Screening and the UK eye screening programmes
Screening looks for disease in people who feel well, so it must meet a higher standard than ordinary care. This lesson covers the Wilson and Jungner criteria, the biases that make screening look better than it is, the three UK programmes that examine the eye, and why glaucoma is not among them.
What screening is
Screening is the testing of apparently healthy people to find those at higher risk of a condition, so that they can be offered further tests or early treatment. It differs from diagnosis in two ways. The person has no complaint and has been invited by the health service, and the screening test only sorts people into higher and lower risk. A positive screen leads to a diagnostic test, not straight to treatment.
Because the people invited are well, any harm done falls on someone who had not asked for help. False positives cause anxiety and unnecessary investigation. False negatives give false reassurance. A programme is therefore justified only if its benefits clearly outweigh these harms across the whole population. In the UK that judgement is made by the UK National Screening Committee (UK NSC), which advises ministers in all four nations.
The Wilson and Jungner criteria
In 1968 Wilson and Jungner set out ten principles for the World Health Organization that a screening programme should satisfy. They remain the standard exam answer, and the UK NSC's own longer criteria grew from them.
- The condition should be an important health problem.
- There should be an accepted treatment for patients with recognised disease.
- Facilities for diagnosis and treatment should be available.
- There should be a recognisable latent or early symptomatic stage.
- There should be a suitable test or examination.
- The test should be acceptable to the population.
- The natural history of the condition should be adequately understood.
- There should be an agreed policy on whom to treat as patients.
- The cost of case-finding should be balanced against medical spending as a whole.
- Case-finding should be a continuing process, not a single project.
They group into four ideas: the condition (important, understood, with an early stage), the test (suitable and acceptable), the treatment (effective, available, with agreed indications) and the programme (affordable and continuous). The UK NSC adds that there should be evidence from high quality randomised trials that the programme reduces death or illness, and that benefit should outweigh harm, including overdiagnosis.
Biases that flatter screening
People whose disease was found by screening nearly always appear to do better than those who presented with symptoms. Much of this is artefact, which is why randomised trials are demanded.
| Bias | Mechanism | Effect |
|---|---|---|
| Lead-time bias | Screening moves the date of diagnosis earlier. If the outcome happens at the same time anyway, the time from diagnosis to outcome is longer | Survival appears longer although nobody lives longer |
| Length-time bias | Slowly progressive disease spends longer in the detectable symptom-free stage, so a periodic screen picks up more slow cases than fast ones | Screen-detected cases look milder than the disease as a whole |
| Overdiagnosis | The extreme of length-time bias: screening finds disease that would never have caused harm in the person's lifetime | People are treated without benefit and outcomes look excellent |
The NHS Diabetic Eye Screening Programme
Diabetic retinopathy fits the criteria well. It is a leading cause of sight loss in working age adults, it has a long symptom-free stage visible on a photograph, and laser or intravitreal treatment given in time prevents blindness. The details below are for England. The other UK nations run similar programmes.
Everyone with diabetes aged 12 and over is invited. At the appointment visual acuity is recorded, the pupils are dilated with drops, and digital photographs are taken of each retina. The images are graded by trained graders, with a second grader and an arbitration grader for abnormal or disputed images. The interval depends on the result. People whose last two screens were both negative are invited every 24 months. Those with a single negative result or with background retinopathy are invited every 12 months.
| Grade | Meaning | Action |
|---|---|---|
| R0 | No retinopathy | Routine recall |
| R1 | Background retinopathy: microaneurysms, haemorrhages, exudates | Routine recall at 12 months |
| R2 | Pre-proliferative: venous beading, multiple blot haemorrhages, intraretinal microvascular abnormalities | Referable. Digital surveillance or routine referral to the hospital eye service |
| R3A | Active proliferative: new vessels on the disc or elsewhere, pre-retinal or vitreous haemorrhage, tractional detachment | Urgent referral |
| R3S | Stable proliferative retinopathy after laser treatment | Monitored. Reactivation is regraded R3A |
| M0 / M1 | No maculopathy / referable maculopathy | M1 is referable |
| P0 / P1 | No laser scars / photocoagulation scars present | Recorded alongside R and M |
| U | Ungradable image | Slit lamp examination |
M1 means any of: exudate within one disc diameter of the centre of the fovea, a group of exudates within the macula, retinal thickening within one disc diameter of the fovea, or a microaneurysm or haemorrhage within one disc diameter of the fovea only if acuity is 6/12 or worse. Current grading guidance also divides R2 into a lower risk group, watched in digital surveillance, and a higher risk group, referred routinely. Each eye receives a grade such as R1 M0 P0, and the worse eye decides the outcome.
