ABSTRACT Objectives There is a lack of consensus regarding what constitutes cognitively normal performance on the Montreal Cognitive Assessment (MoCA) based on demographic characteristics. Further, research regarding normative data on the MoCA for middle‐aged individuals is relatively limited. The current study sought to provide age‐ and education‐corrected normative data for the MoCA in a large epidemiological cohort of cognitively healthy middle‐aged and older adults with characteristics similar to the original validation sample of the MoCA. Methods Participants were from Generation 3 and Omni 2 cohorts of the Framingham Heart Study (n = 2637; 91.43% non‐Hispanic White) who were determined to be cognitively unimpaired at the time of MoCA assessment (Mean age = 53.56 years, age range = 32–83 years, 63.71% ≥ college‐educated). Normative data were generated by age in 10‐year intervals and education (≤ high school, some college, or ≥ college degree). Analysis of variance was used to examine the relationship between MoCA performance, age, and education. Results The average MoCA score across all participants was close to the revised MCI cutoff of 23 (M = 24.69, SD = 3.03). The average MoCA score for individuals over the age of 60 was below the recently suggested MCI cutoff score of 23 points. Similarly, individuals above the age of 70 scored below the revised cutoff score of 23 points, irrespective of level of education. Further, performance of participants below the age of 40 who were college educated was similar to the frequently used original MCI cutoff score of 26 (M = 26.28, SD = 2.41). Conclusions Results are consistent with previous literature suggesting that the original MoCA cutoff score of 26 may result in a high rate of false positives. Findings indicate that the recently suggested MCI cutoff score of 23 on the MoCA may also be artificially high. Using inappropriate normative data for the MoCA can impact diagnostic accuracy as well as misclassification in research settings. These findings highlight the need for the use of demographically appropriate, population‐based normative data for the MoCA in clinical and research settings.

Importance:Traumatic brain injury (TBI) has been associated with mortality and chronic conditions, including dementia. However, research is limited regarding the disease process by which TBI leads to long-term mortality. Objective:To assess TBI incidence over decades and its association with long-term all-cause and dementia-related mortality among participants in the Framingham Heart Study (FHS), with the hypothesis that the association would be largely attributable to dementia-related mortality. Design, Setting, and Participants:This cohort study used data from the original and offspring cohorts of the FHS, a multigenerational, community-based cohort study conducted from 1948 to 2022. Participants with TBI were matched 1:3 to unexposed participants based on birth year, sex, and generational cohort. TBI data were collected from comprehensive medical record review and study examination visit records from the time of enrollment until death or conclusion of data collection. Data analysis was performed from February 1, 2023, to November 1, 2024. Exposure:Traumatic brain injury. Main Outcomes and Measures:The primary outcomes assessed were all-cause and dementia-related mortality. Time-to-event analyses were conducted using data from the FHS original and offspring cohorts. Results:This study included 10 333 FHS participants: 5209 from the original cohort (mean [SD] age at enrollment, 44 [9] years; 55.2% female; mean follow-up, 35 [15] years) and 5124 from the offspring cohort (mean [SD] age at enrollment, 36 [11] years; 51.5% female; mean follow-up, 39 [11] years). A total of 886 participants (17.0%) in the original cohort and 1243 (24.3%) in the offspring cohort experienced at least 1 TBI, with incidence rates of 7.02 (95% CI, 6.63-7.40) and 9.11 (95% CI, 8.70-9.51) TBI events per 1000 person-years and a mean (SD) age at TBI of 74 (16) and 71 (15) years, respectively. Falls were the most common mechanism of TBI in both the original (65.4%) and offspring (82.8%) cohorts. TBI was associated with all-cause mortality (hazard ratio [HR], 1.15 [95% CI, 1.06-1.26]) in a dose-dependent fashion with TBI severity (mild: HR, 1.06 [95% CI, 0.96-1.16]; moderate to severe: HR, 1.82 [95% CI, 1.48-2.25]). TBI was also associated with dementia-related mortality in a dose-dependent fashion (mild TBI: HR, 1.60 [95% CI, 1.31-1.97]; moderate to severe TBI: HR, 3.67 [95% CI, 2.31-5.80]) but not with non-dementia-related mortality. Conclusions and Relevance:In this study of FHS community-based original and offspring cohorts followed up over 7 decades, most TBIs occurred late in life and were related to falls. TBI was associated with an increased risk of long-term, all-cause mortality in a dose-dependent fashion, with dementia-related mortality largely contributing to this association. Preventing falls to reduce TBI could have important implications for dementia and mortality.

