Length-time bias
- Bias
Screening mostly finds slow cases, which spend longer without symptoms and already had a better outlook.
What it is
Length-time bias happens because a screening test is more likely to find disease that stays longer in the preclinical stage. Slow, indolent forms pile up in that stage, while aggressive forms quickly cause symptoms, often between two screening rounds. So screen-detected cases have a better prognosis because of their biology, not because they were found early.
How it arises
- A screening round is a snapshot of the preclinical stage at one moment.
- Cases that stay longer in that stage show up more in the snapshot: the slow ones.
- Fast cases slip out of the snapshot because they cause symptoms before the next round.
Worked example
- Data (made up): a tumour has two forms with 100 new cases a year each; the fast one stays 1 year in the preclinical stage, the slow one 9 years.
- Preclinical prevalence of the fast form: 100 × 1 = 100 people.
- Preclinical prevalence of the slow form: 100 × 9 = 900 people.
- A screening round that finds everything preclinical detects 100 + 900 = 1 000 cases.
- Share of slow forms among those detected: 900 / 1 000 = 90 %, although both forms have the same incidence (50 % each).
- Reading: screened cases are mostly slow and will live longer because of their biology; fast forms will show up as interval cancers.
Markers
- Length-time bias
- Interval cancer
- P ≈ I × D
How to prevent or control it
- Evaluate screening with a randomised trial comparing mortality in the whole invited population with a control group, not just in detected cases.
- Analyse by intention to screen: count every case in the invited arm, including interval cancers.
- Distrust comparisons between screen-detected and symptom-detected cases.
Common pitfalls
- Crediting screening with the better prognosis of detected cases without considering that they are a sample of slow forms.
- Mixing up length-time and lead-time bias: length-time is about which cases are found; lead time, about when.
Test yourself
Think of the answer, then tap the question to see it.
Why does screening mostly find the slow forms of a disease?
Because they spend longer in the preclinical stage: at any moment there are more slow than fast cases waiting to be found.
What is an interval cancer?
A cancer diagnosed from symptoms between two screening rounds; it tends to grow fast.
How does length-time bias differ from lead-time bias?
Length-time bias is about which cases screening finds (the slow ones); lead-time bias is about finding them earlier, which lengthens apparent survival.
Connections
Educational content on epidemiological and statistical methods; the examples use made-up numbers to teach the calculation. Not a clinical or legal guide: current Colombian rules are in the official sources.
Sources: Cancer Screening Overview (PDQ®)–Health Professional Version (National Cancer Institute, 2023); Screening programmes: a short guide. Increase effectiveness, maximize benefits and minimize harm (World Health Organization, Regional Office for Europe, 2020).
Spotted a mistake?We are always improving for the community. Tell us what is wrong and our team will review it.Report a mistake
Representative model, for educational purposes only.