The programme standards set the pace. An urgent R3A referral should be seen in the hospital eye service within 6 weeks of screening and a routine R2 or M1 referral within 13 weeks. Digital surveillance clinics review people who need a closer watch than yearly screening but not yet hospital care, at intervals such as 3, 6 or 9 months, and may add optical coherence tomography to check suspected maculopathy.
Pregnancy can accelerate retinopathy, so women with pre-existing diabetes are screened when the service learns of the pregnancy and again at 28 weeks. If the first screen shows background retinopathy, an extra screen is done at 16 to 20 weeks. Anything worse than background retinopathy is referred to the hospital eye service.
Newborn examination and childhood vision screening
The Newborn and Infant Physical Examination (NIPE) screens the eyes, heart, hips and testes twice: within 72 hours of birth and again at 6 to 8 weeks, usually in primary care. The eye screen exists mainly to find congenital cataract, which affects about 2 to 3 babies in 10,000 and is the commonest treatable cause of childhood blindness. The urgency comes from amblyopia. A dense cataract blocks the image during the critical period of visual development, and surgery for severe cases is usually done between 6 and 10 weeks of age.
The examiner checks that both eyes are present, of equal size and have clear corneas and round pupils, then tests the red reflex with an ophthalmoscope at arm's length in dim light. An absent, white, partly obscured or asymmetrical reflex is a positive screen. A baby who screens positive at birth should be seen by an ophthalmologist within 2 weeks. One who screens positive at 6 to 8 weeks should be seen by 11 weeks of age.
The UK NSC recommends vision screening for all children aged 4 to 5, in practice during the first year at school. The target is amblyopia, usually caused by squint or refractive error, which is treatable with glasses and occlusion at this age but becomes permanent if left. The service is led by orthoptists. Each child reads a crowded logMAR chart (the Keeler test in the English specification), and acuity worse than 0.20 logMAR in either eye leads to referral. Crowded letters are used because amblyopic eyes read isolated letters deceptively well.
Why there is no glaucoma screening
Glaucoma looks like an ideal target: it is common in older people, symptomless until late, irreversible and treatable. Yet the UK NSC does not recommend population screening for it. At its last published review in 2019 the committee found it unclear whether the available tests are accurate enough for a programme, and unclear whether people found by screening would do better than under current care.
Tested against Wilson and Jungner, glaucoma fails mainly on the test. No single measurement is both sensitive and specific. Intraocular pressure misses the many patients with glaucoma at normal pressure and flags the many people with raised pressure who never develop it. Prevalence in the general population is low, so even a fairly specific test yields a poor positive predictive value and floods clinics with false positives, as the previous lesson showed.
The UK relies instead on opportunistic case-finding: optometrists check the discs, pressures and fields of adults who attend for a routine sight test and refer those with suspicious findings.
Summary
- Screening tests well people to find those at higher risk, and a positive screen leads to a diagnostic test.
- The Wilson and Jungner criteria cover the condition, the test, the treatment and the programme.
- Lead-time bias makes survival look longer because diagnosis is earlier, without any change in outcome.
- Length-time bias arises because screening preferentially detects slowly progressive disease.
- Diabetic eye screening is offered to everyone with diabetes aged 12 and over, every 24 months after two consecutive negative screens and every 12 months otherwise.
- R3A is active proliferative retinopathy and needs urgent referral, while R2 and M1 are referable routinely.
- The newborn and infant examination tests the red reflex within 72 hours and at 6 to 8 weeks, mainly to find congenital cataract.
- Vision screening at age 4 to 5 is orthoptist-led, uses a crowded logMAR test and targets amblyopia.
- The UK NSC does not recommend population screening for glaucoma, mainly because no test is accurate enough.
Check yourself
Quick recall on this lesson. These are easier than the question bank.
Written for revision. It is not a source of clinical advice, and guidance changes.