Forskning
Udskriv Udskriv
Switch language
Region Hovedstaden - en del af Københavns Universitetshospital
Udgivet

Lead-time models should not be used to estimate overdiagnosis in cancer screening

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

Harvard

APA

CBE

MLA

Vancouver

Author

Bibtex

@article{9d2a7b557856464cb2738556905ffd42,
title = "Lead-time models should not be used to estimate overdiagnosis in cancer screening",
abstract = "Lead-time can mean two different things: Clinical lead-time is the lead-time for clinically relevant tumors; that is, those that are not overdiagnosed. Model-based lead-time is a theoretical construct where the time when the tumor would have caused symptoms is not limited by the person's death. It is the average time at which the diagnosis is brought forward for both clinically relevant and overdiagnosed cancers. When screening for breast cancer, clinical lead-time is about 1 year, while model-based lead-time varies from 2 to 7 years. There are two different methods to calculate overdiagnosis in cancer screening--the excess-incidence approach and the lead-time approach--that rely on two different lead-time definitions. Overdiagnosis when screening with mammography has varied from 0 to 75 %. We have explained that these differences are mainly caused by using different definitions and methods and not by variations in data. High levels of overdiagnosis of cancer have usually been explained by detection of many slow-growing tumors with long lead-times. This theory can be tested by studying if slow-growing tumors accumulate in the absence of screening, which they don't. Thus, it is likely that the natural history of many subclinical cancers is spontaneous regression.",
author = "Per-Henrik Zahl and J{\o}rgensen, {Karsten Juhl} and G{\o}tzsche, {Peter C}",
year = "2014",
month = sep,
doi = "10.1007/s11606-014-2812-2",
language = "English",
volume = "29",
pages = "1283--6",
journal = "Journal of General Internal Medicine",
issn = "0884-8734",
publisher = "Springer New York LLC",
number = "9",

}

RIS

TY - JOUR

T1 - Lead-time models should not be used to estimate overdiagnosis in cancer screening

AU - Zahl, Per-Henrik

AU - Jørgensen, Karsten Juhl

AU - Gøtzsche, Peter C

PY - 2014/9

Y1 - 2014/9

N2 - Lead-time can mean two different things: Clinical lead-time is the lead-time for clinically relevant tumors; that is, those that are not overdiagnosed. Model-based lead-time is a theoretical construct where the time when the tumor would have caused symptoms is not limited by the person's death. It is the average time at which the diagnosis is brought forward for both clinically relevant and overdiagnosed cancers. When screening for breast cancer, clinical lead-time is about 1 year, while model-based lead-time varies from 2 to 7 years. There are two different methods to calculate overdiagnosis in cancer screening--the excess-incidence approach and the lead-time approach--that rely on two different lead-time definitions. Overdiagnosis when screening with mammography has varied from 0 to 75 %. We have explained that these differences are mainly caused by using different definitions and methods and not by variations in data. High levels of overdiagnosis of cancer have usually been explained by detection of many slow-growing tumors with long lead-times. This theory can be tested by studying if slow-growing tumors accumulate in the absence of screening, which they don't. Thus, it is likely that the natural history of many subclinical cancers is spontaneous regression.

AB - Lead-time can mean two different things: Clinical lead-time is the lead-time for clinically relevant tumors; that is, those that are not overdiagnosed. Model-based lead-time is a theoretical construct where the time when the tumor would have caused symptoms is not limited by the person's death. It is the average time at which the diagnosis is brought forward for both clinically relevant and overdiagnosed cancers. When screening for breast cancer, clinical lead-time is about 1 year, while model-based lead-time varies from 2 to 7 years. There are two different methods to calculate overdiagnosis in cancer screening--the excess-incidence approach and the lead-time approach--that rely on two different lead-time definitions. Overdiagnosis when screening with mammography has varied from 0 to 75 %. We have explained that these differences are mainly caused by using different definitions and methods and not by variations in data. High levels of overdiagnosis of cancer have usually been explained by detection of many slow-growing tumors with long lead-times. This theory can be tested by studying if slow-growing tumors accumulate in the absence of screening, which they don't. Thus, it is likely that the natural history of many subclinical cancers is spontaneous regression.

U2 - 10.1007/s11606-014-2812-2

DO - 10.1007/s11606-014-2812-2

M3 - Journal article

C2 - 24590736

VL - 29

SP - 1283

EP - 1286

JO - Journal of General Internal Medicine

JF - Journal of General Internal Medicine

SN - 0884-8734

IS - 9

ER -

ID: 44633547