Random effects models for assessing diagnostic accuracy of traditional Chinese doctors in absence of a gold standard

Zheyu Wang, Xiao Hua Zhou

Research output: Contribution to journalArticle

Abstract

Two common problems in assessing the accuracy of traditional Chinese medicine (TCM) doctors in detecting a particular symptom are the unknown true symptom status and the ordinal-scale of the symptom status. Wang et al. (Biostatistics 2011; DOI: 10.1093/biostatistics/kxq075) proposed a nonparametric maximum likelihood method for estimating the accuracy of different TCM doctors without a gold standard when the true symptom status is measured on an ordinal-scale. A key assumption of their work is that the diagnosis results are independent conditional on the gold standard. This assumption can be violated in many practical situations. In this paper, we propose a random effects modeling approach that extends their method to incorporate dependence structure among different tests or doctors. The proposed method is illustrated on a real data set from TCM, which contains the diagnostic results from five doctors for the same patients regarding symptoms related to Chills disease. The same data set was analyzed by Wang et al. under the conditional independence assumption. In addition, we also discuss an ad hoc test for the model fitting and a likelihood ratio test on the random effects.

Original languageEnglish (US)
Pages (from-to)661-671
Number of pages11
JournalStatistics in Medicine
Volume31
Issue number7
DOIs
StatePublished - Mar 30 2012
Externally publishedYes

Fingerprint

Traditional Chinese Medicine
Diagnostic Accuracy
Random Effects Model
Chinese Traditional Medicine
Biostatistics
Gold
Ordinal Scale
Random Effects
Nonparametric Maximum Likelihood
Chills
Conditional Independence
Model Fitting
Dependence Structure
Maximum Likelihood Method
Likelihood Ratio Test
Diagnostics
Unknown
Modeling
Standards
Datasets

Keywords

  • Diagnostic test
  • EM algorithm
  • Random effects models
  • Repeated ordinal data
  • Traditional Chinese medicine (TCM)

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Random effects models for assessing diagnostic accuracy of traditional Chinese doctors in absence of a gold standard. / Wang, Zheyu; Zhou, Xiao Hua.

In: Statistics in Medicine, Vol. 31, No. 7, 30.03.2012, p. 661-671.

Research output: Contribution to journalArticle

@article{207d12ccc4ad48c09aa44f5763386757,
title = "Random effects models for assessing diagnostic accuracy of traditional Chinese doctors in absence of a gold standard",
abstract = "Two common problems in assessing the accuracy of traditional Chinese medicine (TCM) doctors in detecting a particular symptom are the unknown true symptom status and the ordinal-scale of the symptom status. Wang et al. (Biostatistics 2011; DOI: 10.1093/biostatistics/kxq075) proposed a nonparametric maximum likelihood method for estimating the accuracy of different TCM doctors without a gold standard when the true symptom status is measured on an ordinal-scale. A key assumption of their work is that the diagnosis results are independent conditional on the gold standard. This assumption can be violated in many practical situations. In this paper, we propose a random effects modeling approach that extends their method to incorporate dependence structure among different tests or doctors. The proposed method is illustrated on a real data set from TCM, which contains the diagnostic results from five doctors for the same patients regarding symptoms related to Chills disease. The same data set was analyzed by Wang et al. under the conditional independence assumption. In addition, we also discuss an ad hoc test for the model fitting and a likelihood ratio test on the random effects.",
keywords = "Diagnostic test, EM algorithm, Random effects models, Repeated ordinal data, Traditional Chinese medicine (TCM)",
author = "Zheyu Wang and Zhou, {Xiao Hua}",
year = "2012",
month = "3",
day = "30",
doi = "10.1002/sim.4275",
language = "English (US)",
volume = "31",
pages = "661--671",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "7",

}

TY - JOUR

T1 - Random effects models for assessing diagnostic accuracy of traditional Chinese doctors in absence of a gold standard

AU - Wang, Zheyu

AU - Zhou, Xiao Hua

PY - 2012/3/30

Y1 - 2012/3/30

N2 - Two common problems in assessing the accuracy of traditional Chinese medicine (TCM) doctors in detecting a particular symptom are the unknown true symptom status and the ordinal-scale of the symptom status. Wang et al. (Biostatistics 2011; DOI: 10.1093/biostatistics/kxq075) proposed a nonparametric maximum likelihood method for estimating the accuracy of different TCM doctors without a gold standard when the true symptom status is measured on an ordinal-scale. A key assumption of their work is that the diagnosis results are independent conditional on the gold standard. This assumption can be violated in many practical situations. In this paper, we propose a random effects modeling approach that extends their method to incorporate dependence structure among different tests or doctors. The proposed method is illustrated on a real data set from TCM, which contains the diagnostic results from five doctors for the same patients regarding symptoms related to Chills disease. The same data set was analyzed by Wang et al. under the conditional independence assumption. In addition, we also discuss an ad hoc test for the model fitting and a likelihood ratio test on the random effects.

AB - Two common problems in assessing the accuracy of traditional Chinese medicine (TCM) doctors in detecting a particular symptom are the unknown true symptom status and the ordinal-scale of the symptom status. Wang et al. (Biostatistics 2011; DOI: 10.1093/biostatistics/kxq075) proposed a nonparametric maximum likelihood method for estimating the accuracy of different TCM doctors without a gold standard when the true symptom status is measured on an ordinal-scale. A key assumption of their work is that the diagnosis results are independent conditional on the gold standard. This assumption can be violated in many practical situations. In this paper, we propose a random effects modeling approach that extends their method to incorporate dependence structure among different tests or doctors. The proposed method is illustrated on a real data set from TCM, which contains the diagnostic results from five doctors for the same patients regarding symptoms related to Chills disease. The same data set was analyzed by Wang et al. under the conditional independence assumption. In addition, we also discuss an ad hoc test for the model fitting and a likelihood ratio test on the random effects.

KW - Diagnostic test

KW - EM algorithm

KW - Random effects models

KW - Repeated ordinal data

KW - Traditional Chinese medicine (TCM)

UR - http://www.scopus.com/inward/record.url?scp=84858007332&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84858007332&partnerID=8YFLogxK

U2 - 10.1002/sim.4275

DO - 10.1002/sim.4275

M3 - Article

C2 - 21626532

AN - SCOPUS:84858007332

VL - 31

SP - 661

EP - 671

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 7

ER -