Methodological quality of the review: High confidence
Author: Mathers M, Keyes M, Wright M.
Region: United Kingdom (UK), United States of America (USA), Israel, Sweden, Australia, Canada, Japan and Austria.
Sector: Vision screening
Sub-sector: Screening programmes, children’s vision screening, effectiveness, visual acuity.
Equity focus: Children aged up to 16
Review type: Effectiveness review
Quantitative synthesis method: Narrative analysis
Qualitative synthesis methods: Not applicable
A ‘screen’ is a procedure that can be performed quickly to determine whether a health concern may exist. While screening can enable important health problems to be identified in the presymptomatic or early symptomatic period so that effective interventions can be offered, unnecessary screening can be costly for participants and society. The variance on the prevalence of common visual disorders is part due to variation in the age at which children are studied. Variance also reflects major differences in screening programs from which data are obtained. This variability in vision screening results from the lack of consistent and conclusive evidence no a number of factors related to vision screening.
To determine (1) the effectiveness of children’s vision screening programmes; (2) at what age children should attend vision screening; and (3) what form vision screening programmes should take to be the most effective.
A total of two randomized controlled trials (RCTs), 33 non-RCTs and 8 systematic reviews were included in the review. Largely because of the study designs used, most of the evidence was categorized as level III-3, with a lesser amount of evidence pertaining to level III-2. The majority of studies were low quality. The two RCTs identified were medium quality and evidence from systematic reviews were largely based on non-RCTs .
Evidence from this review supported children’s vision screening in the preschool period (three to five years of age), but not subsequently at school entry or in the later primary school years. Screening by orthoptists, or non-vision health professionals (such as nurses) with appropriate training and the option for secondary screening, was suggested by the evidence. Referral criteria and outcomes used in the studies reviewed did not necessarily reflect or measure the effects of the vision condition on functional or measure the effects of the vision condition on functional vision. Future studies should consider functional vision in order to appropriately evaluate the benefits of screening.
The authors noted that future research should encompass high-quality RCTs to determine whether vision screening leads to a substantial decrease in the presence of correctable visual acuity deficits. There is a need to examine the effectiveness of vision screening at school entry and rigorous trials are required to determine whether screening in the neo-natal period is indeed a necessity.
The search focused on identifying studies that examined the effectiveness of vision screening programmes for children aged from birth to 16 years old. Studies evaluating not only screening, but also screening personnel, referral pathways, treatment and consideration of outcomes were identified. The search was limited to studies in English only and studies published from 1990 onwards.
Studies eligible for inclusion were (1) systematic reviews; (2) RCTs (3) pseudo-randomized controlled trials; (4) non-RCTs (comparative studies with concurrent controls; and comparative studies without concurrent controls). Case series were excluded.
The authors conducted a search using the databases Medline, CINAHL, Embase and Cochrane Database of Systematic Reviews from 1990-2008. A manual search of reviews and contacting experts in the field was also conducted as part of the search strategy.
Two researchers independently assessed each abstract of potentially eligible papers, extracted data and rated their quality. The authors conducted a narrative synthesis of included studies, which seems appropriate due to the heterogeneity of included studies.
The authors noted that included studies used samples of limited generalizability or very small samples.
The authors did not restrict the search for literature to a specific income setting. Nevertheless, studies eligible for inclusion in the review were high-income countries including the UK, USA, Israel, Sweden, Australia, Canada, Japan and Austria. Due to the heterogeneity of included studies, findings might not be applied to low- and middle-income settings.
Overall, there is high confidence in the conclusions about the effects of this study. The authors conducted a thorough search of the literature to ensure that all relevant studies were included in the review. They used appropriate methods in terms of study selection, data extraction and quality assessment of included studies, minimizing risk of bias. The only limitation identified was not avoiding language bias when searching for the literature. Nevertheless, due to the limitations of studies included in the review, the authors did not draw strong policy conclusions and appropriately reported limitations of the review.