Document Detail

Misuse of the linear mixed model when evaluating risk factors of cognitive decline.
MedLine Citation:
PMID:  21965187     Owner:  NLM     Status:  MEDLINE    
The linear mixed model (LMM), which is routinely used to describe change in outcomes over time and its association with risk factors, assumes that a unit change in any predictor is associated with a constant change in the outcome. When it is used on psychometric tests, this assumption may not hold. Indeed, psychometric tests usually suffer from ceiling and/or floor effects and curvilinearity (i.e., varying sensitivity to change). The authors aimed to determine the consequences of such misspecification when evaluating predictors of cognitive decline. As an alternative to the LMM, they considered 2 mixed models based on latent processes that handle discrete and bounded outcomes. Model differences are illustrated here using data on 4 psychometric tests from the Personnes Agées QUID (PAQUID) Study (1989-2004). The type I error of the Wald test for risk-factor regression parameters was formally assessed in a simulation study. It demonstrated that type I errors in the LMM could be dramatically inflated for some tests, such that spurious associations with risk factors were found. In particular, confusion between effects on mean level and effects on change over time was highlighted. The authors recommend use of the alternative mixed models when studying psychometric tests and more generally quantitative scales (quality of life, activities of daily living).
Cécile Proust-Lima; Jean-François Dartigues; Hélène Jacqmin-Gadda
Related Documents :
23632257 - An update to mitotool: using a new scoring system for faster mtdna haplogroup determina...
24808297 - Visual analytics for converging-business-ecosystem intelligence.
22120027 - A multidisciplinary cross-cultural measurement of functioning after stroke: rasch analy...
22136797 - Correlation and discriminant analysis between clinical, endoscopic, thoracic x-ray and ...
22064797 - Constrained non-negative matrix factorization for image representation.
21406057 - Challenges for the development of new treatments for severe asthma: a pharmaceutical pe...
Publication Detail:
Type:  Journal Article; Research Support, Non-U.S. Gov't     Date:  2011-09-30
Journal Detail:
Title:  American journal of epidemiology     Volume:  174     ISSN:  1476-6256     ISO Abbreviation:  Am. J. Epidemiol.     Publication Date:  2011 Nov 
Date Detail:
Created Date:  2011-10-24     Completed Date:  2011-12-13     Revised Date:  2013-06-27    
Medline Journal Info:
Nlm Unique ID:  7910653     Medline TA:  Am J Epidemiol     Country:  United States    
Other Details:
Languages:  eng     Pagination:  1077-88     Citation Subset:  IM    
Institut de Sante´ Publique, d’E´ pide´miologie et de De´veloppement, Universite´ Bordeaux Segalen, 146 rue Le´o Saignat, 33076 Bordeaux Cedex, France.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Cognition Disorders / epidemiology*,  etiology
Dementia / epidemiology,  etiology
Educational Status
France / epidemiology
Linear Models*
Longitudinal Studies
Models, Statistical
Neuropsychological Tests
Risk Factors

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine

Previous Document:  Proximity to food establishments and body mass index in the Framingham Heart Study offspring cohort ...
Next Document:  The Long Head of the Biceps Tendon Has Minimal Effect on In Vivo Glenohumeral Kinematics: A Biplane ...