On the potential for illogic with logically defined outcomes

Research output: Contribution to journalArticle

Abstract

Logically defined outcomes are commonly used in medical diagnoses and epidemiological research. When missing values in the original outcomes exist, the method of handling the missingness can have unintended consequences, even if the original outcomes are missing completely at random. In this note, we consider 2 binary original outcomes, which are missing completely at random. For estimating the prevalence of a logically defined "or" outcome, we discuss the properties of 4 estimators: the complete-case estimator, the available-case estimator, the maximum likelihood estimator (MLE), and a moment-based estimator. With the exception of the available-case case estimator, all the estimators are consistent. The MLE exhibits superior performance and should be generally adopted.

Original languageEnglish (US)
Pages (from-to)800-804
Number of pages5
JournalBiostatistics
Volume8
Issue number4
DOIs
StatePublished - Oct 2007

Fingerprint

Estimator
Research
Missing Completely at Random
Maximum Likelihood Estimator
Binary Outcomes
Missing Values
Exception
Moment
Maximum likelihood estimator

Keywords

  • Available-case estimator
  • Complete-case estimator
  • Hypertension
  • Maximum likelihood estimator
  • Missing data
  • Moment-based estimator

ASJC Scopus subject areas

  • Medicine(all)
  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

On the potential for illogic with logically defined outcomes. / Li, Xianbin; Caffo, Brian S; Scharfstein, Daniel O.

In: Biostatistics, Vol. 8, No. 4, 10.2007, p. 800-804.

Research output: Contribution to journalArticle

@article{26c7f2b1413d44ffa335608203eefa26,
title = "On the potential for illogic with logically defined outcomes",
abstract = "Logically defined outcomes are commonly used in medical diagnoses and epidemiological research. When missing values in the original outcomes exist, the method of handling the missingness can have unintended consequences, even if the original outcomes are missing completely at random. In this note, we consider 2 binary original outcomes, which are missing completely at random. For estimating the prevalence of a logically defined {"}or{"} outcome, we discuss the properties of 4 estimators: the complete-case estimator, the available-case estimator, the maximum likelihood estimator (MLE), and a moment-based estimator. With the exception of the available-case case estimator, all the estimators are consistent. The MLE exhibits superior performance and should be generally adopted.",
keywords = "Available-case estimator, Complete-case estimator, Hypertension, Maximum likelihood estimator, Missing data, Moment-based estimator",
author = "Xianbin Li and Caffo, {Brian S} and Scharfstein, {Daniel O}",
year = "2007",
month = "10",
doi = "10.1093/biostatistics/kxm006",
language = "English (US)",
volume = "8",
pages = "800--804",
journal = "Biostatistics",
issn = "1465-4644",
publisher = "Oxford University Press",
number = "4",

}

TY - JOUR

T1 - On the potential for illogic with logically defined outcomes

AU - Li, Xianbin

AU - Caffo, Brian S

AU - Scharfstein, Daniel O

PY - 2007/10

Y1 - 2007/10

N2 - Logically defined outcomes are commonly used in medical diagnoses and epidemiological research. When missing values in the original outcomes exist, the method of handling the missingness can have unintended consequences, even if the original outcomes are missing completely at random. In this note, we consider 2 binary original outcomes, which are missing completely at random. For estimating the prevalence of a logically defined "or" outcome, we discuss the properties of 4 estimators: the complete-case estimator, the available-case estimator, the maximum likelihood estimator (MLE), and a moment-based estimator. With the exception of the available-case case estimator, all the estimators are consistent. The MLE exhibits superior performance and should be generally adopted.

AB - Logically defined outcomes are commonly used in medical diagnoses and epidemiological research. When missing values in the original outcomes exist, the method of handling the missingness can have unintended consequences, even if the original outcomes are missing completely at random. In this note, we consider 2 binary original outcomes, which are missing completely at random. For estimating the prevalence of a logically defined "or" outcome, we discuss the properties of 4 estimators: the complete-case estimator, the available-case estimator, the maximum likelihood estimator (MLE), and a moment-based estimator. With the exception of the available-case case estimator, all the estimators are consistent. The MLE exhibits superior performance and should be generally adopted.

KW - Available-case estimator

KW - Complete-case estimator

KW - Hypertension

KW - Maximum likelihood estimator

KW - Missing data

KW - Moment-based estimator

UR - http://www.scopus.com/inward/record.url?scp=34848818788&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34848818788&partnerID=8YFLogxK

U2 - 10.1093/biostatistics/kxm006

DO - 10.1093/biostatistics/kxm006

M3 - Article

C2 - 17384078

AN - SCOPUS:34848818788

VL - 8

SP - 800

EP - 804

JO - Biostatistics

JF - Biostatistics

SN - 1465-4644

IS - 4

ER -