Alzheimer’s & Dementia: Diagnosis, Assessment & Disease Monitoring (Dec 2019)

Nonlinear Z‐score modeling for improved detection of cognitive abnormality

  • John Kornak,
  • Julie Fields,
  • Walter Kremers,
  • Sara Farmer,
  • Hilary W. Heuer,
  • Leah Forsberg,
  • Danielle Brushaber,
  • Amy Rindels,
  • Hiroko Dodge,
  • Sandra Weintraub,
  • Lilah Besser,
  • Brian Appleby,
  • Yvette Bordelon,
  • Jessica Bove,
  • Patrick Brannelly,
  • Christina Caso,
  • Giovanni Coppola,
  • Reilly Dever,
  • Christina Dheel,
  • Bradford Dickerson,
  • Susan Dickinson,
  • Sophia Dominguez,
  • Kimiko Domoto‐Reilly,
  • Kelley Faber,
  • Jessica Ferrall,
  • Ann Fishman,
  • Jamie Fong,
  • Tatiana Foroud,
  • Ralitza Gavrilova,
  • Deb Gearhart,
  • Behnaz Ghazanfari,
  • Nupur Ghoshal,
  • Jill Goldman,
  • Jonathan Graff‐Radford,
  • Neill Graff‐Radford,
  • Ian M. Grant,
  • Murray Grossman,
  • Dana Haley,
  • John Hsiao,
  • Robin Hsiung,
  • Edward D. Huey,
  • David Irwin,
  • David Jones,
  • Lynne Jones,
  • Kejal Kantarci,
  • Anna Karydas,
  • Daniel Kaufer,
  • Diana Kerwin,
  • David Knopman,
  • Ruth Kraft,
  • Joel Kramer,
  • Walter Kukull,
  • Maria Lapid,
  • Irene Litvan,
  • Peter Ljubenkov,
  • Diane Lucente,
  • Codrin Lungu,
  • Ian Mackenzie,
  • Miranda Maldonado,
  • Masood Manoochehri,
  • Scott McGinnis,
  • Emily McKinley,
  • Mario Mendez,
  • Bruce Miller,
  • Namita Multani,
  • Chiadi Onyike,
  • Jaya Padmanabhan,
  • Alexander Pantelyat,
  • Rodney Pearlman,
  • Len Petrucelli,
  • Madeline Potter,
  • Rosa Rademakers,
  • Eliana Marisa Ramos,
  • Katherine Rankin,
  • Katya Rascovsky,
  • Erik D. Roberson,
  • Emily Rogalski‐Miller,
  • Pheth Sengdy,
  • Les Shaw,
  • Adam M. Staffaroni,
  • Margaret Sutherland,
  • Jeremy Syrjanen,
  • Carmela Tartaglia,
  • Nadine Tatton,
  • Joanne Taylor,
  • Arthur Toga,
  • John Trojanowski,
  • Ping Wang,
  • Bonnie Wong,
  • Zbigniew Wszolek,
  • Brad Boeve,
  • Adam Boxer,
  • Howard Rosen,
  • ARTFL/LEFFTDS Consortium

DOI
https://doi.org/10.1016/j.dadm.2019.08.003
Journal volume & issue
Vol. 11, no. 1
pp. 797 – 808

Abstract

Read online

Abstract 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.

Keywords