Are clinical diagnoses of Alzheimer's disease and other dementias affected by education and self-reported race?

Jeanne A. Teresi, Ellen Grober, Joseph P. Eimicke, Amy R. Ehrlich

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

A randomized controlled trial examined whether the diagnostic process for Alzheimer's disease and other dementias may be influenced by knowledge of the patient's education and/or self-reported race. Four conditions were implemented: diagnostic team knows (a) race and education, (b) education only, (c) race only, or (d) neither. Diagnosis and clinical staging was established at baseline, at Wave 2, and for a random sample of Wave 3 respondents by a consensus panel. At study end, a longitudinal, "gold standard" diagnosis was made for patients with follow-up data (71%). Group differences in Diagnostic and Statistical Manual of Mental Disorders (4th ed.; American Psychiatric Association, 1994) diagnosis were estimated using logistic regression and generalized estimating equations. Sensitivity and specificity were examined for baseline diagnosis in relation to the gold standard, longitudinal diagnosis. Despite equivalent status on all measured variables across waves, members of the "knows race only" group were less likely than those of other groups to receive a diagnosis of dementia. At final diagnosis, 19% of the "knows race only" group was diagnosed with dementia versus 38% to 40% in the other 3 conditions (p =.038). Examination of sensitivities and specificities of baseline diagnosis in relation to the gold standard diagnosis showed a nonsignificant trend for lower sensitivities in the knowing race conditions (0.3846), as contrasted with knowing education only (0.480) or neither (0.600). The finding that knowledge of race may influence the diagnostic process in some unknown way is timely, given the recent State-of-the-Science conference on Alzheimer's disease prevention, the authors of which called for information about and standardization of the diagnostic process.

Original languageEnglish (US)
Pages (from-to)531-544
Number of pages14
JournalPsychological Assessment
Volume24
Issue number3
DOIs
StatePublished - Sep 2012

Fingerprint

Dementia
Alzheimer Disease
Education
Sensitivity and Specificity
Patient Education
Diagnostic and Statistical Manual of Mental Disorders
Consensus
Randomized Controlled Trials
Logistic Models

Keywords

  • Alzheimer's disease
  • Bias
  • Clinical diagnosis
  • Education
  • Self-reported race

ASJC Scopus subject areas

  • Psychiatry and Mental health
  • Clinical Psychology

Cite this

Are clinical diagnoses of Alzheimer's disease and other dementias affected by education and self-reported race? / Teresi, Jeanne A.; Grober, Ellen; Eimicke, Joseph P.; Ehrlich, Amy R.

In: Psychological Assessment, Vol. 24, No. 3, 09.2012, p. 531-544.

Research output: Contribution to journalArticle

@article{501c60b7df1749f3a9ccb7928d256241,
title = "Are clinical diagnoses of Alzheimer's disease and other dementias affected by education and self-reported race?",
abstract = "A randomized controlled trial examined whether the diagnostic process for Alzheimer's disease and other dementias may be influenced by knowledge of the patient's education and/or self-reported race. Four conditions were implemented: diagnostic team knows (a) race and education, (b) education only, (c) race only, or (d) neither. Diagnosis and clinical staging was established at baseline, at Wave 2, and for a random sample of Wave 3 respondents by a consensus panel. At study end, a longitudinal, {"}gold standard{"} diagnosis was made for patients with follow-up data (71{\%}). Group differences in Diagnostic and Statistical Manual of Mental Disorders (4th ed.; American Psychiatric Association, 1994) diagnosis were estimated using logistic regression and generalized estimating equations. Sensitivity and specificity were examined for baseline diagnosis in relation to the gold standard, longitudinal diagnosis. Despite equivalent status on all measured variables across waves, members of the {"}knows race only{"} group were less likely than those of other groups to receive a diagnosis of dementia. At final diagnosis, 19{\%} of the {"}knows race only{"} group was diagnosed with dementia versus 38{\%} to 40{\%} in the other 3 conditions (p =.038). Examination of sensitivities and specificities of baseline diagnosis in relation to the gold standard diagnosis showed a nonsignificant trend for lower sensitivities in the knowing race conditions (0.3846), as contrasted with knowing education only (0.480) or neither (0.600). The finding that knowledge of race may influence the diagnostic process in some unknown way is timely, given the recent State-of-the-Science conference on Alzheimer's disease prevention, the authors of which called for information about and standardization of the diagnostic process.",
keywords = "Alzheimer's disease, Bias, Clinical diagnosis, Education, Self-reported race",
author = "Teresi, {Jeanne A.} and Ellen Grober and Eimicke, {Joseph P.} and Ehrlich, {Amy R.}",
year = "2012",
month = "9",
doi = "10.1037/a0027008",
language = "English (US)",
volume = "24",
pages = "531--544",
journal = "Psychological Assessment",
issn = "1040-3590",
publisher = "American Psychological Association Inc.",
number = "3",

