Intended for healthcare professionals


The future of preschool vision screening services in Britain

BMJ 1997; 315 doi: (Published 15 November 1997) Cite this as: BMJ 1997;315:1247

We need better research on which to base policy decisions

  1. Jugnoo S Rahi (j.rahi{at}, Medical Research Council clinical training fellowa,
  2. Carol Dezateux, Senior lecturer in paediatric epidemiologya
  1. a Department of Epidemiology and Public Health, Institute of Child Health, London WC1N 1EH

    What should be done in Britain about existing and proposed programmes to screen preschool children for the related conditions of amblyopia, refractive error, and strabismus? Those who have visited this debate before may not be surprised that the recently published systematic review on preschool vision screening from the NHS Centre for Reviews and Dissemination identified serious deficiencies in the research evidence which informs current policy.1 However, the conclusions that “Providers currently offering screening programmes should consider discontinuing them” and, “From an ethical point of view, it is appropriate to continue to screen only in the context of a controlled trial of treatment” will undoubtedly prove more controversial.

    The available research evidence has been interpreted differently in similar reviews from North America,2 3where it has been concluded that preschool vision screening should continue. Why have reviewers reached different conclusions from essentially the same data? By integrating existing information in an unbiased manner, systematic reviews should provide data for rational decision making4 and stimulate focused debate about policy and future research needs. There are, however, inherent difficulties when reviews are based largely on observational rather than experimental data, as their interpretation is likely to be less straightforward.

    The main purpose of screening and surveillance for visual defects throughout childhood is the early detection and treatment of relevant ophthalmic disorders to minimise their impact on a developing child. As up to 6% of preschool children will have an ocular or vision defect requiring treatment or follow up,1 some form of preschool vision screening service has been offered in most health districts for the past 20 years.5 In addition, children currently receive a further examination at school entry. As the United Kingdom review has policy implications for modifying or discontinuing existing preschool services,1 it is important that its recommendations are viewed in a broader context.

    The key question for policymakers is whether a programme of preschool vision screening offers benefits over existing vision screening at the school entry examination (currently the subject of a separate review from the Centre for Reviews and Dissemination). This question probably cannot be answered definitively by reviewing systematically identified literature of the quality currently available. While ideally policy should be informed by evidence from a randomised controlled trial, further work would be required to clarify its purpose and design as well as the most appropriate measures of outcome. The role of other quantitative methods could be explored—for example, a decision analysis to assess the effectiveness, including cost effectiveness, of differing preventive strategies, which would help determine research priorities.6

    Perhaps the most challenging finding of the review for the ophthalmic community is the lack of evidence from trials to support the main treatments for amblyopia—namely, occlusion combined with spectacle correction as necessary.7 As the almost universal use of these treatments in clinical practice is derived from extensive clinical and basic scientific research on stimulus deprivation amblyopia,7 8prospective studies of the natural history of untreated amblyopia may be difficult. However, clinical uncertainty remains about when to start and how long to continue with occlusion therapy, how to modify treatment according to the response achieved, and finally how to monitor concordance with treatment.9

    All these issues should be addressed in any future treatment trials, which will depend critically on the involvement and collaboration of ophthalmic professionals. Another important issue is the age above which treatment of pre-existing unilateral amblyopia may not succeed. Although treatment is considered to be most effective in early childhood, there is some evidence from observational studies that improvements in acuity can occur, sometimes spontaneously, in some affected adults,10 and this merits further exploration.

    The review also identifies the need to understand better the long term functional consequences of untreated unilateral amblyopia, about which only limited information is available.11 One component of this is being investigated through a national collaborative study of the incidence, causes, and outcomes of loss of vision in the non-amblyopic eye of individuals with pre-existing unilateral amblyopia.12 However, there remains the important question of the disability attributable to amblyopia per se, which partly reflects the inherent difficulties of measuring and interpreting visual function in children.

    The findings of this review present us with the dilemma of what to do when there is little evidence, particularly of the right sort, specifically supporting the benefits of an existing service and none proving it is ineffective or detrimental. In the current climate there is a danger that existing, but incompletely researched, services may be discontinued prematurely. The challenge raised by this review is how to secure a sounder evidence base for policymakers which reflects “clinical reality and its inherent difficulties.”13 Implementing this research agenda will require a close partnership between those concerned with planning, providing, evaluating, and using these services.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.