Search

Background: In late‐life, multiple chronic conditions (MCCs) has been associated with cognitive decline and Alzheimer's Disease and related dementias. However, this relationship has not been investigated in a middle‐aged population, a time when MCCs first begin to develop, or in rural and low socioeconomic settings, where individuals are at a higher risk of MCCs. We examined the relationship between MCCs and midlife cognitive function among Black and White men and women from a rural, community‐based cohort in Bogalusa, Louisiana. Methods: This cross‐sectional study includes 1,223 Black (33%) and White (67%) individuals (mean age 48.10 ± 5.26; 59% female) from the Bogalusa Heart Study with a complete neuropsychological (NP) assessment, and no history of stroke. Six chronic conditions were selected from the CMS (Centers for Medicare & Medicaid Services) Data Warehouse, and defined using self‐reported data, medication use, and physiological/laboratory measures (Table 1). MCCs was defined as the coexistence of ≥2 conditions. NP performance was measured by global cognitive z‐scores (GCS) and by domain specific z‐scores for attention and processing, episodic memory, and executive function. All z‐scores were standardized for age, sex, and race. Regression models assessed the association between MCCs and cognitive outcomes, adjusting for alcohol consumption, smoking status, and education. Results: A total of 66% participants had MCCs; 44.4% (n = 543) had 2 CCs, 17.8% (n = 218) had 3 CCs, and 3.8% (n = 46) had 4+ CCs. Compared to no chronic conditions, having 2,3, and ≥4 CCs was significantly associated with poorer executive function z‐scores after adjusting for covariates (all p's < 0.05). Having three MCCs had the worst impact on executive function z‐scores (‐0.88(.20), p <0.001) (Table 2). The most prevalent conditions in this group were hypertension, dyslipidemia, and diabetes which co‐occurred in 87.7% (n = 165) of participants with three CCs. Conclusion: This study suggests that the presence of MCCs is associated with lower executive function performance as early as midlife. Hypertension and dyslipidemia commonly occur together and are particularly prevalent in midlife. These findings highlight the importance of addressing chronic conditions earlier in life, a less studied and critical time window for potential prevention of adverse cognitive outcomes in late life. [ABSTRACT FROM AUTHOR]

Background: Early detection of Mild Cognitive Impairment (MCI) and Alzheimer's Disease and Related Dementias (ADRD) is key to optimal management. Digital tests allow for the reliable capture of meaningful, process‐based neuropsychological test parameters that traditional paper/pencil measures fail to detect, thus enabling earlier detection. We assessed how subtle functional decline relates to process features extracted from the digital Clock Drawing Test (dCDT) and the digital Trail Making Test‐Part B (dTrB). Methods: 629 community dwelling participants (58.5% female) were assessed with the Functional Assessment Questionnaire (FAQ), the MMSE, the digital clock drawing test (dCDT) and the digital Trail Making Test‐Part B (dTrB). Three FAQ groups were constructed: FAQ = 0‐2 (minimal functional disability, n = 490); FAQ = 3‐5 (subtle functional disability, n = 90); FAQ = 6‐8 (mild functional disability, n = 49). Results: Participants with minimal functional disability were younger (p = 0.027) and scored better on the MMSE (p = 0.001) than other groups. Groups were equated for education. dCDT command/copy component placement was more accurate among participants with minimal functional disability compared to other groups (command dCDT p = 0.001 versus mild functional disability; copy dCDT p = 0.006 versus both subtle and mild functional disability). On the dTrB, participants with minimal disability drew faster than those with subtle or mild disability (mean stroke velocity, p = 0.001). Analyses for percent pen lift time found no difference between participants with minimal versus subtle disability; however, both groups differed from participants with mild disability (p = 0.001). Optimal group classification using stepwise nominal regression with the minimal functional disability group as the reference was achieved using dCDT copy component placement and dTrB mean stroke velocity (X2 = 28.37, p = 0.001), where dTrB mean velocity entered first (X2 = 17.73, p = 0.001) followed by dCDT copy component placement (X2 = 11.19, p = 0.004). Both variables contributed comparably to group classification (Wald statistic range 5.82‐8.88, p = 0.012, respectively). Conclusions: The digital assessment described above, targeting executive, scanning, visuoconstructional and motor domains, captures highly nuanced, discrete behaviors with precision/ operationalism heretofore unattainable. Consequently, it is reasonable to posit that digital cognitive assessments leveraging the Boston Process Approach can both objectively detect emergent cognitive impairment and infer its likely impact on functional activities, critical for practical assessment in clinical care. [ABSTRACT FROM AUTHOR]

