Frequentist and Bayesian approaches to prevalence estimation using examples from Johne's disease.

Locksley L. McV Messam, Adam J. Branscum, Michael T. Collins, Ian Gardner

Research output: Contribution to journalReview article

41 Citations (Scopus)

Abstract

Although frequentist approaches to prevalence estimation are simple to apply, there are circumstances where it is difficult to satisfy assumptions of asymptotic normality and nonsensical point estimates (greater than 1 or less than 0) may result. This is particularly true when sample sizes are small, test prevalences are low and imperfect sensitivity and specificity of diagnostic tests need to be incorporated into calculations of true prevalence. Bayesian approaches offer several advantages including direct computation of range-respecting interval estimates (e.g. intervals between 0 and 1 for prevalence) without the requirement of transformations or large-sample approximations, direct probabilistic interpretation, and the flexibility to model in a straightforward manner the probability of zero prevalence. In this review, we present frequentist and Bayesian methods for animal- and herd-level true prevalence estimation based on individual and pooled samples. We provide statistical methods for detecting differences between population prevalence and frequentist methods for sample size and power calculations. All examples are motivated using Mycobacterium avium subspecies paratuberculosis infection and we provide WinBUGS code for all examples of Bayesian estimation.

Original languageEnglish (US)
Pages (from-to)1-23
Number of pages23
JournalAnimal health research reviews / Conference of Research Workers in Animal Diseases
Volume9
Issue number1
DOIs
StatePublished - Jan 1 2008

Fingerprint

paratuberculosis
sampling
Mycobacterium avium
Bayesian theory
diagnostic techniques
statistical analysis
herds
infection
animals

ASJC Scopus subject areas

  • Animal Science and Zoology

Cite this

Frequentist and Bayesian approaches to prevalence estimation using examples from Johne's disease. / McV Messam, Locksley L.; Branscum, Adam J.; Collins, Michael T.; Gardner, Ian.

In: Animal health research reviews / Conference of Research Workers in Animal Diseases, Vol. 9, No. 1, 01.01.2008, p. 1-23.

Research output: Contribution to journalReview article

@article{a9854521e3a04320af7431ba17d68521,
title = "Frequentist and Bayesian approaches to prevalence estimation using examples from Johne's disease.",
abstract = "Although frequentist approaches to prevalence estimation are simple to apply, there are circumstances where it is difficult to satisfy assumptions of asymptotic normality and nonsensical point estimates (greater than 1 or less than 0) may result. This is particularly true when sample sizes are small, test prevalences are low and imperfect sensitivity and specificity of diagnostic tests need to be incorporated into calculations of true prevalence. Bayesian approaches offer several advantages including direct computation of range-respecting interval estimates (e.g. intervals between 0 and 1 for prevalence) without the requirement of transformations or large-sample approximations, direct probabilistic interpretation, and the flexibility to model in a straightforward manner the probability of zero prevalence. In this review, we present frequentist and Bayesian methods for animal- and herd-level true prevalence estimation based on individual and pooled samples. We provide statistical methods for detecting differences between population prevalence and frequentist methods for sample size and power calculations. All examples are motivated using Mycobacterium avium subspecies paratuberculosis infection and we provide WinBUGS code for all examples of Bayesian estimation.",
author = "{McV Messam}, {Locksley L.} and Branscum, {Adam J.} and Collins, {Michael T.} and Ian Gardner",
year = "2008",
month = "1",
day = "1",
doi = "10.1017/S1466252307001314",
language = "English (US)",
volume = "9",
pages = "1--23",
journal = "Animal Health Research Reviews",
issn = "1466-2523",
publisher = "Cambridge University Press",
number = "1",

}

TY - JOUR

T1 - Frequentist and Bayesian approaches to prevalence estimation using examples from Johne's disease.

AU - McV Messam, Locksley L.

AU - Branscum, Adam J.

AU - Collins, Michael T.

AU - Gardner, Ian

PY - 2008/1/1

Y1 - 2008/1/1

N2 - Although frequentist approaches to prevalence estimation are simple to apply, there are circumstances where it is difficult to satisfy assumptions of asymptotic normality and nonsensical point estimates (greater than 1 or less than 0) may result. This is particularly true when sample sizes are small, test prevalences are low and imperfect sensitivity and specificity of diagnostic tests need to be incorporated into calculations of true prevalence. Bayesian approaches offer several advantages including direct computation of range-respecting interval estimates (e.g. intervals between 0 and 1 for prevalence) without the requirement of transformations or large-sample approximations, direct probabilistic interpretation, and the flexibility to model in a straightforward manner the probability of zero prevalence. In this review, we present frequentist and Bayesian methods for animal- and herd-level true prevalence estimation based on individual and pooled samples. We provide statistical methods for detecting differences between population prevalence and frequentist methods for sample size and power calculations. All examples are motivated using Mycobacterium avium subspecies paratuberculosis infection and we provide WinBUGS code for all examples of Bayesian estimation.

AB - Although frequentist approaches to prevalence estimation are simple to apply, there are circumstances where it is difficult to satisfy assumptions of asymptotic normality and nonsensical point estimates (greater than 1 or less than 0) may result. This is particularly true when sample sizes are small, test prevalences are low and imperfect sensitivity and specificity of diagnostic tests need to be incorporated into calculations of true prevalence. Bayesian approaches offer several advantages including direct computation of range-respecting interval estimates (e.g. intervals between 0 and 1 for prevalence) without the requirement of transformations or large-sample approximations, direct probabilistic interpretation, and the flexibility to model in a straightforward manner the probability of zero prevalence. In this review, we present frequentist and Bayesian methods for animal- and herd-level true prevalence estimation based on individual and pooled samples. We provide statistical methods for detecting differences between population prevalence and frequentist methods for sample size and power calculations. All examples are motivated using Mycobacterium avium subspecies paratuberculosis infection and we provide WinBUGS code for all examples of Bayesian estimation.

UR - http://www.scopus.com/inward/record.url?scp=50549083791&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=50549083791&partnerID=8YFLogxK

U2 - 10.1017/S1466252307001314

DO - 10.1017/S1466252307001314

M3 - Review article

C2 - 18346298

AN - SCOPUS:50549083791

VL - 9

SP - 1

EP - 23

JO - Animal Health Research Reviews

JF - Animal Health Research Reviews

SN - 1466-2523

IS - 1

ER -