Towards comprehensive population-based screening for diabetic retinopathy: operation of the North Wales diabetic...

To examine whether population-based retinopathy screening using a central diabetes register and employing various screening methods can achieve a high degree of population coverage to meet National Screening Committee (NSC) targets. To identify the main barriers to achieving comprehensive population coverage. To analyse referrals to ophthalmology and their outcome as a measure of previously unmet need. To assess the influence of modality of screening.

Three local health board areas in North Wales.

Establishment of a district diabetes register to hold records of all patients and subserve call-recall of general practitioner (GP) sole-care patients for screening by optometrists or digital photography. Hospital attenders were screened in diabetic clinic by direct ophthalmoscopy. Data were collected for years 1 and 2 of operation of the scheme.

The system held a screening record for 86% of diabetic patients after year 1 and 93% after year 2. Failure to attend was the major barrier to comprehensive population screening, but this improved in year 2 (P<0.001). Both optometrists and photography identified substantial unmet need: 1% of all GP sole-care patients required immediate laser treatment. Photography was more sensitive than optometrist screening, but the additional retinopathy identified was mostly minor not requiring treatment.

Lack of screening is much more important than modality of screening as a cause of missed sight-threatening retinopathy. A central, district-based patient register system identifies those patients not screened for further follow-up and can produce a high level of population coverage, close to NSC targets. Such schemes are needed, particularly to support primary care diabetes management.