Conclusion about the association between valve surgery and mortality in an infective endocarditis cohort changed after adjusting for survivor bias

Imad M. Tleyjeh, Hassan M K Ghomrawi, James M. Steckelberg, Victor Manuel Montori, Tanya L. Hoskin, Felicity T Enders, W Charles Huskins, Farouk Mookadam, Walter R. Wilson, Valerie Zimmerman, Larry M. Baddour

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

Objective: Survivor bias commonly weakens observational studies, even those published in premier journals. It occurs because patients who live longer are more likely to receive treatment than those who die early. We sought to quantify the effect of survivor bias on the association between valve surgery and mortality in infective endocarditis (IE). Study Design and Setting: The study cohort included 546 IE patients. We compared the hazard ratios (HR) resulting from two propensity score analysis approaches that adjusted for survivor bias (time-dependent variable and matching on follow-up time) with those achieved using the same models but without that adjustment (time-fixed variable). Results: In the total cohort, the HR of surgery in the time-dependent model was 1.9 (95% confidence interval [CI] = 1.1-3.2; P = 0.03) vs. 0.9 (95% CI = 0.5-1.4; P = 0.53) in the time-fixed model. In the propensity score-matched subset, the HR of surgery was 1.3 (95% CI = 0.5-3.1; P = 0.56) and 0.8 (95% CI = 0.4-1.7; P = 0.57) in the subset with and without matching on follow-up time, respectively. Conclusion: Adjusting for survivor bias changed the conclusion about the association between valve surgery and mortality in IE. Researchers should be aware of this bias when evaluating observational studies of treatment efficacy.

Original languageEnglish (US)
Pages (from-to)130-135
Number of pages6
JournalJournal of Clinical Epidemiology
Volume63
Issue number2
DOIs
StatePublished - Feb 2010

Fingerprint

Endocarditis
Survivors
Mortality
Confidence Intervals
Propensity Score
Observational Studies
Cohort Studies
Research Personnel

Keywords

  • Endocarditis
  • Observational studies
  • Propensity score
  • Surgery
  • Survivor bias
  • Time dependent

ASJC Scopus subject areas

  • Epidemiology

Cite this

Conclusion about the association between valve surgery and mortality in an infective endocarditis cohort changed after adjusting for survivor bias. / Tleyjeh, Imad M.; Ghomrawi, Hassan M K; Steckelberg, James M.; Montori, Victor Manuel; Hoskin, Tanya L.; Enders, Felicity T; Huskins, W Charles; Mookadam, Farouk; Wilson, Walter R.; Zimmerman, Valerie; Baddour, Larry M.

In: Journal of Clinical Epidemiology, Vol. 63, No. 2, 02.2010, p. 130-135.

Research output: Contribution to journalArticle

Tleyjeh, Imad M. ; Ghomrawi, Hassan M K ; Steckelberg, James M. ; Montori, Victor Manuel ; Hoskin, Tanya L. ; Enders, Felicity T ; Huskins, W Charles ; Mookadam, Farouk ; Wilson, Walter R. ; Zimmerman, Valerie ; Baddour, Larry M. / Conclusion about the association between valve surgery and mortality in an infective endocarditis cohort changed after adjusting for survivor bias. In: Journal of Clinical Epidemiology. 2010 ; Vol. 63, No. 2. pp. 130-135.
@article{48c986b50ded4552b92235e2777fef76,
title = "Conclusion about the association between valve surgery and mortality in an infective endocarditis cohort changed after adjusting for survivor bias",
abstract = "Objective: Survivor bias commonly weakens observational studies, even those published in premier journals. It occurs because patients who live longer are more likely to receive treatment than those who die early. We sought to quantify the effect of survivor bias on the association between valve surgery and mortality in infective endocarditis (IE). Study Design and Setting: The study cohort included 546 IE patients. We compared the hazard ratios (HR) resulting from two propensity score analysis approaches that adjusted for survivor bias (time-dependent variable and matching on follow-up time) with those achieved using the same models but without that adjustment (time-fixed variable). Results: In the total cohort, the HR of surgery in the time-dependent model was 1.9 (95{\%} confidence interval [CI] = 1.1-3.2; P = 0.03) vs. 0.9 (95{\%} CI = 0.5-1.4; P = 0.53) in the time-fixed model. In the propensity score-matched subset, the HR of surgery was 1.3 (95{\%} CI = 0.5-3.1; P = 0.56) and 0.8 (95{\%} CI = 0.4-1.7; P = 0.57) in the subset with and without matching on follow-up time, respectively. Conclusion: Adjusting for survivor bias changed the conclusion about the association between valve surgery and mortality in IE. Researchers should be aware of this bias when evaluating observational studies of treatment efficacy.",
keywords = "Endocarditis, Observational studies, Propensity score, Surgery, Survivor bias, Time dependent",
author = "Tleyjeh, {Imad M.} and Ghomrawi, {Hassan M K} and Steckelberg, {James M.} and Montori, {Victor Manuel} and Hoskin, {Tanya L.} and Enders, {Felicity T} and Huskins, {W Charles} and Farouk Mookadam and Wilson, {Walter R.} and Valerie Zimmerman and Baddour, {Larry M.}",
year = "2010",
month = "2",
doi = "10.1016/j.jclinepi.2008.06.022",
language = "English (US)",
volume = "63",
pages = "130--135",
journal = "Journal of Clinical Epidemiology",
issn = "0895-4356",
publisher = "Elsevier USA",
number = "2",

