Clarifying the debate on population-based screening for breast cancer with mammography: A systematic review of randomized controlled trials on mammography with Bayesian meta-analysis and causal model

Tony Hsiu Hsi Chen, Amy Ming Fang Yen, Jean Ching Yuan Fann, Paula Gordon, Sam Li Sheng Chen, Sherry Yueh Hsia Chiu, Chen Yang Hsu, King Jen Chang, Won Chul Lee, Khay Guan Yeoh, Hiroshi Saito, Supannee Promthet, Chisato Hamashima, Alimin Maidin, Fredie Robinson, Li Zhong Zhao

研究成果: 雜誌貢獻回顧型文獻

11 引文 (Scopus)

摘要

Background: The recent controversy about using mammography to screen for breast cancer based on randomized controlled trials over 3 decades in Western countries has not only eclipsed the paradigm of evidence-based medicine, but also puts health decision-makers in countries where breast cancer screening is still being considered in a dilemma to adopt or abandon such a well-established screening modality. Methods: We reanalyzed the empirical data from the Health Insurance Plan trial in 1963 to the UK age trial in 1991 and their follow-up data published until 2015. We first performed Bayesian conjugated meta-analyses on the heterogeneity of attendance rate, sensitivity, and over-detection and their impacts on advanced stage breast cancer and death from breast cancer across trials using Bayesian Poisson fixed- and random-effect regression model. Bayesian meta-analysis of causal model was then developed to assess a cascade of causal relationships regarding the impact of both attendance and sensitivity on 2 main outcomes. Results: The causes of heterogeneity responsible for the disparities across the trials were clearly manifested in 3 components. The attendance rate ranged from 61.3% to 90.4%. The sensitivity estimates show substantial variation from 57.26% to 87.97% but improved with time from 64% in 1963 to 82% in 1980 when Bayesian conjugated meta-analysis was conducted in chronological order. The percentage of over-detection shows a wide range from 0% to 28%, adjusting for long lead-time. The impacts of the attendance rate and sensitivity on the 2 main outcomes were statistically significant. Causal inference made by linking these causal relationships with emphasis on the heterogeneity of the attendance rate and sensitivity accounted for the variation in the reduction of advanced breast cancer (none-30%) and of mortality (none-31%). We estimated a 33% (95% CI: 24-42%) and 13% (95% CI: 6-20%) breast cancer mortality reduction for the best scenario (90% attendance rate and 95% sensitivity) and the poor scenario (30% attendance rate and 55% sensitivity), respectively. Conclusion: Elucidating the scenarios from high to low performance and learning from the experiences of these trials helps screening policy-makers contemplate on how to avoid errors made in ineffective studies and emulate the effective studies to save women lives. Abbreviations: CI = confidence interval, DAG = directed acyclic graphic, HIP = Health Insurance Plan, I/E = incidence of interval cancer/expected incidence, IC = interval cancer, LNS = last negative screening, MCMC = Markov Chain Monte Carlo, NBSS = National Breast Screening Study, NSO = number of screenings required for over-detecting, PCDP = preclinical detectable phase, RR = relative risk.
原文英語
文章編號e5684
期刊Medicine (United States)
96
發行號3
DOIs
出版狀態已發佈 - 2017

指紋

Bayes Theorem
Mammography
Meta-Analysis
Randomized Controlled Trials
Breast Neoplasms
Population
Health Insurance
Markov Chains
Mortality
Evidence-Based Medicine
Incidence
Administrative Personnel
Early Detection of Cancer
Neoplasms
Breast
Learning
Confidence Intervals
Health

ASJC Scopus subject areas

  • Medicine(all)

引用此文

Clarifying the debate on population-based screening for breast cancer with mammography : A systematic review of randomized controlled trials on mammography with Bayesian meta-analysis and causal model. / Chen, Tony Hsiu Hsi; Yen, Amy Ming Fang; Fann, Jean Ching Yuan; Gordon, Paula; Chen, Sam Li Sheng; Chiu, Sherry Yueh Hsia; Hsu, Chen Yang; Chang, King Jen; Lee, Won Chul; Yeoh, Khay Guan; Saito, Hiroshi; Promthet, Supannee; Hamashima, Chisato; Maidin, Alimin; Robinson, Fredie; Zhao, Li Zhong.

於: Medicine (United States), 卷 96, 編號 3, e5684, 2017.

