Preventive health screening in Singapore: what early detection can and cannot do

A useful screening test identifies a condition early enough to change treatment or prognosis. Screening programmes are judged not only by how many abnormalities they find, but by whether they reduce illness, complications or death without causing excessive harm from false positives.

This is the single most useful concept for anyone evaluating a screening claim, and it is almost never explained in marketing material. Screening can make survival statistics improve dramatically without a single person living longer.

Lead time bias

Imagine a cancer that would be diagnosed from symptoms at age 67 and cause death at 70 — three years of measured survival. Now screen the same person and detect it at 64. If the disease course is unchanged, death still occurs at 70, but measured survival is now six years. Survival has apparently doubled while the person gained nothing at all. They simply spent three additional years knowing they had cancer.

Length time bias

Screening at intervals preferentially detects slow-growing disease, because fast-growing disease tends to appear between screening rounds. The screened group therefore contains a higher proportion of indolent disease with naturally better outcomes — which flatters the screening programme without the screening having changed anything.

So when a test is marketed with a survival statistic, the useful question is not whether the number is impressive but whether it comes from a comparison that could distinguish real benefit from these two effects.

Screening medicine has long-established criteria. They are not complicated, and applying them to any test being offered is a reasonable thing for a patient to do.

A test failing several of these is not necessarily useless — it may be appropriate for a specific higher-risk individual. But it is not population screening, and it should not be sold as though it were.

Singapore's 2026 Screening Test Review Committee updated national guidance using local epidemiology and current evidence. This is a better starting point than choosing tests purely because they are included in a premium package.

Cardiovascular screening deserves particular emphasis because it is unglamorous and clears every criterion above comfortably. Hypertension, diabetes and lipid disorders are common, silent for years, detectable cheaply, and treatable in ways that demonstrably reduce heart attacks and strokes. No advanced biomarker panel currently on the market has that combination of evidence.

False positives can trigger repeat testing, biopsies, radiation exposure, cost and anxiety. Overdiagnosis can detect abnormalities that would never have caused harm during a person's lifetime. These trade-offs become more important as screening expands into very broad imaging or biomarker panels.

Overdiagnosis, explained plainly

Some conditions detected by screening would never have progressed to cause symptoms. Because there is currently no reliable way to identify which ones, they are generally treated — meaning surgery, medication or surveillance for a condition that was never going to cause trouble. The person experiences all the cost, risk and anxiety of illness with none of the benefit of prevention, and typically believes screening saved their life.

Biological-age tools estimate aspects of physiological ageing using laboratory or molecular data. They may be useful for trend tracking and research-informed risk discussions, but they do not replace conventional screening for blood pressure, diabetes, cancer or other established conditions.

Judged against the criteria above, biological-age testing sits in a different category: it is measuring something plausibly meaningful, but it has not been shown that acting on the result improves outcomes, and there is no agreed protocol for what to do when the number is unfavourable. That does not make it worthless — trend data over time can be genuinely informative, and it interests people in their health — but it is a different proposition from a test with mortality evidence behind it.

Aquila's biological-age and longevity assessment is best understood as an additional layer of health profiling rather than a substitute for guideline-based screening.

Population screening is recommended for some cancers because evidence shows benefit in defined groups. Other cancers are screened only in people at higher risk. Singapore MOH specifically notes that lung and liver cancer screening are targeted rather than general-population tests because lower-yield screening can produce many false positives.

Family history, smoking exposure, inherited syndromes and previous abnormal findings can move a person into a different screening pathway.

Multi-cancer blood tests are the current frontier and deserve the same scrutiny as anything else. They can detect signals from several cancers at once, which is genuinely impressive technology — but no such test has yet demonstrated reduced cancer mortality in a completed randomised trial. Our guides to multi-cancer early detection and the SPOT-MAS test cover what the results can and cannot tell you.

• What is my baseline risk? Age, sex, family history, exposures and existing disease.
• Which tests have evidence for someone like me? Not merely which tests are available.
• What happens if the result is abnormal? Follow-up should be planned before the test is ordered.

That third question is the one most often skipped, and it is the one that prevents the worst outcomes. A test you would not act on is a test worth declining. A test whose abnormal result would lead somewhere you are unwilling to go — a biopsy, a scan, a specialist referral — is worth thinking about before rather than after.