Background: Poor diet has been associated with insidious onset dementia. The Alternate Healthy Eating Index (AHEI) can capture overall nutrient quality and individual/population eating behaviors. Higher adherence to AHEI has been associated with a lower risk of incident dementia and better cognitive function in late life. However, most evidence comes from high‐income populations from urban areas. Here, we examined the adherence to AHEI and its association with cognitive function in midlife among Black and White individuals from a community cohort in the rural, low SES area of Bogalusa, Louisiana. Method: This cross‐sectional analysis considered 1,040 participants from the Bogalusa Heart Study (mean age 48.1 ± 61% female, 31% Black) with a complete neuropsychological (NP) assessment and food frequency questionnaires. (AHEI‐2010 score, range 0‐110). Regression models using GEE accounted for clustering effects of neighborhood factors/SES using census‐track level codes and estimated the association between quintiles (Q1‐Q5) of AHEI‐2010 and a global cognitive z‐score (GCS) and cognitive domains. Result: Individuals in Q5 (healthiest) tended to be female (68%, p = 0.03), significantly older (49.3 ± 5, p<0.001), with higher education (³ college 56%, p<0.001), and AHEI‐2010 scores from 53.38‐79.2. Compared with participants in Q1. Those in Q5 had higher GCS [ß (95% CI):1.71 (0.85, 2.57); p<0.001] after adjusting for age, sex, race, and the ACS‐Index of concentration at the extremes (to account for neighborhood‐level SES). Similarly, those with the healthiest diet had higher NP z‐scores within the domains of executive function, attention and speed processing information, and episodic memory. However, these associations were not significant after adjusting for education. (Table 1) For episodic memory, individuals in quintile 4 had higher NP z‐scores after adjusting for education [0.36 (0.07, 0.65); p = 0.01], and the trend across quintiles was also significant in all models and after adjusting for education [p = 0.03]. Conclusion: As early as midlife, Black and White individuals from rural settings with greater adherence to healthy dietary patterns have better NP performance across different neurocognitive domains. These results strengthen previous evidence suggesting that diet quality may be a potential avenue of intervention to prevent cognitive decline and aid health disparities. [ABSTRACT FROM AUTHOR]

Background: Smartphone apps for digital health with decentralized data collection enable more frequent and precise brain‐health monitoring via the abundance of embedded sensors. Such methods may improve the sensitivity of traditional metrics and representation among diverse populations impacted by health disparities, especially in ADRD studies. Here, we describe a framework for implementing a smartphone‐based digital platform for precision brain health assessment in a longitudinal cohort of Black and White men and women. Methods: Participants are being recruited as part of a multi‐site study from the Bogalusa Heart Study (BHS), a community‐based cohort in rural Louisiana that began in 1973. The digital platform consists of a smartphone application that is used remotely every three months for two weeks. It includes tasks that assess cognition, speech, gait and balance, and questionnaires that ensure instructions are appropriately followed and to monitor usage experiences. The BHS study personnel handle the initial screening, enrollment, and application setup. To overcome internet connection and time/availability issues, remote phone calls from BHS staff and text messages from a local area code using RingCentral software were implemented. Messages included short reminders, a link to step‐by‐step instructions to complete the application setup, and examples of smartphone tasks. (Figure 1.) Participants could schedule a call to receive phone instructions to complete the process. Results: A total of 94 participants [age 56 ± 4.5, women n = 60 (69%), Black n = 16 (19%)] have been enrolled in the study thus far. Of those, 77 (82%) registered and set up the application (69 at the study site and 8 remotely via text/phone), and 62 (81%) completed at least one set of tasks. There was no significant difference in completion rates between race [White n = 52 (84%) and Black n = 10 (77%); p = 0.54] and sex groups [women n = 45 (82%) and men n = 17 (85%); p = 0.75]. Conclusion: Results showed that individuals from a rural community‐based cohort are willing to use digital technologies for brain health assessments. These findings highlight the importance of utilizing low‐cost, scalable platforms and adapting research protocols to meet the requirements of specific populations, particularly those from low socioeconomic and underrepresented backgrounds. [ABSTRACT FROM AUTHOR]

