Misclassification in more than one factor in a case-control study

A combination of Mantel-Haenszel and maximum likelihood approaches

S. W. Duffy, Thomas E. Rohan, N. E. Day

Research output: Contribution to journalArticle

23 Citations (Scopus)

Abstract

Misclassification can arise in the case-control setting in both the risk factor of interest, and in a potential confounder. If true status is not ascertainable, but an independent replicated sample is available, adjusted maximum likelihood estimates of the individual binomial proportions can be obtained, and hence a relative risk estimate of the Mantel-Haenszel type constructed. An interval estimate on the latter can be calculated using the asymptotic properties of the likelihood function. The method is illustrated by application to a case-control study of breast cancer risk and such lifestyle attributes as diet, smoking and alcohol consumption, in which misclassification was anticipated and consequently repeatability data were collected.

Original languageEnglish (US)
Pages (from-to)1529-1536
Number of pages8
JournalStatistics in Medicine
Volume8
Issue number12
StatePublished - 1989
Externally publishedYes

Fingerprint

Likelihood Functions
Misclassification
Case-control Study
Maximum Likelihood
Case-Control Studies
Case-control
Relative Risk
Smoking
Repeatability
Alcohol
Risk Factors
Likelihood Function
Maximum Likelihood Estimate
Breast Cancer
Alcohol Drinking
Estimate
Asymptotic Properties
Life Style
Proportion
Attribute

ASJC Scopus subject areas

  • Epidemiology

Cite this

Misclassification in more than one factor in a case-control study : A combination of Mantel-Haenszel and maximum likelihood approaches. / Duffy, S. W.; Rohan, Thomas E.; Day, N. E.

In: Statistics in Medicine, Vol. 8, No. 12, 1989, p. 1529-1536.

Research output: Contribution to journalArticle

@article{149eff6ef9d746aa9f7d14deeab1f0b8,
title = "Misclassification in more than one factor in a case-control study: A combination of Mantel-Haenszel and maximum likelihood approaches",
abstract = "Misclassification can arise in the case-control setting in both the risk factor of interest, and in a potential confounder. If true status is not ascertainable, but an independent replicated sample is available, adjusted maximum likelihood estimates of the individual binomial proportions can be obtained, and hence a relative risk estimate of the Mantel-Haenszel type constructed. An interval estimate on the latter can be calculated using the asymptotic properties of the likelihood function. The method is illustrated by application to a case-control study of breast cancer risk and such lifestyle attributes as diet, smoking and alcohol consumption, in which misclassification was anticipated and consequently repeatability data were collected.",
author = "Duffy, {S. W.} and Rohan, {Thomas E.} and Day, {N. E.}",
year = "1989",
language = "English (US)",
volume = "8",
pages = "1529--1536",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "12",

}

TY - JOUR

T1 - Misclassification in more than one factor in a case-control study

T2 - A combination of Mantel-Haenszel and maximum likelihood approaches

AU - Duffy, S. W.

AU - Rohan, Thomas E.

AU - Day, N. E.

PY - 1989

Y1 - 1989

N2 - Misclassification can arise in the case-control setting in both the risk factor of interest, and in a potential confounder. If true status is not ascertainable, but an independent replicated sample is available, adjusted maximum likelihood estimates of the individual binomial proportions can be obtained, and hence a relative risk estimate of the Mantel-Haenszel type constructed. An interval estimate on the latter can be calculated using the asymptotic properties of the likelihood function. The method is illustrated by application to a case-control study of breast cancer risk and such lifestyle attributes as diet, smoking and alcohol consumption, in which misclassification was anticipated and consequently repeatability data were collected.

AB - Misclassification can arise in the case-control setting in both the risk factor of interest, and in a potential confounder. If true status is not ascertainable, but an independent replicated sample is available, adjusted maximum likelihood estimates of the individual binomial proportions can be obtained, and hence a relative risk estimate of the Mantel-Haenszel type constructed. An interval estimate on the latter can be calculated using the asymptotic properties of the likelihood function. The method is illustrated by application to a case-control study of breast cancer risk and such lifestyle attributes as diet, smoking and alcohol consumption, in which misclassification was anticipated and consequently repeatability data were collected.

UR - http://www.scopus.com/inward/record.url?scp=0024801527&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0024801527&partnerID=8YFLogxK

M3 - Article

VL - 8

SP - 1529

EP - 1536

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 12

ER -