INS NYC 2024 Program

Poster	Poster Session 06 Program Schedule 02/15/2024 04:00 pm - 05:15 pm Room: Majestic Complex (Posters 61-120) Poster Session 06: Aging \| MCI \| Neurodegenerative Disease - PART 2 Final Abstract #108 Measuring Ambulatory Cognition in Older Adults at risk for Alzheimer’s Disease Andrea Weinstein, University of Pittsburgh, Pittsburgh, United States Scott Rothenberger, University of Pittsburgh, Pittsburgh, United States Raeanne Moore, University of California, San Diego, San Diego, United States Beth Snitz, University of Pittsburgh, Pittsburgh, United States Ann Cohen, University of Pittsburgh, Pittsburgh, United States Meryl Butters, University of Pittsburgh, Pittsburgh, United States Category: Dementia (Alzheimer's Disease) Keyword 1: dementia - Alzheimer's disease Keyword 2: computerized neuropsychological testing Keyword 3: mild cognitive impairment Objective: Characterizing very early stages of Alzheimer’s disease (AD) is imperative for prevention and treatment efforts, but AD neuropathology develops years prior to impaired neuropsychological test performance. As such, there is a need for cognitive assessment methods and tools that accurately capture very early AD risk. The current study used smartphone-based mobile cognitive tests (MCT) to probe daily cognitive performance. The primary goal was to examine whether mean or variability in MCT cognitive performance was more sensitive to established cognitive assessments. A secondary goal was to examine MCT performance in relation to PET AD biomarkers (amyloid-beta [Aβ], tau). Participants and Methods: Participants were recruited from ongoing University of Pittsburgh studies. Inclusion criteria were: age 65+, able to use a smartphone, absence of severe psychiatric or neurological disorder, completion of parent study NIH Toolbox cognition and PET imaging in past year, CDR<1. The MCT protocol consisted of 4 burst assessments per day over 10 days; data from Memory Matrix (spatial memory) and Color Trick (processing speed and inhibitory control) are presented. NIH Toolbox assessments were used for validated cognitive comparison. Aβ and tau were measured with Pittsburgh compound B (PiB)-PET and AV-1451-PET neuroimaging. MCT task performance was estimated using mixed-effects location scale models, allowing for simultaneous modeling of both the mean and intraindividual variability as functions of covariates. For each MCT task, we modeled performance (mean and/or variability) in relation to NIH Toolbox cognitive performance (Dimensional Change Card Sort, Pattern Matching, Picture Naming, Reading, Sorting) and Aβ and tau as predictors with binary groupings for positivity. Models controlled for age, education, and session day; model fit was assessed using Bayesian Information Criterion (BIC). Results: Forty participants (mean age = 73.6 years, range = 67-91; mean education = 15 years, range = 10-19 years; 21 female; 77% white) completed the protocol. Participants completed 91% (61% to 100%) of all MCT assessments. When comparing MCT mean vs. variability in Color Trick correct trial reaction time (RT), mean performance was more consistently associated with Toolbox performance (3/5 models). For example, Color Trick mean RT was associated with Dimensional Change Card Sort (β=-0.1 [-0.18, -0.03], p=.007, BIC=295.87). Mean and variability in Memory Matrix performance was associated with Dimensional Change Card Sort (mean β=1.47 [0.59-2.36], variability β=-.79 [-1.21- -0.37], p’s<.001, BIC=3856.87). When comparing MCT mean vs. variability in relation to Aβ and tau positivity, within-person variability was more strongly related to PET AD markers than mean performance for both Memory Matrix and Color Trick. For instance, Aβ positivity was associated with greater Color Trick RT variability (β=0.31 [0.01-0.62], p=.044, BIC=278.16). Tau positivity was associated with less variability in Color Trick RT (β=-0.57 [-0.88- -0.25], p<.001, BIC=204.33) and Memory Matrix accuracy (β=-1.61 [-3.19- -0.02], p=.047, BIC=2426.65). Conclusions: MCT measures of cognition are feasible to collect in older adults. Both mean and variability in MCT performance show sensitivity to established measures of cognitive performance and Aβ and tau in nondemented older adults. Ongoing work will measure MCT performance patterns in relation to patterns in daily life activities and cognitive decline (K23AG076663).