Some evidence supports an association between traumatic brain injury (TBI) and greater risk of dementia, but the role of cognitive resilience in this association is poorly understood. 2,050 participants from the Framingham Heart Study Offspring cohort who were aged ≥60 year and had a plasma total tau (t-tau) measure at Exam 8 (2005-2008), and a neuropsychological (NP) exam visit within five years were included. Plasma t-tau was measured using the Simoa assay (Quanterix). NP factor scores were previously derived for memory, language, and executive function using confirmatory factor analysis. Information on TBIs was collected by comprehensive review of medical records, health history updates, exams, and self-report. TBI occurrence and severity were operationalized using modified ACRM & VA/DoD criteria, respectively. Cognitive resilience was operationalized using a residual approach by regressing each NP factor score on the plasma t-tau measure, adjusting for age at Exam 8, sex, education, time from blood draw, and APOE ε4 genotype. The adjusted residuals were then regressed on history of TBI (yes versus no), and severity of TBI (moderate-to-severe versus mild versus none). The sample was, on average, 67 years of age at Exam 8, 54% female, and college educated. No differences were observed in plasma t-tau levels between those with and without TBI. Having a history of TBI was significantly associated with a reduction in resilience in executive function (β: -0.110; 95% CI: -0.175, -0.044; p: 0.001) as compared to not having a history of TBI. No significant associations were observed between history of TBI and resilience in memory or language. Greater TBI severity was significantly associated with worse resilience in executive function in a dose-response manner (P trend : <0.001), with the association being strongest in the moderate-to-severe TBI group (β: -0.209; 95% CI: -0.340, -0.078; p: 0.002) followed by the mild TBI group (β: -0.082; 95% CI: -0.155, -0.010; p: 0.026). Having a TBI was associated with worse resilience to neurodegeneration in executive function, and most strongly among individuals with moderate-to-severe TBI. These results suggest that having a TBI may increase vulnerability to late-life executive dysfunction after accounting for a primary neurodegenerative disease process.

Background:Genome-wide association studies (GWAS) have identified over 1,000 blood pressure (BP) loci and over 80 loci for Alzheimer's disease (AD). Considering BP is an AD risk factor, identifying pleiotropy in BP and cognitive performance measures may indicate mechanistic links between BP and AD. Methods:Genome-wide scans for pleiotropy in BP variables-systolic (SBP), diastolic (DBP), mean arterial (MAP), and pulse pressure (PP)-and co-calibrated scores for cognitive domains (executive function, language, and memory) were performed using generalized linear mixed models and 116,075 longitudinal measures from 25,726 participants of clinic-based and prospective cohorts. GWAS was conducted using PLACO to estimate each SNP's main effect and interaction with age, and their joint effect on pleiotropy. Effects of genome-wide significant (GWS) pleiotropic SNPs on cognition as direct or mediated through BP were evaluated using Mendelian randomization. Potential contribution of genes in top-ranked pleiotropic loci to cognitive resilience was assessed by comparing their expression in brain tissue from pathologically confirmed AD cases with and without clinical symptoms. Results:Pleiotropy GWAS identified GWS associations with APOE and 11 novel loci. In the total sample, pleiotropy was identified for SBP and language with JPH2 ( P Joint =6.09×10 -9 ) and GATA3 ( P G×Age =1.42×10 -8 ), MAP and executive function with PAX2 ( P G×Age =4.22×10 -8 ), MAP and language with LOC105371656 ( P G×Age =1.75×10 -8 ), and DBP and language with SUFU ( P G =2.10×10 -8 ). In prospective cohorts, pleiotropy was found for SBP and language with RTN4 ( P G×Age =1.49×10 -8 ), DBP and executive function with ULK2 ( P Joint =2.85×10 -8 ), PP and memory with SORBS2 ( P G =2.33×10 -8 ), and DBP and memory with LOC100128993 ( P G×Age =2.81×10 -8 ). In clinic-based cohorts, pleiotropy was observed for PP and language with ADAMTS3 ( P G =2.37×10 -8 ) and SBP and memory with LINC02946 ( P G×Age =3.47×10 -8 ). Five GWS pleiotropic loci influence cognition directly, and genes at six pleiotropic loci were differentially expressed between pathologically confirmed AD cases with and without clinical symptoms. Conclusion:Our results provide insight into the underlying mechanisms of high BP and AD. Ongoing efforts to harmonize BP and cognitive measures across several cohorts will improve the power of discovering, replicating, and generalizing novel associations with pleiotropic loci.

