Exact inference for the risk ratio with an imperfect diagnostic test

J. Reiczigel, J. Singer, Zs Lang

Research output: Contribution to journalArticle

Abstract

SUMMARY The risk ratio quantifies the risk of disease in a study population relative to a reference population. Standard methods of estimation and testing assume a perfect diagnostic test having sensitivity and specificity of 100%. However, this assumption typically does not hold, and this may invalidate naive estimation and testing for the risk ratio. We propose procedures that control for sensitivity and specificity of the diagnostic test, given the risks are measured by proportions, as it is in cross-sectional studies or studies with fixed follow-up times. These procedures provide an exact unconditional test and confidence interval for the true risk ratio. The methods also cover the case when sensitivity and specificity differ in the two groups (differential misclassification). The resulting test and confidence interval may be useful in epidemiological studies as well as in clinical and vaccine trials. We illustrate the method with real-life examples which demonstrate that ignoring sensitivity and specificity of the diagnostic test may lead to considerable bias in the estimated risk ratio.

Original languageEnglish
Pages (from-to)187-193
Number of pages7
JournalEpidemiology and Infection
Volume145
Issue number1
DOIs
Publication statusPublished - Jan 1 2017

Fingerprint

Routine Diagnostic Tests
Odds Ratio
Sensitivity and Specificity
Confidence Intervals
Population
Epidemiologic Studies
Vaccines
Cross-Sectional Studies
Clinical Trials

Keywords

  • exact unconditional test
  • Key words Exact confidence interval
  • misclassification
  • prevalence ratio
  • relative risk

ASJC Scopus subject areas

  • Epidemiology
  • Infectious Diseases

Cite this

Exact inference for the risk ratio with an imperfect diagnostic test. / Reiczigel, J.; Singer, J.; Lang, Zs.

In: Epidemiology and Infection, Vol. 145, No. 1, 01.01.2017, p. 187-193.

Research output: Contribution to journalArticle

Reiczigel, J. ; Singer, J. ; Lang, Zs. / Exact inference for the risk ratio with an imperfect diagnostic test. In: Epidemiology and Infection. 2017 ; Vol. 145, No. 1. pp. 187-193.
@article{1808fca6f18c43cfb9d3ffce89e8d420,
title = "Exact inference for the risk ratio with an imperfect diagnostic test",
abstract = "SUMMARY The risk ratio quantifies the risk of disease in a study population relative to a reference population. Standard methods of estimation and testing assume a perfect diagnostic test having sensitivity and specificity of 100{\%}. However, this assumption typically does not hold, and this may invalidate naive estimation and testing for the risk ratio. We propose procedures that control for sensitivity and specificity of the diagnostic test, given the risks are measured by proportions, as it is in cross-sectional studies or studies with fixed follow-up times. These procedures provide an exact unconditional test and confidence interval for the true risk ratio. The methods also cover the case when sensitivity and specificity differ in the two groups (differential misclassification). The resulting test and confidence interval may be useful in epidemiological studies as well as in clinical and vaccine trials. We illustrate the method with real-life examples which demonstrate that ignoring sensitivity and specificity of the diagnostic test may lead to considerable bias in the estimated risk ratio.",
keywords = "exact unconditional test, Key words Exact confidence interval, misclassification, prevalence ratio, relative risk",
author = "J. Reiczigel and J. Singer and Zs Lang",
year = "2017",
month = "1",
day = "1",
doi = "10.1017/S0950268816002028",
language = "English",
volume = "145",
pages = "187--193",
journal = "Epidemiology and Infection",
issn = "0950-2688",
publisher = "Cambridge University Press",
number = "1",

}

TY - JOUR

T1 - Exact inference for the risk ratio with an imperfect diagnostic test

AU - Reiczigel, J.

AU - Singer, J.

AU - Lang, Zs

PY - 2017/1/1

Y1 - 2017/1/1

N2 - SUMMARY The risk ratio quantifies the risk of disease in a study population relative to a reference population. Standard methods of estimation and testing assume a perfect diagnostic test having sensitivity and specificity of 100%. However, this assumption typically does not hold, and this may invalidate naive estimation and testing for the risk ratio. We propose procedures that control for sensitivity and specificity of the diagnostic test, given the risks are measured by proportions, as it is in cross-sectional studies or studies with fixed follow-up times. These procedures provide an exact unconditional test and confidence interval for the true risk ratio. The methods also cover the case when sensitivity and specificity differ in the two groups (differential misclassification). The resulting test and confidence interval may be useful in epidemiological studies as well as in clinical and vaccine trials. We illustrate the method with real-life examples which demonstrate that ignoring sensitivity and specificity of the diagnostic test may lead to considerable bias in the estimated risk ratio.

AB - SUMMARY The risk ratio quantifies the risk of disease in a study population relative to a reference population. Standard methods of estimation and testing assume a perfect diagnostic test having sensitivity and specificity of 100%. However, this assumption typically does not hold, and this may invalidate naive estimation and testing for the risk ratio. We propose procedures that control for sensitivity and specificity of the diagnostic test, given the risks are measured by proportions, as it is in cross-sectional studies or studies with fixed follow-up times. These procedures provide an exact unconditional test and confidence interval for the true risk ratio. The methods also cover the case when sensitivity and specificity differ in the two groups (differential misclassification). The resulting test and confidence interval may be useful in epidemiological studies as well as in clinical and vaccine trials. We illustrate the method with real-life examples which demonstrate that ignoring sensitivity and specificity of the diagnostic test may lead to considerable bias in the estimated risk ratio.

KW - exact unconditional test

KW - Key words Exact confidence interval

KW - misclassification

KW - prevalence ratio

KW - relative risk

UR - http://www.scopus.com/inward/record.url?scp=84986538599&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84986538599&partnerID=8YFLogxK

U2 - 10.1017/S0950268816002028

DO - 10.1017/S0950268816002028

M3 - Article

C2 - 27608542

AN - SCOPUS:84986538599

VL - 145

SP - 187

EP - 193

JO - Epidemiology and Infection

JF - Epidemiology and Infection

SN - 0950-2688

IS - 1

ER -