the general population if all children under 6 years of age undergo a specific eye test (universal vision screening). On the one hand, the few studies available do not allow robust conclusions on the benefit of vision screening, while on the other, potential harms have hardly been investigated. IQWiG does not therefore consider it justifiable at present to implement an intensified screening programme for amblyopia or its risk factors in preschool children. This is the conclusion of the final report "Screening for visual impairment in children younger than 6 years” published by the Institute on 2 June 2008.
Vision is already tested in paediatric screening tests
In Germany, it is estimated that about 4 to 5% of preschool children are affected by amblyopia, depending on the threshold value specified for normal vision. This disorder usually originates in one eye, when images transmitted to the brain by the two eyes are different. For example, this may be the case in people with strabismus or major differences in refraction between the right and left eye. The brain "favours” one eye by suppressing images from the other eye. This may result in reduced visual acuity in the "non-favoured” eye, which cannot be corrected by vision aids such as spectacles or contact lenses.
For quite some time now health professionals have been discussing whether it would be meaningful for specially trained staff to perform a vision test in all children up to 6 years of age. This test would be implemented in Germany in addition to the legally embedded paediatric screening tests. The aim would be to detect amblyopia or its risk factors at an earlier stage, and through suitable treatment prevent, or at least reduce, its potentially lifelong consequences. The German Federal Joint Committee therefore commissioned IQWiG to assess whether current evidence shows a benefit from universal vision screening.
Only a few studies available - most show limited robustness of results
The authors of the IQWiG report conclude that the available studies provide neither evidence nor indications of a benefit from universal vision screening. On the one hand, only one study was available that directly compared the benefit of screening with no screening, while on the other, only a few studies were available that compared screening strategies of varying intensity or screening strategies conducted at different ages. In most cases, the study results were inconsistent and were not reliably interpretable due to methodological limitations.
Potential harms cannot be excluded
In addition, studies are lacking that would enable an estimate of potential harms (as with all medical procedures, screening programmes may also produce adverse effects). These may, for example, occur if a test produces false-positive results, i.e. if visual impairment is diagnosed even though vision is normal. The waiting period between the first false-positive result and the later correct diagnosis places an unnecessary burden upon affected children and parents. In the reverse case, harmful effects may also be produced if a false-negative test result leads to a false sense of security in affected families. In this case, visual impairment may be detected and treated too late, for example, if the child's visual problems are classified as irrelevant because of the "passed” vision test.
Danger of overtreatment
Moreover, the tests cannot detect with sufficient reliability those children who, despite actually having visual impairment, are not in need of treatment. In some cases, visual impairment may be reversible without treatment, and not cause permanent damage. This shortcoming of the tests and the resulting potential "overtreatment” seem more problematical the younger the children are at the time of screening. These children may suffer harmful effects if, for example, the unnecessary prescription of an eye patch leads to labelling affected children as "sick”, causing potential negative psychosocial consequences for them and their families.
IQWiG proposes supplementary measures
As the benefit of universal screening for amblyopia or its risk factors is uncertain, but potential harms of screening are possible, additional studies should first be conducted. They should either assess the benefit of additional screening programmes before their widespread implementation or accompany programmes implemented without evidence of a benefit.
Inspired by comments on the preliminary report, the Institute also developed proposals on how additional screening tests could be designed, also under the conditions of the uncertain evidence base. From IQWiG's point of view, two alternatives would be suited in principle for the implementation of a regionally restricted and temporary screening programme. In any case, the benefit of these programmes should be assessed in accompanying studies. These should be designed in such a way that not only the benefit for children (improvement of visual acuity) can be assessed, but also the accuracy of the applied tests with regard to the avoidance of overtreatment.
Procedure of report production
The preliminary results (preliminary report) were published by IQWiG on 29 October 2007. The deadline for comments was 26 November. Unclear aspects of the 11 comments received were discussed in an oral debate on 8 January 2008 in Cologne. The written comments as well as the meeting minutes of the debate were published in a separate document on the IQWiG website. After the debate, the preliminary report was revised and the final report sent to the contracting agency (Federal Joint Committee) on 2 April 2008.
The executive summary provides an overview of the background, methods, and further results of the final report.
Contact: Tel. ++49(0)221-35685-0, email@example.com