Intended for healthcare professionals

Editorials

A national screening programme for diabetic retinopathy

BMJ 2001; 323 doi: https://doi.org/10.1136/bmj.323.7303.4 (Published 07 July 2001) Cite this as: BMJ 2001;323:4

Needs to learn the lessons of existing screening programmes

  1. Ulrich Freudenstein (base{at}joanduli.freeserve.co.uk), locum consultant in public health medicine,
  2. Julia Verne (jverne{at}doh.gov.uk), consultant in public health medicine
  1. Wiltshire Health Authority, Devizes SN10 5EQ
  2. NHS Executive, Bristol BS34 8SR

    Retinopathy is the biggest single cause of blindness in the United Kingdom.1 Laser coagulation of high risk lesions detected by screening can significantly reduce the likelihood of blindness and deteriorating vision. 2 3 Screening for diabetic retinopathy has been available in some areas of the United Kingdom since the late 1980s, but access is uneven, screening techniques of differing effectiveness have been used, 3 4 quality assurance may not be an integral part, and the resources available are variable. A national screening programme has now been recommended, but several organisational issues need to be tackled if this programme is not to repeat the problems incurred by earlier national screening programmes.

    In 1999 the UK National Screening Committee asked the British Diabetic Association (now Diabetes UK) to convene an advisory panel to produce a model for a cost effective national screening programme. The panel's recommendations are now published on the national screening committee's website (www.diabetic-retinopathy.screening.nhs.uk/index.html). The preferred method for screening is digital retinal photography. This technology has secondary advantages of easy storage and retrieval of images, which facilitates quality assurance, training, and patient education. However, a baseline assessment of the current position in one English health region has identified some of the issues to be addressed before a comprehensive risk reduction programme can be introduced.

    The South West region of England has a population of around 4.9 million. Its eight health districts vary in size from 0.5 to one million people. Two districts currently screen for retinopathy using retinal photography, one of them using digital images. Both screen most patients in a general practice setting. Two more districts have partial population coverage, one by a hospital based retinal photography programme and the other by optometrist based retinal biomicroscopy. In the other districts opticians and general practitioners provide screening on a case by case basis or through individual practice based recall programmes.

    An accurate list of diabetics is essential to planning a screening service. Both district wide screening programmes using retinal photography observed a rise of a third in the number of recorded diabetic patients after the introduction of a central recall register. This was probably due to previously incomplete practice registers. The effect of under-recording will need to be taken into account where district wide registers have not yet been established.

    Screening by retinal photography requires trained non-medical staff. There is presently no system for accreditation, so a national training infrastructure will be needed to provide accredited training. Screeners will also need a career structure, for example in quality control or programme management if the recruitment and retention problems experienced by the cervical cytology screening programme are not to be repeated.5 Additional roles for screeners might include communication of results to patients and patient education, because alerting patients to the development of retinal changes may help to improve their efforts at glycaemic and blood pressure control. This would, however, reduce the number of screening examinations per session. Extrapolating from a district programmes where screeners conduct little patient education, we estimate that 21 full time screeners would be needed for the whole region to deliver a retinal photography screening programme with annual patient review. This figure rises to 30 if the screeners were also to provide patient education.

    The introduction of a national programme will, at least in the first rounds, significantly increase ophthalmology workloads. None of the departments of ophthalmology in the region could provide data on workload generated by diabetic patients, though two were currently auditing that workload. The lack of information extends also to the management of cataracts detected at screening. Two programmes in the South West (an optometrist based service in its 5th biennial screening round and retinal photography service in its 3rd round) referred 8% and 6.2% respectively of those screened to ophthalmologists. Both refer 4% for retinopathy. The additional patients are largely referred for cataracts, and the higher rate resulted from a decision to refer all patients with cataracts. National criteria for referral of screen detected cataracts are needed.

    Users and commissioners need to be reassured about the quality of screening and treatment of patients. The advisory panel has made recommendations on potential quality criteria, but the success of quality assurance depends on the method of screening used. Though indirect slit lamp examination of the retina is as good as retinal photography when provided by skilled staff, quality assurance would require repeated examination of at least a sample of patients. Moreover, the small numbers of patients per examiner would limit interpretation. In contrast, retinal photography can be stored, electronically transmitted, and reviewed, and numbers per screener are large.

    The advisory panel has made a strong case for introducing a national risk reduction programme for diabetic retinopathy. Careful planning is required to ensure that lessons are learnt from existing programmes, particularly about workforce issues, quality assurance, and impact on secondary services. The programme must also be linked to overall management for diabetic patients.

    References

    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    View Abstract