Partially latent class models for case-control studies of childhood pneumonia aetiology

Research output: Contribution to journalArticle

Abstract

In population studies on the aetiology of disease, one goal is the estimation of the fraction of cases that are attributable to each of several causes. For example, pneumonia is a clinical diagnosis of lung infection that may be caused by viral, bacterial, fungal or other pathogens. The study of pneumonia aetiology is challenging because directly sampling from the lung to identify the aetiologic pathogen is not standard clinical practice in most settings. Instead, measurements from multiple peripheral specimens are made. The paper introduces the statistical methodology designed for estimating the population aetiology distribution and the individual aetiology probabilities in the Pneumonia Etiology Research for Child Health study of 9500 children for seven sites around the world. We formulate the scientific problem in statistical terms as estimating the mixing weights and latent class indicators under a partially latent class model (PLCM) that combines heterogeneous measurements with different error rates obtained from a case-control study. We introduce the PLCM as an extension of the latent class model. We also introduce graphical displays of the population data and inferred latent class frequencies. The methods are tested with simulated data, and then applied to Pneumonia Etiology Research for Child Health data. The paper closes with a brief description of extensions of the PLCM to the regression setting and to the case where conditional independence between the measures is relaxed.

Original languageEnglish (US)
Pages (from-to)97-114
Number of pages18
JournalJournal of the Royal Statistical Society. Series C: Applied Statistics
Volume65
Issue number1
DOIs
StatePublished - Jan 1 2016

Fingerprint

Latent Class Model
Case-control Study
Latent Class
Lung
Health
Graphical Display
Conditional Independence
Infection
Error Rate
Regression
Latent class model
Etiology
Pneumonia
Childhood
Methodology
Term
Children

Keywords

  • Aetiology
  • Bayesian method
  • Case-control
  • Latent class
  • Measurement error
  • Pneumonia

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

@article{549ba39e130c4648b55e4e88947b489f,
title = "Partially latent class models for case-control studies of childhood pneumonia aetiology",
abstract = "In population studies on the aetiology of disease, one goal is the estimation of the fraction of cases that are attributable to each of several causes. For example, pneumonia is a clinical diagnosis of lung infection that may be caused by viral, bacterial, fungal or other pathogens. The study of pneumonia aetiology is challenging because directly sampling from the lung to identify the aetiologic pathogen is not standard clinical practice in most settings. Instead, measurements from multiple peripheral specimens are made. The paper introduces the statistical methodology designed for estimating the population aetiology distribution and the individual aetiology probabilities in the Pneumonia Etiology Research for Child Health study of 9500 children for seven sites around the world. We formulate the scientific problem in statistical terms as estimating the mixing weights and latent class indicators under a partially latent class model (PLCM) that combines heterogeneous measurements with different error rates obtained from a case-control study. We introduce the PLCM as an extension of the latent class model. We also introduce graphical displays of the population data and inferred latent class frequencies. The methods are tested with simulated data, and then applied to Pneumonia Etiology Research for Child Health data. The paper closes with a brief description of extensions of the PLCM to the regression setting and to the case where conditional independence between the measures is relaxed.",
keywords = "Aetiology, Bayesian method, Case-control, Latent class, Measurement error, Pneumonia",
author = "Zhenke Wu and Knoll, {Maria Deloria} and Hammitt, {Laura L} and Scott Zeger",
year = "2016",
month = "1",
day = "1",
doi = "10.1111/rssc.12101",
language = "English (US)",
volume = "65",
pages = "97--114",
journal = "Journal of the Royal Statistical Society. Series C: Applied Statistics",
issn = "0035-9254",
publisher = "Wiley-Blackwell",
number = "1",

}

TY - JOUR

T1 - Partially latent class models for case-control studies of childhood pneumonia aetiology

AU - Wu, Zhenke

AU - Knoll, Maria Deloria

AU - Hammitt, Laura L

AU - Zeger, Scott

PY - 2016/1/1

Y1 - 2016/1/1

N2 - In population studies on the aetiology of disease, one goal is the estimation of the fraction of cases that are attributable to each of several causes. For example, pneumonia is a clinical diagnosis of lung infection that may be caused by viral, bacterial, fungal or other pathogens. The study of pneumonia aetiology is challenging because directly sampling from the lung to identify the aetiologic pathogen is not standard clinical practice in most settings. Instead, measurements from multiple peripheral specimens are made. The paper introduces the statistical methodology designed for estimating the population aetiology distribution and the individual aetiology probabilities in the Pneumonia Etiology Research for Child Health study of 9500 children for seven sites around the world. We formulate the scientific problem in statistical terms as estimating the mixing weights and latent class indicators under a partially latent class model (PLCM) that combines heterogeneous measurements with different error rates obtained from a case-control study. We introduce the PLCM as an extension of the latent class model. We also introduce graphical displays of the population data and inferred latent class frequencies. The methods are tested with simulated data, and then applied to Pneumonia Etiology Research for Child Health data. The paper closes with a brief description of extensions of the PLCM to the regression setting and to the case where conditional independence between the measures is relaxed.

AB - In population studies on the aetiology of disease, one goal is the estimation of the fraction of cases that are attributable to each of several causes. For example, pneumonia is a clinical diagnosis of lung infection that may be caused by viral, bacterial, fungal or other pathogens. The study of pneumonia aetiology is challenging because directly sampling from the lung to identify the aetiologic pathogen is not standard clinical practice in most settings. Instead, measurements from multiple peripheral specimens are made. The paper introduces the statistical methodology designed for estimating the population aetiology distribution and the individual aetiology probabilities in the Pneumonia Etiology Research for Child Health study of 9500 children for seven sites around the world. We formulate the scientific problem in statistical terms as estimating the mixing weights and latent class indicators under a partially latent class model (PLCM) that combines heterogeneous measurements with different error rates obtained from a case-control study. We introduce the PLCM as an extension of the latent class model. We also introduce graphical displays of the population data and inferred latent class frequencies. The methods are tested with simulated data, and then applied to Pneumonia Etiology Research for Child Health data. The paper closes with a brief description of extensions of the PLCM to the regression setting and to the case where conditional independence between the measures is relaxed.

KW - Aetiology

KW - Bayesian method

KW - Case-control

KW - Latent class

KW - Measurement error

KW - Pneumonia

UR - http://www.scopus.com/inward/record.url?scp=84957845518&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84957845518&partnerID=8YFLogxK

U2 - 10.1111/rssc.12101

DO - 10.1111/rssc.12101

M3 - Article

AN - SCOPUS:84957845518

VL - 65

SP - 97

EP - 114

JO - Journal of the Royal Statistical Society. Series C: Applied Statistics

JF - Journal of the Royal Statistical Society. Series C: Applied Statistics

SN - 0035-9254

IS - 1

ER -