}

TY - JOUR

T1 - Conclusion about the association between valve surgery and mortality in an infective endocarditis cohort changed after adjusting for survivor bias

AU - Tleyjeh, Imad M.

AU - Ghomrawi, Hassan M K

AU - Steckelberg, James M.

AU - Montori, Victor Manuel

AU - Hoskin, Tanya L.

AU - Enders, Felicity T

AU - Huskins, W Charles

AU - Mookadam, Farouk

AU - Wilson, Walter R.

AU - Zimmerman, Valerie

AU - Baddour, Larry M.

PY - 2010/2

Y1 - 2010/2

N2 - Objective: Survivor bias commonly weakens observational studies, even those published in premier journals. It occurs because patients who live longer are more likely to receive treatment than those who die early. We sought to quantify the effect of survivor bias on the association between valve surgery and mortality in infective endocarditis (IE). Study Design and Setting: The study cohort included 546 IE patients. We compared the hazard ratios (HR) resulting from two propensity score analysis approaches that adjusted for survivor bias (time-dependent variable and matching on follow-up time) with those achieved using the same models but without that adjustment (time-fixed variable). Results: In the total cohort, the HR of surgery in the time-dependent model was 1.9 (95% confidence interval [CI] = 1.1-3.2; P = 0.03) vs. 0.9 (95% CI = 0.5-1.4; P = 0.53) in the time-fixed model. In the propensity score-matched subset, the HR of surgery was 1.3 (95% CI = 0.5-3.1; P = 0.56) and 0.8 (95% CI = 0.4-1.7; P = 0.57) in the subset with and without matching on follow-up time, respectively. Conclusion: Adjusting for survivor bias changed the conclusion about the association between valve surgery and mortality in IE. Researchers should be aware of this bias when evaluating observational studies of treatment efficacy.

AB - Objective: Survivor bias commonly weakens observational studies, even those published in premier journals. It occurs because patients who live longer are more likely to receive treatment than those who die early. We sought to quantify the effect of survivor bias on the association between valve surgery and mortality in infective endocarditis (IE). Study Design and Setting: The study cohort included 546 IE patients. We compared the hazard ratios (HR) resulting from two propensity score analysis approaches that adjusted for survivor bias (time-dependent variable and matching on follow-up time) with those achieved using the same models but without that adjustment (time-fixed variable). Results: In the total cohort, the HR of surgery in the time-dependent model was 1.9 (95% confidence interval [CI] = 1.1-3.2; P = 0.03) vs. 0.9 (95% CI = 0.5-1.4; P = 0.53) in the time-fixed model. In the propensity score-matched subset, the HR of surgery was 1.3 (95% CI = 0.5-3.1; P = 0.56) and 0.8 (95% CI = 0.4-1.7; P = 0.57) in the subset with and without matching on follow-up time, respectively. Conclusion: Adjusting for survivor bias changed the conclusion about the association between valve surgery and mortality in IE. Researchers should be aware of this bias when evaluating observational studies of treatment efficacy.

KW - Endocarditis

KW - Observational studies

KW - Propensity score

KW - Surgery

KW - Survivor bias

KW - Time dependent

UR - http://www.scopus.com/inward/record.url?scp=72749093341&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=72749093341&partnerID=8YFLogxK

U2 - 10.1016/j.jclinepi.2008.06.022

DO - 10.1016/j.jclinepi.2008.06.022

M3 - Article

C2 - 20122499

AN - SCOPUS:72749093341

VL - 63

SP - 130

EP - 135

JO - Journal of Clinical Epidemiology

JF - Journal of Clinical Epidemiology

SN - 0895-4356

IS - 2

ER -