BackgroundCardiovascular health (CVH) is a modifiable risk factor for Alzheimer's disease (AD). However, studies examining the association between mid-age CVH, as indicated by Life's Essential 8 (LE8) health metrics, and digital cognitive performance or AD risk are limited.ObjectiveTo examine the associations between mid-age CVH, assessed by LE8 scores during ages 45 to 65, and digital Clock Drawing Test (dCDT) performance as well as the incidence of AD.MethodsWe included 1198 participants (51.6% women) from the Framingham Heart Study (FHS) Offspring cohort. Linear regression and Cox proportional hazards models were applied to examine the associations between mid-age CVH and dCDT performance, as well as the incidence of AD.ResultsOver a median follow-up of 17.5 years, 45 participants developed AD. Each standard deviation (SD) higher mid-age LE8 total score was associated with a 0.16 SD higher level of the dCDT total score (p < 0.001) and a 0.35-fold lower risk of incident AD (HR = 0.65, 95% CI: 0.49-0.87, p = 0.003). The dCDT measures showed stronger associations with mid-age LE8 and AD risk compared to the conventional CDT (cCDT). For example, the drawing score on copy tasks was more strongly associated with LE8 (beta = 0.10, p = 0.007 versus beta = 0.08, p = 0.27) and had higher discrimination for incident AD (C-statistic = 0.89 versus 0.83) compared to the cCDT.ConclusionsOur results highlight the potential of digital cognitive assessments for evaluating AD risk and emphasize the importance of mid-age CVH in shaping cognitive outcomes and the development of AD.

Although brain amyloid and tau deposition measured by PET scans are established as biomarkers of Alzheimer's disease (AD), they can emerge decades before symptoms are detectable on traditional neuropsychological (NP) tests. There is a pressing need for early AD detection tools that are more accessible, cost-effective, and non-invasive. The digital clock drawing test (dCDT), a digital version of the clock drawing test, has emerged as a promising cognitive assessment tool that takes minutes to administer and can reveal clinical symptoms earlier than paper-pencil NP tests. This study explored the association between 53 dCDT measures and amyloid and tau PET biomarkers using data from 87 low age risk participants in the Framingham Heart Study. Our findings revealed a significant association between a dCDT measure related to spatial reasoning function and global amyloid burden (P < 0.05), and 4 dCDT measures correlated with tau accumulation after adjusting for multiple comparisons. Notably, the combination of demographic variables and a composite dCDT score achieved a mean area under the receiver operating characteristics curve of 0.86 in detecting amyloid positivity. These results highlight the potential of dCDT measures as effective predictors of amyloid and tau pathology in preclinical AD.

