A latent variable model approach for assembling and scoring screening tests for dementia

Jianzhao Shen, Paul Crane, Sujuan Gao

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

In dementia studies, the diagnosis of dementia often relies on results of screening tests aimed at measuring various dimensions of cognitive functions. The current practice of scoring a screening test involves simply summing the correct responses from each item. However, this method may be imprecise and inefficient in the predictive power of the score for dementia. We propose a latent variable model approach for the scoring and item selection of such tests. We model the item responses to be random variables based on latent variables. We also model the disease outcomes to be a function of the latent variables. Maximum likelihood estimates are obtained by maximizing the joint likelihood functions of disease and the item responses over a specified distribution function for the latent variables. Variances of model parameters are estimated using a nonparametric bootstrap method. We illustrate the approach using a screening test for dementia from a community-based study.

Original languageEnglish
Pages (from-to)399-407
Number of pages9
JournalJournal of Alzheimer's Disease
Volume5
Issue number5
StatePublished - Oct 2003

Fingerprint

Dementia
Likelihood Functions
Cognition
Joints

Keywords

  • Dementia
  • Item selection and scoring
  • Latent variable model

ASJC Scopus subject areas

  • Neuropsychology and Physiological Psychology

Cite this

A latent variable model approach for assembling and scoring screening tests for dementia. / Shen, Jianzhao; Crane, Paul; Gao, Sujuan.

In: Journal of Alzheimer's Disease, Vol. 5, No. 5, 10.2003, p. 399-407.

Research output: Contribution to journalArticle

@article{b6f6c740d49541119ca157f0b2c7d140,
title = "A latent variable model approach for assembling and scoring screening tests for dementia",
abstract = "In dementia studies, the diagnosis of dementia often relies on results of screening tests aimed at measuring various dimensions of cognitive functions. The current practice of scoring a screening test involves simply summing the correct responses from each item. However, this method may be imprecise and inefficient in the predictive power of the score for dementia. We propose a latent variable model approach for the scoring and item selection of such tests. We model the item responses to be random variables based on latent variables. We also model the disease outcomes to be a function of the latent variables. Maximum likelihood estimates are obtained by maximizing the joint likelihood functions of disease and the item responses over a specified distribution function for the latent variables. Variances of model parameters are estimated using a nonparametric bootstrap method. We illustrate the approach using a screening test for dementia from a community-based study.",
keywords = "Dementia, Item selection and scoring, Latent variable model",
author = "Jianzhao Shen and Paul Crane and Sujuan Gao",
year = "2003",
month = "10",
language = "English",
volume = "5",
pages = "399--407",
journal = "Journal of Alzheimer's Disease",
issn = "1387-2877",
publisher = "IOS Press",
number = "5",

}

TY - JOUR

T1 - A latent variable model approach for assembling and scoring screening tests for dementia

AU - Shen, Jianzhao

AU - Crane, Paul

AU - Gao, Sujuan

PY - 2003/10

Y1 - 2003/10

N2 - In dementia studies, the diagnosis of dementia often relies on results of screening tests aimed at measuring various dimensions of cognitive functions. The current practice of scoring a screening test involves simply summing the correct responses from each item. However, this method may be imprecise and inefficient in the predictive power of the score for dementia. We propose a latent variable model approach for the scoring and item selection of such tests. We model the item responses to be random variables based on latent variables. We also model the disease outcomes to be a function of the latent variables. Maximum likelihood estimates are obtained by maximizing the joint likelihood functions of disease and the item responses over a specified distribution function for the latent variables. Variances of model parameters are estimated using a nonparametric bootstrap method. We illustrate the approach using a screening test for dementia from a community-based study.

AB - In dementia studies, the diagnosis of dementia often relies on results of screening tests aimed at measuring various dimensions of cognitive functions. The current practice of scoring a screening test involves simply summing the correct responses from each item. However, this method may be imprecise and inefficient in the predictive power of the score for dementia. We propose a latent variable model approach for the scoring and item selection of such tests. We model the item responses to be random variables based on latent variables. We also model the disease outcomes to be a function of the latent variables. Maximum likelihood estimates are obtained by maximizing the joint likelihood functions of disease and the item responses over a specified distribution function for the latent variables. Variances of model parameters are estimated using a nonparametric bootstrap method. We illustrate the approach using a screening test for dementia from a community-based study.

KW - Dementia

KW - Item selection and scoring

KW - Latent variable model

UR - http://www.scopus.com/inward/record.url?scp=0345304262&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0345304262&partnerID=8YFLogxK

M3 - Article

VL - 5

SP - 399

EP - 407

JO - Journal of Alzheimer's Disease

JF - Journal of Alzheimer's Disease

SN - 1387-2877

IS - 5

ER -