Background: Digital cognitive assessments (DCA) enable quick, sensitive, and reliable assessment of neurocognitive operations that underlie dementia and mild cognitive impairment (MCI). We examined how well the DCTclockTM DCA, Mini‐Mental State Examination (MMSE), and Philadelphia (repeatable) Veral Learning Test (PrVLT) can classify memory clinic patients by cognitive impairment profile. Method: 106 serial referrals to an outpatient memory clinic were assessed with the DCTclock, MMSE, and P(r)VLT. Using statistical criteria for subtle cognitive impairment (SCI, Edmonds et al., 2015) and MCI (Jak et al., 2008), patients were classified into cognitively normal (CN, n = 21), SCI (n = 33), amnestic MCI (aMCI, n = 22), and dysexecutive/mixed MCI (d/mMCI, n = 30) groups. A series of stepwise logistic regression analyses were conducted with CN participants as the criterion group. Result: In the first analysis (Table 1), P(r)VLT‐recognition (X2= 24.39, p<0.001) entered the model first followed by the DCTclock Score (X2= 16.68, p<0.001). P(r)VLT‐recognition classified aMCI patients (Wald = 11.21, p<0.001). DCTclock (Wald = 8.97, p<0.001) and P(r)VLT‐recognition (Wald = 4.06, p<0.001) classified d/mMCI patients. In the second analysis the four command DCTclock composites were analyzed, excluding the MMSE and P(r)VLT‐recognition scores, Spatial Reasoning classified patients into their respective groups (SCI, Wald = 3.68, p<0.055; aMCI, Wald = 8.84, p<0.003; d/mMCI, Wald = 19.12 p<0.001). Third, analyzing DCTclock command Spatial Reasoning, MMSE, and P(r)VLT‐recognition, the P(r)VLT score, again, entered first (X2= 24.39, p<0.001) followed by DCTclock Spatial Reasoning (X2= 18.95, p<0.001). DCTclock Spatial Reasoning classified SCI patients (Wald = 4.66, p<0.031), P(r)VLT‐recognition classified aMCI patients (Wald = 9.11, p<0.003), and both DCTclock command Spatial Reasoning (Wald = 10.56, p<0.001) and P(r)VLT‐recognition (Wald = 3.32, p<.068) classified d/mMCI patients. Conclusion: Unlike the MMSE that takes 10‐15 minutes to administer and is scored subjectively, all DCTclock data are digitally obtained and automatically scored, requiring no examiner interface for administration and interpretation, thus obviating clinical judgment, ensuring inter‐rater reliability, and enabling more efficient cognitive assessments. In this series of logistic regression analyses, a combination of DCTclock and P(r)VLT clearly outperformed the MMSE in classifying non‐demented memory clinic patients into cognitively normal, subtle cognitive impairment, and MCI subtypes. These results support the efficacy of DCAs in clinical trials and to screen for emergent cognitive impairment in primary care and specialty medical practices. [ABSTRACT FROM AUTHOR]

Background: Digital assessment technology provides rapid and reliable assessment of neurocognitive abilities associated with Alzheimer's Disease and Related Dementia (ADRD) and Mild Cognitive Impairment (MCI). Dysexecutive and amnestic impairment typify many ADRD and MCI disorders. Here, we evaluate a 10‐minute digitally administered protocol assessing these neurocognitive constructs. Method: 229 community dwelling participants were assessed (age = 68.32+9.85; education = 15.01+2.58, MMSE = 26.84+2.78; 144 females). Executive abilities were assessed with three trials of 5‐digits backward digit span test (BDST) and the 'animal' fluency test. Episodic memory was assessed with a 6‐word Philadelphia (repeatable) Verbal Learning Test (PrVLT) consisting of 2 immediate free recall (IFR) trials, delayed free recall (DFR), and delayed recognition. Traditional pencil and paper testing was not used. Data were obtained with an iPad processed using automatic speech recognition. Result: A K‐means cluster analysis (k = 3) classified participants with possible dementia (pDem = 31), MCI (n = 96), and cognitively normal (CN, n = 100; Table 1). Groups differed from each other on all three P(r)VLT parameters (p = 0.001). CN participants outperformed other groups on executive measures (p = 0.001). All between‐group analyses yielded large effect sizes (η2 = 0.277–0.642). Logistic regression analysis using P(r)VLT‐DFR, P(r)VLT‐recognition, BDST, and 'animal' fluency parameters classify MCI and pDem participants into their respective groups compared to CN participants (p = 0.003). For the pDem group, within‐group t‐tests found rapid forgetting (PrVLT‐DFT

