TY - JOUR
T1 - Nonlinear Z-score modeling for improved detection of cognitive abnormality
AU - ARTFL/LEFFTDS Consortium
AU - Kornak, John
AU - Fields, Julie
AU - Kremers, Walter
AU - Farmer, Sara
AU - Heuer, Hilary W.
AU - Forsberg, Leah
AU - Brushaber, Danielle
AU - Rindels, Amy
AU - Dodge, Hiroko
AU - Weintraub, Sandra
AU - Besser, Lilah
AU - Appleby, Brian
AU - Bordelon, Yvette
AU - Bove, Jessica
AU - Brannelly, Patrick
AU - Caso, Christina
AU - Coppola, Giovanni
AU - Dever, Reilly
AU - Dheel, Christina
AU - Dickerson, Bradford
AU - Dickinson, Susan
AU - Dominguez, Sophia
AU - Domoto-Reilly, Kimiko
AU - Faber, Kelley
AU - Ferrall, Jessica
AU - Fishman, Ann
AU - Fong, Jamie
AU - Foroud, Tatiana
AU - Gavrilova, Ralitza
AU - Gearhart, Deb
AU - Ghazanfari, Behnaz
AU - Ghoshal, Nupur
AU - Goldman, Jill
AU - Graff-Radford, Jonathan
AU - Graff-Radford, Neill
AU - Grant, Ian M.
AU - Grossman, Murray
AU - Haley, Dana
AU - Hsiao, John
AU - Hsiung, Robin
AU - Huey, Edward D.
AU - Irwin, David
AU - Jones, David
AU - Jones, Lynne
AU - Kantarci, Kejal
AU - Karydas, Anna
AU - Kaufer, Daniel
AU - Kerwin, Diana
AU - Knopman, David
AU - Kraft, Ruth
N1 - Publisher Copyright:
© 2019
PY - 2019/12
Y1 - 2019/12
N2 - Introduction: Conventional Z-scores are generated by subtracting the mean and dividing by the standard deviation. More recent methods linearly correct for age, sex, and education, so that these “adjusted” Z-scores better represent whether an individual's cognitive performance is abnormal. Extreme negative Z-scores for individuals relative to this normative distribution are considered indicative of cognitive deficiency. Methods: In this article, we consider nonlinear shape constrained additive models accounting for age, sex, and education (correcting for nonlinearity). Additional shape constrained additive models account for varying standard deviation of the cognitive scores with age (correcting for heterogeneity of variance). Results: Corrected Z-scores based on nonlinear shape constrained additive models provide improved adjustment for age, sex, and education, as indicated by higher adjusted-R2. Discussion: Nonlinearly corrected Z-scores with respect to age, sex, and education with age-varying residual standard deviation allow for improved detection of non-normative extreme cognitive scores.
AB - Introduction: Conventional Z-scores are generated by subtracting the mean and dividing by the standard deviation. More recent methods linearly correct for age, sex, and education, so that these “adjusted” Z-scores better represent whether an individual's cognitive performance is abnormal. Extreme negative Z-scores for individuals relative to this normative distribution are considered indicative of cognitive deficiency. Methods: In this article, we consider nonlinear shape constrained additive models accounting for age, sex, and education (correcting for nonlinearity). Additional shape constrained additive models account for varying standard deviation of the cognitive scores with age (correcting for heterogeneity of variance). Results: Corrected Z-scores based on nonlinear shape constrained additive models provide improved adjustment for age, sex, and education, as indicated by higher adjusted-R2. Discussion: Nonlinearly corrected Z-scores with respect to age, sex, and education with age-varying residual standard deviation allow for improved detection of non-normative extreme cognitive scores.
KW - Generalized additive models
KW - Heterogenous variance modeling
KW - Neuropsychological testing scores
KW - Nonlinear Z-score correction
KW - Shape constrained additive models
UR - http://www.scopus.com/inward/record.url?scp=85075806346&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85075806346&partnerID=8YFLogxK
U2 - 10.1016/j.dadm.2019.08.003
DO - 10.1016/j.dadm.2019.08.003
M3 - Article
AN - SCOPUS:85075806346
SN - 2352-8729
VL - 11
SP - 797
EP - 808
JO - Alzheimer's and Dementia: Diagnosis, Assessment and Disease Monitoring
JF - Alzheimer's and Dementia: Diagnosis, Assessment and Disease Monitoring
ER -