Methodological quality of the review: Low confidence
Author: Kemper AR, Margolis PA, Downs SM, Bordley WC.
Region: Not specified
Sector: Visual screening
Sub-sector: Screening, vision, amblyopia, tests for detection
Equity focus: Pre-school children
Review type: Effectiveness review
Quantitative synthesis method: Narrative analysis
Qualitative synthesis methods: Not applicable
The prevalence of amblyopia in the USA is estimated at between 2% and 5%. Amblyopia is known to be the leading cause of monocular vision loss among adults between 20 and 70 years. It is therefore recommended that all children between three and four years should receive screening to detect amblyopia and strabismus. Despite this, many children do not receive vision screening. Two approaches are available to primary care providers to screen amblyopia: traditional vision screening and photorefractive screening. Despite the different screening tests available, it is uncertain which test performs best in the primary care setting.
To test characteristics and the quality of evidence on available screening tests for the detection of amblyopia in preschool-aged children and to support primary care practitioners when choosing a strategy for screening.
The authors identified 9,551 citations. Out of these, 33 potentially met the criteria for inclusion. Overall, only four studies met all inclusion criteria. The review of reference list of these articles did not identify other studies for inclusion.
Of the four included studies, one addressed traditional vision screening while three addressed photo screening. The review found that there is variation in the gold standard ophthalmologic examination. Authors suggest the evaluation of vision screening tests in primary care setting using a standardized ophthalmologic examination.
The authors included studies that compared the results of commercially available screening tests in pre-school aged children to ophthalmologic examination.
Authors conducted a search in Medline from inception 1966 until 1999. The search included a number of terms associated with vision tests, amblyopia, strabismus, refractive errors, visual acuity or cataract. The review was restricted studies in English, human studies, and to subjects over the age of 12 years. The searches were also supplemented by manual searching the bibliographies of included studies.
The authors suggested that two reviewers abstracted the included studies to check accuracy. The authors resolved all differences from their individual review by consensus.
None if the included studies took place in primary care settings using usual screening procedures which makes generalization of these articles to a primary care setting more difficult.
Geographic focus of the studies included in the review was not reported.
There is low confidence in the conclusions about the effects of this study. The review was based on literature searches of one database (Medline), covering appropriate search period. Although the searches were also supplemented by searching the bibliographies of included studies manually, grey literature and contacting experts for further potentially relevant studies were not part of the search strategy. Therefore, we cannot be confident that relevant studies were not omitted in the review. Review authors did not report on the methods used to extract data and did not report assessing the quality and risk of bias of included studies, as such it is not clear if findings of included studies were reliably reported impacting on the validity of the review.