Background: Digital cognitive assessments (DCA) enable quick, sensitive, and reliable assessment of neurocognitive operations that underlie dementia and mild cognitive impairment (MCI). We examined how well the DCTclockTM DCA, Mini‐Mental State Examination (MMSE), and Philadelphia (repeatable) Veral Learning Test (PrVLT) can classify memory clinic patients by cognitive impairment profile. Method: 106 serial referrals to an outpatient memory clinic were assessed with the DCTclock, MMSE, and P(r)VLT. Using statistical criteria for subtle cognitive impairment (SCI, Edmonds et al., 2015) and MCI (Jak et al., 2008), patients were classified into cognitively normal (CN, n = 21), SCI (n = 33), amnestic MCI (aMCI, n = 22), and dysexecutive/mixed MCI (d/mMCI, n = 30) groups. A series of stepwise logistic regression analyses were conducted with CN participants as the criterion group. Result: In the first analysis (Table 1), P(r)VLT‐recognition (X2 = 24.39, p<0.001) entered the model first followed by the DCTclock Score (X2 = 16.68, p<0.001). P(r)VLT‐recognition classified aMCI patients (Wald = 11.21, p<0.001). DCTclock (Wald = 8.97, p<0.001) and P(r)VLT‐recognition (Wald = 4.06, p<0.001) classified d/mMCI patients. In the second analysis the four command DCTclock composites were analyzed, excluding the MMSE and P(r)VLT‐recognition scores, Spatial Reasoning classified patients into their respective groups (SCI, Wald = 3.68, p<0.055; aMCI, Wald = 8.84, p<0.003; d/mMCI, Wald = 19.12 p<0.001). Third, analyzing DCTclock command Spatial Reasoning, MMSE, and P(r)VLT‐recognition, the P(r)VLT score, again, entered first (X2 = 24.39, p<0.001) followed by DCTclock Spatial Reasoning (X2 = 18.95, p<0.001). DCTclock Spatial Reasoning classified SCI patients (Wald = 4.66, p<0.031), P(r)VLT‐recognition classified aMCI patients (Wald = 9.11, p<0.003), and both DCTclock command Spatial Reasoning (Wald = 10.56, p<0.001) and P(r)VLT‐recognition (Wald = 3.32, p<.068) classified d/mMCI patients. Conclusion: Unlike the MMSE that takes 10‐15 minutes to administer and is scored subjectively, all DCTclock data are digitally obtained and automatically scored, requiring no examiner interface for administration and interpretation, thus obviating clinical judgment, ensuring inter‐rater reliability, and enabling more efficient cognitive assessments. In this series of logistic regression analyses, a combination of DCTclock and P(r)VLT clearly outperformed the MMSE in classifying non‐demented memory clinic patients into cognitively normal, subtle cognitive impairment, and MCI subtypes. These results support the efficacy of DCAs in clinical trials and to screen for emergent cognitive impairment in primary care and specialty medical practices. [ABSTRACT FROM AUTHOR]

Background: Amyloid beta plaques (APs) containing amyloid beta peptides (Aβ), blood‐brain barrier (BBB) breakdown, and progressive cognitive decline are hallmarks of Alzheimer's disease (AD) pathology. Pathophysiological changes that initiate and perpetuate these hallmarks remain unresolved. We hypothesize that chronic extravasation of immunoglobulin Gs (IgGs) and Aβ via chronically impaired BBB promotes their interaction with pyramidal neurons (PNs). This interaction impairs PN function, consequentially initiates and perpetuates these AD‐associated hallmarks. Method: We tested our hypothesis associating neurocognitive decline with BBB breakdown in humans and a mouse model. In humans, antemortem neurocognitive status and postmortem BBB breakdown in their Brodmann area 39/40 were analyzed [AD, mild cognitive impairment (MCI), and no cognitive impairment (NCI) subjects, n=15]. Additionally, we tested the association between chronic BBB compromise, Aβ injection, and long‐term memory in CD1 mice (10 months, male, n=33) with Pertussis toxin (PT) injection‐induced BBB breach. This mouse‐model comprised three injection groups: Saline only, PT+Saline, and PT+Aβ42. We used the immunohistochemistry technique and image‐analysis software to track and quantify the fate of blood‐borne Aβ42 in mouse brain by injecting synthetic, fluorescent‐tagged monomeric Aβ42 (F‐Aβ42). We used Lashley's III Maze and Water Maze to evaluate the impact of BBB breach and Aβ42 injection on animals' long‐term memory. Result: Brodmann areas 39/40 of AD subjects showed a greater incidence and extent of BBB breach than MCI and NCI subjects. Cognitive test results demonstrating reduced cognitive function for Brodmann areas 39/40 correlated with the extent of BBB breach. Likewise, PT+Aβ42 mice exhibited increased BBB permeability and selectively heightened immunoreactivity between PNs and extravasated IgGs compared to other groups. Since these IgGs target "self" epitopes in the brain parenchyma, we refer to this subset of IgGs as brain‐reactive autoantibodies (BrAbs). Additionally, in PT+Aβ42 mice, we observed intraneuronal accumulations of F‐Aβ42 in cortical and hippocampal PNs and significant impairment in long‐term memory. Conclusion: In AD pathogenesis, BBB compromise induces chronic extravasation of BrAbs and Aβ, an event that serves as a mechanistic link that initiates and perpetuates Aβ deposition in PNs. Deposition of Aβ may perturb PN function and lead to progressive cognitive decline. [ABSTRACT FROM AUTHOR]

