What Makes a Good Screening Test

A screening test can measure something accurately and still be unsuitable for population screening. Policymakers must judge the condition, target population, test, next steps and complete programme together. The central question is whether offering the pathway will help people more than it harms them.

A good test is only one part of screening A screening programme is an organised pathway rather than a test offered in isolation. It includes eligibility, information, testing, follow up, intervention, monitoring and quality assurance. The UK National Screening Committee uses detailed screening criteria covering the condition, test, intervention, whole programme and implementation. Reducing these to four simple rules can hide important evidence and safety questions. A promising test does not justify a programme by itself. The complete pathway must produce more benefit than harm at a reasonable cost. The health problem and population must be clear The target condition should be an important health problem because of its frequency, severity or both. A rare condition is not automatically unsuitable when its consequences are serious and early action can help. The target population must also be defined. A test may be worthwhile for a higher risk group even when screening the whole population would create too much harm. The condition's natural history should be understood. Natural history means how the condition develops over time without screening or intervention. There should be a detectable early stage, risk marker or measurable change linked reliably to serious or treatable disease. Policymakers should also consider whether better prevention or ordinary clinical care would achieve more benefit. The test must measure reliably Analytical validity asks whether a test reliably measures the marker it claims to measure. A laboratory test can fail this standard if results vary too much between samples, equipment or laboratories. Clinical validity asks how well the result identifies the target condition in the intended screening population. Evidence from people with symptoms may not transfer directly to people offered screening. Sensitivity, specificity and predictive values should be understood in that population. A threshold must be defined because changing it alters who is referred and who may be missed. The entire testing process must be acceptable, from invitation and sample collection to receiving results. A technically accurate test may still have poor uptake or create unacceptable burdens. The pathway must provide clinical utility Clinical utility asks whether using the test within the complete pathway leads to worthwhile outcomes. This includes what happens after positive, negative, unclear or incidental findings. For many programmes, earlier intervention should improve health compared with usual care. Useful outcomes can include less serious illness, fewer deaths, improved quality of life or reduced complications. Some screening programmes mainly provide reliable information that supports informed choices. In those programmes, the information must be accurate, understandable and valuable to the person screened. The person being screened should have a realistic prospect of benefit. Finding a marker without useful information, monitoring or action is not enough. Earlier diagnosis can create misleading appearances Finding a condition earlier does not prove that screening improves outcomes. The diagnosis date can move forward even when the eventual outcome remains unchanged. This effect is called lead time bias. It can make survival after diagnosis appear longer simply because the clock started earlier. A programme should therefore not be judged only by earlier stage at diagnosis or longer survival after diagnosis. High quality comparative evidence should examine outcomes that match the programme's purpose. Randomised trials are often the strongest design for showing reduced illness or mortality when they are feasible. Other evidence and modelling may support questions that trials cannot answer alone. Benefits and harms must be assessed together The programme should consider false positives, false negatives, uncertain findings, complications, overdiagnosis and overtreatment. It should also consider reassurance, earlier useful action and improved informed choice. The balance depends on the target population, threshold, screening interval and follow up pathway. A test can be useful diagnostically but unsuitable for proactive screening in a lower risk population. Clinical utility includes these combined effects rather than accuracy alone. The next lesson examines screening harms in more detail. Delivery must be safe, fair and sustainable A programme needs enough trained staff, equipment and capacity for testing, diagnosis, treatment and monitoring. Finding more cases without timely follow up can create delay and harm. Quality assurance checks whether the pathway meets agreed standards and improves when problems appear. It should cover invitation, testing, results, referrals, interventions and outcomes. People need accessible, balanced information and a genuine choice to accept or decline. Services should examine whether access and outcomes differ unfairly between population groups. Cost effectiveness compares health benefits, harms and resource use across the whole pathway. It also considers opportunity cost, meaning what other healthcare cannot be provided with the same resources. Recommendations can change Screening policy can change when disease patterns, tests, treatments, evidence or service capacity change. A test that was previously unsuitable may become useful within a better pathway. A programme can also become less worthwhile if prevalence falls or usual care improves. Ongoing evaluation matters after implementation as well as before it. Not recommending population screening does not mean that the condition is unimportant or that the test has no clinical use. Individual diagnostic testing follows a different question and pathway. Do not wait for screening if you develop new, persistent or concerning symptoms. Contact an appropriate healthcare service for clinical assessment. This lesson explains general screening policy principles. It cannot determine whether a particular test or programme is suitable for you.

A screening test is worthwhile only when it performs reliably inside a complete pathway that offers useful benefit, controls harm, supports informed choice and can be delivered fairly and sustainably.