Reference
Glossary
27 terms used in preventive screening, defined so each one stands on its own. Every entry is linkable.
Tests and modalities
- Whole-body MRI
- Magnetic resonance imaging covering most of the body in a single session, typically brain, chest, abdomen, pelvis and spine, without ionising radiation. Used in preventive screening to look for structural disease in people without symptoms. No randomised trial has shown that it reduces mortality in asymptomatic adults. Our assessment, grade C →
- Multi-cancer early detection test
- A blood test that looks for fragments of tumour DNA circulating in plasma and, if it finds them, predicts the likely tissue of origin. In the PATHFINDER 2 study of 35,878 adults, episode sensitivity was 39.3% across all cancers with 99.6% specificity, rising to 73.7% sensitivity for the twelve cancers responsible for about two-thirds of US cancer deaths. Our assessment, grade B minus →
- Spiroergometry
- An exercise test on a bicycle or treadmill with simultaneous measurement of expired gases, producing VO₂ max, the maximum rate of oxygen consumption. Cardiorespiratory fitness measured this way carries one of the strongest mortality associations of any measurable variable, with no observed upper limit of benefit. Our assessment, grade B →
- Coronary artery calcium score
- A low-dose CT scan of the heart that quantifies calcified plaque in the coronary arteries and returns a number, the Agatston score. Zero means no detectable calcified plaque. It appears by name in the 2018 ACC/AHA cholesterol guideline for intermediate-risk adults, and in Germany requires an individual justified indication under §83 StrlSchG. Our assessment, grade B →
- Epigenetic clock
- An algorithm that estimates biological age or pace of ageing from DNA methylation patterns at selected CpG sites in a blood sample. Repeated-measures work shows estimates fluctuate substantially with meals, acute stress and pollution exposure, and that technical reproducibility does not predict biological stability. Our assessment, grade D →
- Apolipoprotein B
- A blood measurement that counts atherogenic lipoprotein particles rather than the cholesterol carried inside them. One ApoB molecule sits on each such particle. It entered the US cholesterol guideline as a recommended measurement in 2026, particularly where triglycerides are 150 mg/dL or above, in diabetes, or where achieved LDL-C is under 70 mg/dL. What to test by age →
- Lipoprotein(a)
- An inherited lipoprotein whose blood concentration is almost entirely genetically determined and stable across adult life. The 2026 ACC/AHA guideline recommends measuring it at least once in all adults as a Class I recommendation. Roughly 20% of people are at or above 125 nmol/L, about a 1.4-fold increase in ASCVD risk; at or above 250 nmol/L the risk is at least doubled. What to test by age →
- DEXA
- A low-dose X-ray scan measuring bone mineral density and, in some protocols, body composition. The reference standard for diagnosing osteoporosis and inexpensive relative to the rest of a premium panel.
- Polygenic risk score
- A single number summarising the combined effect of many common genetic variants on the probability of a disease. Useful for stratifying populations; its value for an individual decision is contested, and scores derived largely from European cohorts perform worse in other ancestries.
Reading a test result
- Sensitivity
- The proportion of people who have the condition that the test correctly identifies. A test with 40% sensitivity misses six cases in ten. Sensitivity governs what a negative result is worth: the lower it is, the less a negative rules out.
- Specificity
- The proportion of people without the condition that the test correctly clears. A test with 99.6% specificity produces about four false alarms per thousand people screened. Specificity governs what a positive result is worth.
- Positive predictive value
- The probability that somebody with a positive result actually has the condition. Unlike sensitivity and specificity it depends on how common the condition is, which is why a test performing well in a high-risk clinic can produce mostly false positives when applied to a healthy screening population.
- Episode sensitivity
- Sensitivity measured across a complete screening episode rather than a single test run, counting a cancer as detected if the episode flagged it. The figure reported for multi-cancer blood tests in the PATHFINDER studies.
- Confidence interval
- The range within which a true value plausibly lies given the data, conventionally at 95%. A wide interval means the estimate is uncertain. A result quoted without one is a point estimate presented as a fact, which is the most common omission in consumer test reporting.
- Hazard ratio
- The ratio of event rates between two groups over time. A hazard ratio of 2.0 means the event occurs twice as often in one group as the other, which says nothing about how common the event is in absolute terms. A doubled risk of something rare remains rare.
- Regression to the mean
- The tendency of an extreme measurement to be followed by one closer to the average, for statistical rather than causal reasons. Anyone selected for a poor initial result will on average improve on retest with no intervention, which is sufficient on its own to produce before-and-after marketing claims. A worked example →
Screening concepts
- Incidental finding
- An abnormality discovered by a scan performed for another reason, unrelated to any symptom. In apparently asymptomatic adults, pooled prevalence on brain and body MRI is 13.4% for potentially serious findings and 32.1% once indeterminate findings are included, so roughly one scanned person in three leaves with something unresolved. The cost of looking →
- Overdiagnosis
- The correct identification of a disease that would never have caused symptoms or death in that person’s lifetime. Distinct from a false positive: the diagnosis is accurate, but the detection produces only the harms of treatment. The central harm of any screening programme that finds indolent disease.
- Surrogate endpoint
- A measurement used in place of the outcome that actually matters, such as tumour stage at diagnosis standing in for survival. Surrogates make trials faster and cheaper. The history of cancer screening includes technologies that improved a surrogate and produced no survival benefit. How we grade →
- Stage shift
- A change in the distribution of cancer stages at diagnosis toward earlier stages in a screened population. A reasonable surrogate for screening benefit, and not the same as one: stage shift also occurs when screening detects disease that was never going to progress.
- Lead-time bias
- The apparent survival gain created purely by diagnosing a disease earlier. If a condition is found three years sooner but the date of death is unchanged, measured survival from diagnosis rises by three years while nothing about the outcome has improved.
- Number needed to screen
- How many people must be screened over a defined period to prevent one death or adverse outcome. The most honest single figure for describing a screening programme, and the one least often published by commercial providers.
- Asymptomatic
- Having no symptoms of the condition being looked for. The distinction matters because evidence generated in symptomatic patients, where a test is used diagnostically, does not transfer to screening healthy people, where the prevalence is far lower and the balance of benefit and harm changes.
German system
- Check-up 35
- Germany’s statutory preventive health examination, free to statutory insurees every three years from age 35, plus one check between 18 and 34. Covers a physical examination, blood pressure, fasting glucose, a full lipid panel, urinalysis, a one-off hepatitis B and C screen and a consultation. It includes no imaging, tumour markers, genetics or functional testing. What it covers, itemised →
- IGeL
- An individual health service in Germany that statutory insurance does not cover and the patient pays for privately. Most private preventive screening falls into this category by definition, which is why a statutory insuree pays out of pocket for anything beyond the Check-up 35 entitlement.
- Rechtfertigende Indikation
- The determination required under §83 StrlSchG before any examination using ionising radiation in Germany, establishing that the health benefit to that individual outweighs the radiation risk. Screening applications are permitted only after separate federal assessment under §§84 and 14(3), which is why cardiac CT cannot be sold as a routine screening item. What this means for calcium scoring →
- GOÄ
- The German fee schedule for private medical billing. It constrains what physicians may charge privately, subject to multipliers, and is one structural reason a comprehensive private workup costs substantially less in Germany than in the United States.