Background We assessed whether mild cognitive impairment (MCI) subtypes could be empirically derived within the Alzheimer's Disease Neuroimaging Initiative (ADNI) MCI cohort and examined associated biomarkers and clinical outcomes. Methods Cluster analysis was performed on neuropsychological data from 825 MCI ADNI participants. Results Four subtypes emerged: (1) dysnomic (n = 153), (2) dysexecutive (n = 102), (3) amnestic (n = 288), and (4) cluster-derived normal (n = 282) who performed within normal limits on cognitive testing. The cluster-derived normal group had significantly fewer APOE ε4 carriers and fewer who progressed to dementia compared with the other subtypes; they also evidenced cerebrospinal fluid Alzheimer's disease biomarker profiles that did not differ from the normative reference group. Conclusions Identification of empirically derived MCI subtypes demonstrates heterogeneity in MCI cognitive profiles that is not captured by conventional criteria. The large cluster-derived normal group suggests that conventional diagnostic criteria are susceptible to false-positive errors, with the result that prior MCI studies may be diluting important biomarker relationships. [ABSTRACT FROM AUTHOR]

Background: Digital technology offers the ability to obtain nuanced data previously unobtainable. A digital Backward Digit Span (BDT) using voice recognition was administered to memory clinic patients. Our goal was to assess the validity of this technology by (1) analyzing differences between non‐MCI versus MCI patients; (2) examining correlations with other neuropsychology tests; and, (3) analyzing test performance in relation to MRI subcortical regions of interest known to be related to working memory. Method: Jak‐Bondi (2009) criteria classify patients into MCI (n= 67) and non‐MCI (n= 72) groups. Working memory (WM) was assessed with the Backward Digit Span Test (BDT) consisting of 7 trials of 5 digits backwards; all trials were administered; and comprehensive neuropsychological testing was obtained. Only correct test trials were analyzed. BDT variables of interest included latency for each five responses. NeuroquantTM software calculated thalamic, caudate and hippocampal volume (n= 23) controlled for intracranial volume (ICV). Result: Between‐group difference for total time to completion was not significant (M= 55.64 + 18.91 sec.). An analysis of serial order position latencies clearly suggests two distinct epochs (Figure 1). A latency ratio (LR) for serial order 1 versus serial order 3 was calculated. Using this measure, non‐MCI patients presented with slower latency to initiate their response (serial order 1); but faster latency to sustain their response (serial order 3) compared that MCI patients (F[1,52]= 6.26, p<.016). Simple correlations were significant between the BDT‐LR and Judgement of Line Orientation Oblique test items (r= 0.325, p<.019) and WAIS‐III Digit Symbol test performance (r= 0.316, p<.017). Simple correlations were significant or near significant between the BDT‐LR and left thalamic volume (r= 0.468, p< 0.032); right thalamic volume (0.410, p<.065); total thalamic volume (r= 0.470, p< 0.032); and left caudate volume (r= 0.447, p<.042). Conclusion: These data are remarkable for how patients allocate their time to generate a correct response. Correlations between BDT‐LR, subcortical nuclei, and neuropsychological test performance provide evidence for the validity of using discrete latency measures to assess WM in patients with MCI. Intra‐response BDT latency ratios could be a meaningful neurocognitive biomarker for emergent illness. [ABSTRACT FROM AUTHOR]