研究成果: 雜誌貢獻回顧型文獻

Chen, Tony Hsiu Hsi ; Yen, Amy Ming Fang ; Fann, Jean Ching Yuan ; Gordon, Paula ; Chen, Sam Li Sheng ; Chiu, Sherry Yueh Hsia ; Hsu, Chen Yang ; Chang, King Jen ; Lee, Won Chul ; Yeoh, Khay Guan ; Saito, Hiroshi ; Promthet, Supannee ; Hamashima, Chisato ; Maidin, Alimin ; Robinson, Fredie ; Zhao, Li Zhong. / Clarifying the debate on population-based screening for breast cancer with mammography : A systematic review of randomized controlled trials on mammography with Bayesian meta-analysis and causal model. 於: Medicine (United States). 2017 ; 卷 96, 編號 3.
@article{fd7d49d925254097bb7b14c15063ac2a,
title = "Clarifying the debate on population-based screening for breast cancer with mammography: A systematic review of randomized controlled trials on mammography with Bayesian meta-analysis and causal model",
abstract = "Background: The recent controversy about using mammography to screen for breast cancer based on randomized controlled trials over 3 decades in Western countries has not only eclipsed the paradigm of evidence-based medicine, but also puts health decision-makers in countries where breast cancer screening is still being considered in a dilemma to adopt or abandon such a well-established screening modality. Methods: We reanalyzed the empirical data from the Health Insurance Plan trial in 1963 to the UK age trial in 1991 and their follow-up data published until 2015. We first performed Bayesian conjugated meta-analyses on the heterogeneity of attendance rate, sensitivity, and over-detection and their impacts on advanced stage breast cancer and death from breast cancer across trials using Bayesian Poisson fixed- and random-effect regression model. Bayesian meta-analysis of causal model was then developed to assess a cascade of causal relationships regarding the impact of both attendance and sensitivity on 2 main outcomes. Results: The causes of heterogeneity responsible for the disparities across the trials were clearly manifested in 3 components. The attendance rate ranged from 61.3{\%} to 90.4{\%}. The sensitivity estimates show substantial variation from 57.26{\%} to 87.97{\%} but improved with time from 64{\%} in 1963 to 82{\%} in 1980 when Bayesian conjugated meta-analysis was conducted in chronological order. The percentage of over-detection shows a wide range from 0{\%} to 28{\%}, adjusting for long lead-time. The impacts of the attendance rate and sensitivity on the 2 main outcomes were statistically significant. Causal inference made by linking these causal relationships with emphasis on the heterogeneity of the attendance rate and sensitivity accounted for the variation in the reduction of advanced breast cancer (none-30{\%}) and of mortality (none-31{\%}). We estimated a 33{\%} (95{\%} CI: 24-42{\%}) and 13{\%} (95{\%} CI: 6-20{\%}) breast cancer mortality reduction for the best scenario (90{\%} attendance rate and 95{\%} sensitivity) and the poor scenario (30{\%} attendance rate and 55{\%} sensitivity), respectively. Conclusion: Elucidating the scenarios from high to low performance and learning from the experiences of these trials helps screening policy-makers contemplate on how to avoid errors made in ineffective studies and emulate the effective studies to save women lives. Abbreviations: CI = confidence interval, DAG = directed acyclic graphic, HIP = Health Insurance Plan, I/E = incidence of interval cancer/expected incidence, IC = interval cancer, LNS = last negative screening, MCMC = Markov Chain Monte Carlo, NBSS = National Breast Screening Study, NSO = number of screenings required for over-detecting, PCDP = preclinical detectable phase, RR = relative risk.",
keywords = "Breast cancer, Heterogeneity, Mammography, Randomized controlled trial, Screening",
author = "Chen, {Tony Hsiu Hsi} and Yen, {Amy Ming Fang} and Fann, {Jean Ching Yuan} and Paula Gordon and Chen, {Sam Li Sheng} and Chiu, {Sherry Yueh Hsia} and Hsu, {Chen Yang} and Chang, {King Jen} and Lee, {Won Chul} and Yeoh, {Khay Guan} and Hiroshi Saito and Supannee Promthet and Chisato Hamashima and Alimin Maidin and Fredie Robinson and Zhao, {Li Zhong}",
year = "2017",
doi = "10.1097/MD.0000000000005684",
language = "English",
volume = "96",
journal = "Medicine; analytical reviews of general medicine, neurology, psychiatry, dermatology, and pediatries",
issn = "0025-7974",
publisher = "Lippincott Williams and Wilkins",
number = "3",

}

TY - JOUR

T1 - Clarifying the debate on population-based screening for breast cancer with mammography

T2 - A systematic review of randomized controlled trials on mammography with Bayesian meta-analysis and causal model