Poster

02/15/2024
04:00 pm - 05:15 pm
Room: Majestic Complex (Posters 61-120)

Poster Session 06: Aging | MCI | Neurodegenerative Disease - PART 2

Final Abstract #108

Measuring Ambulatory Cognition in Older Adults at risk for Alzheimer’s Disease

Andrea Weinstein, University of Pittsburgh, Pittsburgh, United States
Scott Rothenberger, University of Pittsburgh, Pittsburgh, United States
Raeanne Moore, University of California, San Diego, San Diego, United States
Beth Snitz, University of Pittsburgh, Pittsburgh, United States
Ann Cohen, University of Pittsburgh, Pittsburgh, United States
Meryl Butters, University of Pittsburgh, Pittsburgh, United States

Category: Dementia (Alzheimer's Disease)

Keyword 1: dementia - Alzheimer's disease
Keyword 2: computerized neuropsychological testing
Keyword 3: mild cognitive impairment

Objective:

Characterizing very early stages of Alzheimer’s disease (AD) is imperative for prevention and treatment efforts, but AD neuropathology develops years prior to impaired neuropsychological test performance. As such, there is a need for cognitive assessment methods and tools that accurately capture very early AD risk. The current study used smartphone-based mobile cognitive tests (MCT) to probe daily cognitive performance. The primary goal was to examine whether mean or variability in MCT cognitive performance was more sensitive to established cognitive assessments. A secondary goal was to examine MCT performance in relation to PET AD biomarkers (amyloid-beta [Aβ], tau).

Participants and Methods:

Participants were recruited from ongoing University of Pittsburgh studies. Inclusion criteria were: age 65+, able to use a smartphone, absence of severe psychiatric or neurological disorder, completion of parent study NIH Toolbox cognition and PET imaging in past year, CDR<1. The MCT protocol consisted of 4 burst assessments per day over 10 days; data from Memory Matrix (spatial memory) and Color Trick (processing speed and inhibitory control) are presented. NIH Toolbox assessments were used for validated cognitive comparison. Aβ and tau were measured with Pittsburgh compound B (PiB)-PET and AV-1451-PET neuroimaging. MCT task performance was estimated using mixed-effects location scale models, allowing for simultaneous modeling of both the mean and intraindividual variability as functions of covariates. For each MCT task, we modeled performance (mean and/or variability) in relation to NIH Toolbox cognitive performance (Dimensional Change Card Sort, Pattern Matching, Picture Naming, Reading, Sorting) and Aβ and tau as predictors with binary groupings for positivity. Models controlled for age, education, and session day; model fit was assessed using Bayesian Information Criterion (BIC).

Results:

Forty participants (mean age = 73.6 years, range = 67-91; mean education = 15 years, range = 10-19 years; 21 female; 77% white) completed the protocol. Participants completed 91% (61% to 100%) of all MCT assessments. When comparing MCT mean vs. variability in Color Trick correct trial reaction time (RT), mean performance was more consistently associated with Toolbox performance (3/5 models). For example, Color Trick mean RT was associated with Dimensional Change Card Sort (β=-0.1 [-0.18, -0.03], p=.007, BIC=295.87). Mean and variability in Memory Matrix performance was associated with Dimensional Change Card Sort (mean β=1.47 [0.59-2.36], variability β=-.79 [-1.21- -0.37], p’s<.001, BIC=3856.87). When comparing MCT mean vs. variability in relation to Aβ and tau positivity, within-person variability was more strongly related to PET AD markers than mean performance for both Memory Matrix and Color Trick. For instance, Aβ positivity was associated with greater Color Trick RT variability (β=0.31 [0.01-0.62], p=.044, BIC=278.16). Tau positivity was associated with less variability in Color Trick RT (β=-0.57 [-0.88- -0.25], p<.001, BIC=204.33) and Memory Matrix accuracy (β=-1.61 [-3.19- -0.02], p=.047, BIC=2426.65).

Conclusions:

MCT measures of cognition are feasible to collect in older adults. Both mean and variability in MCT performance show sensitivity to established measures of cognitive performance and Aβ and tau in nondemented older adults. Ongoing work will measure MCT performance patterns in relation to patterns in daily life activities and cognitive decline (K23AG076663).