Premium
O2‐15‐04: ROBUST INDIVIDUALIZED PREDICTION MODELS WHICH ARE APPLICABLE ACROSS DIFFERENT COHORTS
Author(s) -
Maurik Ingrid S.,
Vos Stephanie J.B.,
Bos Isabelle,
Bouwman Femke H.,
Tijms Betty M.,
Teunissen Charlotte E.,
Barkhof Frederik,
Scheltens Philip,
Palmqvist Sebastian,
Hansson Oskar,
Visser Pieter Jelle,
Berkhof Johannes,
Flier Wiesje M.
Publication year - 2018
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2018.06.2727
Subject(s) - cohort , generalizability theory , medicine , proportional hazards model , neuroimaging , dementia , alzheimer's disease neuroimaging initiative , oncology , cohort study , disease , psychology , psychiatry , developmental psychology
impairment (Figure 2), but better longitudinal prediction is provided when age is included in the predictive models (p<.018). A/ T/N group did not predict slope of scores on the psychometric, depression, and neuropsychiatric tests over time. Conclusions:The A/T/N system predicts time to onset of cognitive impairment, and its’ predictive ability is improved by consideration of participant age. Cross-sectional results support those previously reported by others.