What is lead-time bias?

Lead-time bias is the apparent lengthening of survival that occurs simply because screening moves the date of diagnosis earlier; the person may live no longer overall, yet survival measured from diagnosis appears longer, which is why mortality rather than survival is used to judge screening.

Why does the same test perform differently in different populations?

Sensitivity and specificity describe the test itself, but the positive predictive value, how likely a positive result reflects true disease, also depends on how common the disease is; in a low-prevalence population a larger share of positive results will be false positives.

Screening Principles and Test Evaluation

Screening principles and test evaluation are the cross-cutting concepts used to judge whether any screening test and programme are worthwhile. They cover the criteria a condition and test should meet before screening is offered, the measures of test performance such as sensitivity and predictive value, and the biases that can make screening appear beneficial when it is not.

Definition

Screening principles and test evaluation are the set of criteria and quantitative measures used to assess whether screening an asymptomatic population for a disease produces more benefit than harm, including test sensitivity, specificity, predictive value, and the biases peculiar to screening.

Scope

This topic provides the conceptual toolkit that underlies all organ-specific cancer screening: the Wilson and Jungner framework, the operating characteristics of a test, and the lead-time, length-time and overdiagnosis biases that make mortality the decisive endpoint. It is a methodological reference and not clinical advice about any specific test.

Core questions

What conditions must a disease, test and programme satisfy before screening is justified?
How are a screening test's sensitivity, specificity and predictive value defined and interpreted?
Why can earlier diagnosis appear beneficial through lead-time and length-time bias even when it saves no lives?

Key concepts

Wilson and Jungner criteria
Sensitivity and specificity
Positive and negative predictive value
Disease prevalence and predictive value
Detectable preclinical phase
Lead-time bias
Length-time bias
Overdiagnosis
Mortality as the primary endpoint

Mechanisms

Wilson and Jungner (1968) set out criteria a screening programme should meet: the condition should be an important health problem with a recognizable latent stage, there should be a suitable and acceptable test, and an accepted, effective treatment for those detected. A test's value is described by its sensitivity (the proportion of those with disease who test positive) and specificity (the proportion without disease who test negative); how meaningful a positive result is depends additionally on disease prevalence through the positive predictive value. Crucially, because screening detects disease in a detectable preclinical phase, earlier diagnosis automatically lengthens measured survival (lead-time bias) and preferentially detects slow-growing cases (length-time bias), and may detect disease that would never have caused harm (overdiagnosis); these biases mean that a reduction in cause-specific mortality, ideally from a randomized trial, is the endpoint that demonstrates benefit (Rothman, 2008).

Clinical relevance

These principles are the basis on which expert bodies decide whether to recommend a cancer screening programme and how to interpret its evidence, including the weighing of mortality benefit against overdiagnosis seen in breast screening (Marmot, 2012). The entry describes the appraisal framework for reference; it is not guidance for screening any individual.

History

Wilson and Jungner's 1968 monograph for the World Health Organization codified principles for appraising screening that remain a touchstone, and later work revisited and extended them for new contexts such as genomic screening (Andermann, 2008). In parallel, epidemiologists formalized the biases peculiar to screening, lead-time, length-time and overdiagnosis, establishing why randomized evidence of mortality reduction, rather than improved survival, is required to demonstrate that a programme works (Rothman, 2008).

Debates

Whether the classic screening criteria need updating: The Wilson and Jungner criteria were framed for single-disease, treatment-focused screening; advances such as molecular and genomic tests, and greater attention to overdiagnosis and informed choice, have prompted proposals to revise and synthesize the criteria for contemporary screening.

Key figures

J. M. G. Wilson
Gunnar Jungner
Kenneth Rothman
Anne Andermann

Seminal works

wilson-jungner-1968

Frequently asked questions

What is lead-time bias?: Lead-time bias is the apparent lengthening of survival that occurs simply because screening moves the date of diagnosis earlier; the person may live no longer overall, yet survival measured from diagnosis appears longer, which is why mortality rather than survival is used to judge screening.
Why does the same test perform differently in different populations?: Sensitivity and specificity describe the test itself, but the positive predictive value, how likely a positive result reflects true disease, also depends on how common the disease is; in a low-prevalence population a larger share of positive results will be false positives.