Reducing the effects of lead-time bias, length bias and over-detection in evaluating screening mammography

A censored bivariate data approach

Jonathan D. Mahnken, Wenyaw Chan, Daniel H. Freeman, Jean L. Freeman

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Measuring the benefit of screening mammography is difficult due to lead-time bias, length bias and over-detection. We evaluated the benefit of screening mammography in reducing breast cancer mortality using observational data from the SEER-Medicare linked database. The conceptual model divided the disease duration into two phases: preclinical (T0) and symptomatic (T1) breast cancer. Censored information for the bivariate response vector (T0 T1) was observed and used to generate a likelihood function. However, the contribution to the likelihood function for some observations could not be calculated analytically, thus, censoring boundaries for these observations were modified. Inferences about the impact of screening mammography on breast cancer mortality were made based on maximum likelihood estimates derived from this likelihood function. Hazard ratios (95% confidence intervals) of 0.54 (0.48-0.61) and 0.33 (0.26-0.42) for single and regular users (vs. non-users), respectively, demonstrated a protective effect of screening mammography among women 69 years and older. This method reduced the impact of lead-time bias, length bias and over-detection, which biased the estimated hazard ratios derived from standard survival models in favour of screening.

Original languageEnglish (US)
Pages (from-to)643-663
Number of pages21
JournalStatistical Methods in Medical Research
Volume17
Issue number6
DOIs
StatePublished - 2008

Fingerprint

Likelihood Functions
Mammography
Screening
Likelihood Function
Breast Cancer
Breast Neoplasms
Mortality
Hazard
Medicare
Survival Model
Conceptual Model
Censoring
Maximum Likelihood Estimate
Databases
Biased
Confidence Intervals
Confidence interval
Standard Model
Survival

ASJC Scopus subject areas

  • Epidemiology
  • Health Information Management
  • Statistics and Probability

Cite this

Reducing the effects of lead-time bias, length bias and over-detection in evaluating screening mammography : A censored bivariate data approach. / Mahnken, Jonathan D.; Chan, Wenyaw; Freeman, Daniel H.; Freeman, Jean L.

In: Statistical Methods in Medical Research, Vol. 17, No. 6, 2008, p. 643-663.

Research output: Contribution to journalArticle

@article{4aa8e4bc995f4ec4b3d69560f597fd9c,
title = "Reducing the effects of lead-time bias, length bias and over-detection in evaluating screening mammography: A censored bivariate data approach",
abstract = "Measuring the benefit of screening mammography is difficult due to lead-time bias, length bias and over-detection. We evaluated the benefit of screening mammography in reducing breast cancer mortality using observational data from the SEER-Medicare linked database. The conceptual model divided the disease duration into two phases: preclinical (T0) and symptomatic (T1) breast cancer. Censored information for the bivariate response vector (T0 T1) was observed and used to generate a likelihood function. However, the contribution to the likelihood function for some observations could not be calculated analytically, thus, censoring boundaries for these observations were modified. Inferences about the impact of screening mammography on breast cancer mortality were made based on maximum likelihood estimates derived from this likelihood function. Hazard ratios (95{\%} confidence intervals) of 0.54 (0.48-0.61) and 0.33 (0.26-0.42) for single and regular users (vs. non-users), respectively, demonstrated a protective effect of screening mammography among women 69 years and older. This method reduced the impact of lead-time bias, length bias and over-detection, which biased the estimated hazard ratios derived from standard survival models in favour of screening.",
author = "Mahnken, {Jonathan D.} and Wenyaw Chan and Freeman, {Daniel H.} and Freeman, {Jean L.}",
year = "2008",
doi = "10.1177/0962280207087309",
language = "English (US)",
volume = "17",
pages = "643--663",
journal = "Statistical Methods in Medical Research",
issn = "0962-2802",
publisher = "SAGE Publications Ltd",
number = "6",

}

TY - JOUR

T1 - Reducing the effects of lead-time bias, length bias and over-detection in evaluating screening mammography

T2 - A censored bivariate data approach

AU - Mahnken, Jonathan D.

AU - Chan, Wenyaw

AU - Freeman, Daniel H.

AU - Freeman, Jean L.

PY - 2008

Y1 - 2008

N2 - Measuring the benefit of screening mammography is difficult due to lead-time bias, length bias and over-detection. We evaluated the benefit of screening mammography in reducing breast cancer mortality using observational data from the SEER-Medicare linked database. The conceptual model divided the disease duration into two phases: preclinical (T0) and symptomatic (T1) breast cancer. Censored information for the bivariate response vector (T0 T1) was observed and used to generate a likelihood function. However, the contribution to the likelihood function for some observations could not be calculated analytically, thus, censoring boundaries for these observations were modified. Inferences about the impact of screening mammography on breast cancer mortality were made based on maximum likelihood estimates derived from this likelihood function. Hazard ratios (95% confidence intervals) of 0.54 (0.48-0.61) and 0.33 (0.26-0.42) for single and regular users (vs. non-users), respectively, demonstrated a protective effect of screening mammography among women 69 years and older. This method reduced the impact of lead-time bias, length bias and over-detection, which biased the estimated hazard ratios derived from standard survival models in favour of screening.

AB - Measuring the benefit of screening mammography is difficult due to lead-time bias, length bias and over-detection. We evaluated the benefit of screening mammography in reducing breast cancer mortality using observational data from the SEER-Medicare linked database. The conceptual model divided the disease duration into two phases: preclinical (T0) and symptomatic (T1) breast cancer. Censored information for the bivariate response vector (T0 T1) was observed and used to generate a likelihood function. However, the contribution to the likelihood function for some observations could not be calculated analytically, thus, censoring boundaries for these observations were modified. Inferences about the impact of screening mammography on breast cancer mortality were made based on maximum likelihood estimates derived from this likelihood function. Hazard ratios (95% confidence intervals) of 0.54 (0.48-0.61) and 0.33 (0.26-0.42) for single and regular users (vs. non-users), respectively, demonstrated a protective effect of screening mammography among women 69 years and older. This method reduced the impact of lead-time bias, length bias and over-detection, which biased the estimated hazard ratios derived from standard survival models in favour of screening.

UR - http://www.scopus.com/inward/record.url?scp=58749098454&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=58749098454&partnerID=8YFLogxK

U2 - 10.1177/0962280207087309

DO - 10.1177/0962280207087309

M3 - Article

VL - 17

SP - 643

EP - 663

JO - Statistical Methods in Medical Research

JF - Statistical Methods in Medical Research

SN - 0962-2802

IS - 6

ER -