AU - Chen, Tony Hsiu Hsi

AU - Yen, Amy Ming Fang

AU - Fann, Jean Ching Yuan

AU - Gordon, Paula

AU - Chen, Sam Li Sheng

AU - Chiu, Sherry Yueh Hsia

AU - Hsu, Chen Yang

AU - Chang, King Jen

AU - Lee, Won Chul

AU - Yeoh, Khay Guan

AU - Saito, Hiroshi

AU - Promthet, Supannee

AU - Hamashima, Chisato

AU - Maidin, Alimin

AU - Robinson, Fredie

AU - Zhao, Li Zhong

PY - 2017

Y1 - 2017

N2 - Background: The recent controversy about using mammography to screen for breast cancer based on randomized controlled trials over 3 decades in Western countries has not only eclipsed the paradigm of evidence-based medicine, but also puts health decision-makers in countries where breast cancer screening is still being considered in a dilemma to adopt or abandon such a well-established screening modality. Methods: We reanalyzed the empirical data from the Health Insurance Plan trial in 1963 to the UK age trial in 1991 and their follow-up data published until 2015. We first performed Bayesian conjugated meta-analyses on the heterogeneity of attendance rate, sensitivity, and over-detection and their impacts on advanced stage breast cancer and death from breast cancer across trials using Bayesian Poisson fixed- and random-effect regression model. Bayesian meta-analysis of causal model was then developed to assess a cascade of causal relationships regarding the impact of both attendance and sensitivity on 2 main outcomes. Results: The causes of heterogeneity responsible for the disparities across the trials were clearly manifested in 3 components. The attendance rate ranged from 61.3% to 90.4%. The sensitivity estimates show substantial variation from 57.26% to 87.97% but improved with time from 64% in 1963 to 82% in 1980 when Bayesian conjugated meta-analysis was conducted in chronological order. The percentage of over-detection shows a wide range from 0% to 28%, adjusting for long lead-time. The impacts of the attendance rate and sensitivity on the 2 main outcomes were statistically significant. Causal inference made by linking these causal relationships with emphasis on the heterogeneity of the attendance rate and sensitivity accounted for the variation in the reduction of advanced breast cancer (none-30%) and of mortality (none-31%). We estimated a 33% (95% CI: 24-42%) and 13% (95% CI: 6-20%) breast cancer mortality reduction for the best scenario (90% attendance rate and 95% sensitivity) and the poor scenario (30% attendance rate and 55% sensitivity), respectively. Conclusion: Elucidating the scenarios from high to low performance and learning from the experiences of these trials helps screening policy-makers contemplate on how to avoid errors made in ineffective studies and emulate the effective studies to save women lives. Abbreviations: CI = confidence interval, DAG = directed acyclic graphic, HIP = Health Insurance Plan, I/E = incidence of interval cancer/expected incidence, IC = interval cancer, LNS = last negative screening, MCMC = Markov Chain Monte Carlo, NBSS = National Breast Screening Study, NSO = number of screenings required for over-detecting, PCDP = preclinical detectable phase, RR = relative risk.

AB - Background: The recent controversy about using mammography to screen for breast cancer based on randomized controlled trials over 3 decades in Western countries has not only eclipsed the paradigm of evidence-based medicine, but also puts health decision-makers in countries where breast cancer screening is still being considered in a dilemma to adopt or abandon such a well-established screening modality. Methods: We reanalyzed the empirical data from the Health Insurance Plan trial in 1963 to the UK age trial in 1991 and their follow-up data published until 2015. We first performed Bayesian conjugated meta-analyses on the heterogeneity of attendance rate, sensitivity, and over-detection and their impacts on advanced stage breast cancer and death from breast cancer across trials using Bayesian Poisson fixed- and random-effect regression model. Bayesian meta-analysis of causal model was then developed to assess a cascade of causal relationships regarding the impact of both attendance and sensitivity on 2 main outcomes. Results: The causes of heterogeneity responsible for the disparities across the trials were clearly manifested in 3 components. The attendance rate ranged from 61.3% to 90.4%. The sensitivity estimates show substantial variation from 57.26% to 87.97% but improved with time from 64% in 1963 to 82% in 1980 when Bayesian conjugated meta-analysis was conducted in chronological order. The percentage of over-detection shows a wide range from 0% to 28%, adjusting for long lead-time. The impacts of the attendance rate and sensitivity on the 2 main outcomes were statistically significant. Causal inference made by linking these causal relationships with emphasis on the heterogeneity of the attendance rate and sensitivity accounted for the variation in the reduction of advanced breast cancer (none-30%) and of mortality (none-31%). We estimated a 33% (95% CI: 24-42%) and 13% (95% CI: 6-20%) breast cancer mortality reduction for the best scenario (90% attendance rate and 95% sensitivity) and the poor scenario (30% attendance rate and 55% sensitivity), respectively. Conclusion: Elucidating the scenarios from high to low performance and learning from the experiences of these trials helps screening policy-makers contemplate on how to avoid errors made in ineffective studies and emulate the effective studies to save women lives. Abbreviations: CI = confidence interval, DAG = directed acyclic graphic, HIP = Health Insurance Plan, I/E = incidence of interval cancer/expected incidence, IC = interval cancer, LNS = last negative screening, MCMC = Markov Chain Monte Carlo, NBSS = National Breast Screening Study, NSO = number of screenings required for over-detecting, PCDP = preclinical detectable phase, RR = relative risk.

KW - Breast cancer

KW - Heterogeneity

KW - Mammography

KW - Randomized controlled trial

KW - Screening

UR - http://www.scopus.com/inward/record.url?scp=85010366320&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85010366320&partnerID=8YFLogxK

U2 - 10.1097/MD.0000000000005684

DO - 10.1097/MD.0000000000005684

M3 - Review article

C2 - 28099330

AN - SCOPUS:85010366320

VL - 96

JO - Medicine; analytical reviews of general medicine, neurology, psychiatry, dermatology, and pediatries

JF - Medicine; analytical reviews of general medicine, neurology, psychiatry, dermatology, and pediatries

SN - 0025-7974

IS - 3

M1 - e5684

ER -