}

TY - JOUR

T1 - Are clinical diagnoses of Alzheimer's disease and other dementias affected by education and self-reported race?

AU - Teresi, Jeanne A.

AU - Grober, Ellen

AU - Eimicke, Joseph P.

AU - Ehrlich, Amy R.

PY - 2012/9

Y1 - 2012/9

N2 - A randomized controlled trial examined whether the diagnostic process for Alzheimer's disease and other dementias may be influenced by knowledge of the patient's education and/or self-reported race. Four conditions were implemented: diagnostic team knows (a) race and education, (b) education only, (c) race only, or (d) neither. Diagnosis and clinical staging was established at baseline, at Wave 2, and for a random sample of Wave 3 respondents by a consensus panel. At study end, a longitudinal, "gold standard" diagnosis was made for patients with follow-up data (71%). Group differences in Diagnostic and Statistical Manual of Mental Disorders (4th ed.; American Psychiatric Association, 1994) diagnosis were estimated using logistic regression and generalized estimating equations. Sensitivity and specificity were examined for baseline diagnosis in relation to the gold standard, longitudinal diagnosis. Despite equivalent status on all measured variables across waves, members of the "knows race only" group were less likely than those of other groups to receive a diagnosis of dementia. At final diagnosis, 19% of the "knows race only" group was diagnosed with dementia versus 38% to 40% in the other 3 conditions (p =.038). Examination of sensitivities and specificities of baseline diagnosis in relation to the gold standard diagnosis showed a nonsignificant trend for lower sensitivities in the knowing race conditions (0.3846), as contrasted with knowing education only (0.480) or neither (0.600). The finding that knowledge of race may influence the diagnostic process in some unknown way is timely, given the recent State-of-the-Science conference on Alzheimer's disease prevention, the authors of which called for information about and standardization of the diagnostic process.

AB - A randomized controlled trial examined whether the diagnostic process for Alzheimer's disease and other dementias may be influenced by knowledge of the patient's education and/or self-reported race. Four conditions were implemented: diagnostic team knows (a) race and education, (b) education only, (c) race only, or (d) neither. Diagnosis and clinical staging was established at baseline, at Wave 2, and for a random sample of Wave 3 respondents by a consensus panel. At study end, a longitudinal, "gold standard" diagnosis was made for patients with follow-up data (71%). Group differences in Diagnostic and Statistical Manual of Mental Disorders (4th ed.; American Psychiatric Association, 1994) diagnosis were estimated using logistic regression and generalized estimating equations. Sensitivity and specificity were examined for baseline diagnosis in relation to the gold standard, longitudinal diagnosis. Despite equivalent status on all measured variables across waves, members of the "knows race only" group were less likely than those of other groups to receive a diagnosis of dementia. At final diagnosis, 19% of the "knows race only" group was diagnosed with dementia versus 38% to 40% in the other 3 conditions (p =.038). Examination of sensitivities and specificities of baseline diagnosis in relation to the gold standard diagnosis showed a nonsignificant trend for lower sensitivities in the knowing race conditions (0.3846), as contrasted with knowing education only (0.480) or neither (0.600). The finding that knowledge of race may influence the diagnostic process in some unknown way is timely, given the recent State-of-the-Science conference on Alzheimer's disease prevention, the authors of which called for information about and standardization of the diagnostic process.

KW - Alzheimer's disease

KW - Bias

KW - Clinical diagnosis

KW - Education

KW - Self-reported race

UR - http://www.scopus.com/inward/record.url?scp=84874419801&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84874419801&partnerID=8YFLogxK

U2 - 10.1037/a0027008

DO - 10.1037/a0027008

M3 - Article

VL - 24

SP - 531

EP - 544

JO - Psychological Assessment

JF - Psychological Assessment

SN - 1040-3590

IS - 3

ER -