Background: Modifiable lifestyle factors such as diet and physical activity are associated with cognitive function in aging adults. However, less is known regarding the intersection between lifestyle factors and cognitive heterogeneity in this same population. Understanding the types of dietary and physical activity patterns that are associated with differing cognitive profiles in normal aging could provide clinically meaningful data to promote cognitive health. Therefore, we characterized cognitive subgroups within a community‐dwelling sample of non‐demented, non‐depressed older adults and examined whether resulting cognitive subgroups differed on dietary and physical activity measures. Method: 121 non‐demented, non‐depressed participants (M age=67.69; 49.6% female; 60% minority; MMSE=28.62) completed a neuropsychological evaluation. Latent Class Analysis (LCA) utilizing six cognitive scores (CVLT Delayed Free Recall & Recognition Discriminability, 'Animal' and Letter Fluency, Letter‐Number Sequencing, & Boston Naming) revealed 4 cognitive subgroups with 1) high‐normal cognition (50.4%); 2) normal cognition (25.6%); 3) poor memory (15.7%); and 4) poor executive functioning (8.3%). A subset of participants completed the Sedentary Behavior Questionnaire and CHAMPS Physical Activity Questionnaire for Older Adults (n=96), and the Food Frequency Questionnaire (n=89). Weekly sedentary behavior and weekly moderate‐to‐vigorous physical activity (MVPA) total scores and a FFQ‐derived Mediterranean Diet (MedDiet) score were calculated according to published criteria. Differences between the four cognitive subgroups and these lifestyle variables were examined using ANCOVA adjusting for age, sex, race, word reading, depressive symptoms, and body mass index. Result: There was a significant difference between cognitive subgroups (p=.016) for MedDiet consumption. Post‐hoc analyses revealed participants with high‐normal cognition had higher MedDiet scores than those with normal cognition (p=.001). There were no significant differences between cognitive subgroups for sedentary behavior (p=.695) or MVPA (p=.731) scores. Conclusion: LCA‐defined cognitive subgroups within non‐demented, non‐depressed older adults showed group differences in MedDiet consumption (participants with high‐normal cognition greater than those with normal cognition). No group differences were seen with sedentary behavior or physical activity. When choosing 'one thing' to advise non‐demented, non‐depressed older adults about maintaining cognitive health, adherence to a MedDiet may mean the difference between good, better, and best cognitive aging. [ABSTRACT FROM AUTHOR]

Background: Healthy, young adults deviate leftward when bisecting horizontal lines and upward when bisecting vertical lines, a phenomenon termed "pseudoneglect". With aging, there is a reduction in this leftward bias. Alzheimer's disease (AD) can be associated with neglect. Our research assessed directional bias in healthy controls, patients with amnestic mild cognitive impairment (aMCI), and AD with the digital Clock Drawing Test (dCDT) command test condition. Method: The sample included 68 cognitively normal older adults (M= 72.08, SD= 5.1), 16 younger healthy adults (M= 31.34, SD= 5.83), 13 patients with aMCI (M= 72.56, SD= 6.00), and 24 patients with AD (M= 80.34, SD= 5.97). dCDT protocols were selected for correct number placement within 30 degrees of ideal and hands directed toward the correct digits. The estimated center of the clock face was determined by the point where clock hands converged. Horizontal and vertical deviations of this estimated center from the actual center in each group were normalized as percentages of the clock face diameter. Result: Horizontal deviation: Younger adults deviated leftward(M= 3.55%, SD = 2.82%) and differed from older controls (p= 0.003) who were near center with leftward deviation (0.31%, SD 2.58%). AD patients deviated leftward (M= 3.77%, SD = 4.58%) relative to older controls (p= 0.008). aMCI patients also deviated leftward (M= 2.18%, SD=2.36%) and differed from older controls (p= 0.047). Vertical deviation: Every group deviated upward relative to center (M= 1.82%, SD= 5.09%; p< 0.001). Conclusion: On average, clock center estimation deviates upward similarly across age and AD related neurodegeneration. Leftward "pseudoneglect" was present in younger clocks and decreased with age. AD and aMCI were associated with greater leftward deviation compared to healthy older adults. Leftward deviation on the dCDT may be a useful marker predicting cognitive impairment in otherwise "normal" clocks. [ABSTRACT FROM AUTHOR]