[Durgical](https://durgical.com/en/) › [DurgiPedia](https://durgical.com/en/durgipedia/) › [Epidemiology](https://durgical.com/en/epidemiology/) › [Screening](https://durgical.com/en/epidemiology/areas/screening/) › Lead-time bias

# Lead-time bias

- Bias

Screening moves the date of diagnosis earlier; survival looks longer even if the person dies on the same date.

Updated Nov 5, 2026

## Key points

- Lead time: the time between detection by screening and the moment the disease would have been diagnosed from symptoms.
- Survival from diagnosis grows by that amount even if the course does not change.
- Five-year survival is no good for judging screening: lead time inflates it.
- The right end point is disease-specific mortality in the invited population, compared with a randomised control group.
- It adds to length-time bias and overdiagnosis: all three make screening look better than it is.

## What it is

Lead-time bias arises when screening finds a disease before symptoms and survival is counted from diagnosis. The clock starts earlier, so screen-detected cases seem to live longer, even if treatment does not change the date of death. That is why a screening programme's efficacy is measured by disease-specific mortality in a randomised trial, not by case survival.

## How it arises

- The survival clock starts on the date of diagnosis.
- Screening moves that date back, into the subclinical stage.
- If early treatment does not change the course, the date of death stays the same and the whole apparent gain is lead time.

## Worked example

- Data (made up): a person's disease begins at age 50; symptoms would appear at 60 and death at 64.
- Without screening: diagnosis at 60; survival from diagnosis: 64 − 60 = 4 years.
- With screening: diagnosis at 57; if treatment changes nothing, death is still at 64: survival 64 − 57 = 7 years.
- Lead time: 60 − 57 = 3 years; the survival gained (7 − 4 = 3 years) is exactly the lead time.
- Five years after diagnosis, the unscreened person has died and the screened one is alive: five-year survival goes from 0 % to 100 %.
- In a trial of 20 000 people, 10 000 per arm, 30 die of the disease in each arm: 3 per 1 000 in both.
- Reading: survival improved through lead time alone; equal mortality shows that screening saved no lives.

## Markers

- Lead time
- Five-year survival
- Disease-specific mortality

## How to prevent or control it

- Design randomised trials that compare disease mortality between those invited and not invited to screening.
- If survival is used, count it from a date common to both groups (for example, randomisation) and not from diagnosis.
- Correct for the estimated lead time when comparing screen-detected and symptom-detected case series.

## Common pitfalls

- Celebrating a rise in five-year survival after a screening programme starts without looking at mortality.
- Comparing screen-detected with symptom-detected cases as if they were equivalent groups.

## Memory aid

The clock moves earlier, not the life: if the date of death does not change, screening only moved the starting line.

## Test yourself

Think of the answer, then tap the question to see it.

## What is lead time?

The interval between detection of the disease by screening and the date it would have been diagnosed from symptoms.

## Why is five-year survival no good for judging screening?

Because the clock starts earlier in screened people: survival rises even if the date of death does not change.

## Which end point avoids lead-time bias when evaluating screening?

Disease-specific mortality in a randomised trial comparing those invited and not invited.

[Add to my cards (3)](https://durgical.app/estudiar/?add=/en/epidemiology/lead-time-bias/)

## Connections

## Related 5

- [Length-time bias](https://durgical.com/en/epidemiology/length-time-bias/)
- [Overdiagnosis](https://durgical.com/en/epidemiology/overdiagnosis/)
- [Principles of screening](https://durgical.com/en/epidemiology/principles-of-screening/)
- [Survival analysis and censoring](https://durgical.com/en/epidemiology/survival-analysis-and-censoring/)
- [Randomised controlled trial](https://durgical.com/en/epidemiology/randomised-controlled-trial/)

## Related conditions 2

- [Prostate cancer](https://durgical.com/en/conditions/prostate-cancer/)
- [Breast cancer](https://durgical.com/en/conditions/breast-cancer/)

[← All of epidemiology](https://durgical.com/en/epidemiology/)

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](https://www.cancer.gov/about-cancer/screening/hp-screening-overview-pdq) (National Cancer Institute, 2023); [Principles of Epidemiology | Lesson 1 - Section 9](https://archive.cdc.gov/www_cdc_gov/csels/dsepd/ss1978/lesson1/section9.html) (Centers for Disease Control and Prevention, 2012).

## **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.

---
Source: https://durgical.com/en/epidemiology/lead-time-bias/
License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Cite as: DurgiPedia (durgical.com), https://durgical.com/en/epidemiology/lead-time-bias/