Premium
P2‐468: DEVELOPMENT OF A CONSENSUS ROADMAP TO HARMONIZE COGNITIVE ASSESSMENT
Author(s) -
Boccardi Marina,
Monsch Andreas U.,
Cappa Stefano,
Dubois Bruno,
Salmon David P.,
Brodaty Henry,
Georges Jean,
Winblad Bengt,
Frisoni Giovanni B.,
Kliegel Matthias
Publication year - 2019
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.2019.06.2875
Subject(s) - harmonization , cognition , psychology , cognitive test , cognitive decline , normative , medicine , dementia , political science , disease , psychiatry , physics , pathology , acoustics , law
Background: The pFCSRT+IR is widely used to identify prevalent dementia, incident dementia and AD, and amnestic mild cognitive impairment. Our goal was to evaluate and compare the predictive validity of learning and retention measures from the pFCSRT+IR for incident AD in the Baltimore Study of Aging (BLSA).Methods: Learning was defined by the sum of free recall (sumFR) over the three test trials (max1⁄448) and retention was defined by delayed free recall (DFR) tested 15-20 minutes later (max1⁄416). Analysis was restricted to AD onset within 10 years of baseline pFCSRT+IR. Cox proportional hazards models were used to evaluate the effect of baseline learning and retention on risk of AD adjusting for age, gender and education. Time-dependent receiver operating characteristics (ROC) analysis was used to assess the predictive accuracy of learning and retention for AD at follow-up time 3, 5 and 7 years from baseline. Youden’s index was used to select optimal cut scores. Results: The sample, which included 1455 participants without dementia, was primarily Caucasian (80%), had a mean age of 69.8 years, 45.9% were female, mean education was 16.7 years, and average follow-up time was 7.2 years. The mean (SD) scores for sumFR was 32.1 (5.4) and 11.9 (2.3) for DFR (r1⁄40.67 between measures). Fifty-nine participants developed AD during follow-up. For each SD increase in sumFR, risk of AD decreased (HR1⁄40.49, CI: 0.39-0.62); for each SD increase in DFR, AD risk also decreased (HR1⁄40.54, CI: 0.44-0.68) after adjusting for age, education and gender. When both sumFR and DFR were included in the same model, only the effect of sumFR (HR1⁄40.57, CI: 0.41-0.79) remained while DFR was attenuated (HR1⁄40.92, CI: 0.60-1.13). Areas under the ROC curve (AUC) at 3, 5 and 7 years were 0.83, 0.85 and 0.81, for sumFR, and 0.76, 0.80 and 0.72 for delayed FR; differences were not significant. The optimal cut points for sumFR were 28 at year 3 (sensitivity 0.70, specificity 0.84) and year 5 (sensitivity 0.72, specificity 0.85) and 29 at year 7 (sensitivity 0.67, specificity 0.82). Conclusions: Both learning and retention predict incident AD. In models including both, only learning remains as an independent predictor.