Although traumatic brain injury (TBI) is known to be associated with short term mortality, its effects on long-term mortality remain less clear. TBI is also a well-known risk factor for dementia. We hypothesized that TBI would be associated with long-term mortality, particularly dementia-related mortality. The Framingham Heart Study (FHS) is a population-based cohort study which recruited its original cohort between 1948-1950, and its offspring cohort between 1971-1975. Information on TBIs was collected across the lifespan by comprehensive review of medical records, FHS health history updates and exams and self-report. TBI occurrence and severity were operationalized using modified American Congress of Rehabilitation Medicine and Veterans Affairs/Department of Defense criteria, respectively. Dementia diagnosis was operationalized using Diagnostic and Statistical Manual of Mental Disorders, 4 th Edition, and National Institute of Neurological and Communicative Diseases and Stroke/AD and Related Disorders Association criteria. Participants with history of TBI were matched to those without TBI for age and sex, with a 1:3 ratio. Participants who died within a year of TBI were excluded. Cox proportional hazards models assessed TBI risk for all-cause mortality and cause-specific competing risk models assessed risk for dementia-related mortality. Over the course of study, 72.82% died, with 12.32% of deaths from dementia-related mortality. 1,688 participants with TBI (Mean age at TBI = 69 ( SD = 15.5) years, 59.89% women) and 5,064 age- and sex-matched participants without TBI were followed for an average of 15±12 years after TBI. Hazard ratios (HRs) for all-cause mortality were 1.09 (95%CI 1.00-1.20) and 1.66 (95%CI 1.31-2.09 for mild and moderate-to-severe TBI respectively. Hazard ratios (HRs) for dementia-related mortality were 1.63 (95%CI 1.32–2.02) and 2.67 (95%CI 1.57–4.56) for mild and moderate-to-severe TBI respectively. Hazard ratios (HRs) for non-dementia related mortality were 0.96 (95%CI 0.89–1.04) and 1.15 (95%CI 0.95–1.38) for mild and moderate-to-severe TBI respectively. In a multi-generational community-based sample followed since the 1940s, mild and moderate-to-severe TBIs were associated with all-cause mortality with the effect driven by dementia-related mortality.

The deficit of unawareness of cognitive impairment (cognitive anosognosia) is known to be associated with adverse health outcomes, caregiver burden, and worse cognitive outcomes. A better understanding of cognitive self-awareness and the ability to self-judge cognitive performance among the general population would enable a rational design of cognitive screening and improve how subjective cognitive decline and self-reported errors at tasks like medication administration are interpreted. Participants were enrolled in the Framingham Heart Study, which is a community-based cohort with three generations of participants. Through ancillary studies, FHS participants underwent neuropsychological assessments every 4 to 5 years. For those with possible cognitive impairment, study examiners identified, and a panel later adjudicated, a clinical diagnosis (intact, mild cognitive impairment [MCI], or dementia, and their subtypes). Before the neuropsychological testing, participants were asked, “Do you have any concerns about your memory or thinking”. If “yes,” they were further asked: “Do you think this is normal for your age?” Our study included 5,110 participants of FHS with self-reported assessment of cognition (mean age: 67±15 years; 55.1% women, 66.2% with some college degree or above). As shown in Table 1 , among 359 (7.0% of the participants) diagnosed with MCI or dementia, 47.4% (170) had a self-report consistent with unawareness of cognitive impairment (72 or 45.0% of MCI, 98 or 49.2% of dementia). These individuals had NO self-reported concerns at all about memory and thinking. Even more concerning, a larger group of people (p = 0.025) with MCI or dementia who DID self-report concerns about memory/thinking (88 with MCI, or 55.0%, 101 with dementia, 50.7%), stated that they thought the problem was normal for their age (124 participants, 65.6%). Most of the Framingham Heart Study participants clinically diagnosed with pathologic cognitive impairment self-reported that they either had no memory or thinking problems, or, more often, self-reported memory and thinking problems that they felt were normal for their age. This is consistent with cognitive anosognosia, which can represent a range of unawareness. Since subjective cognitive decline is used to identify MCI, further research could develop complementary MCI diagnosis markers.

Preclinical measures of Alzheimer’s disease (AD) risk include markers of amyloid and tau measured via PET scans and decline in verbal memory. Audio features from recorded speech collected during neuropsychological testing offer an alternative for monitoring preclinical AD symptoms that is nonintrusive, economical, and scalable but warrants further validation. This study aims to examine whether there is a stronger association between acoustic metrics and PET imaging biomarkers compared to a traditional verbal memory test. This study involved 196 Framingham Heart Study dementia-free participants with available acoustic, as well as PET amyloid and tau imaging data (refer to Table 1 for demographics). For each participant, 65 acoustic features were extracted from speech samples recorded during the Wechsler's Memory Scale Logical Memory immediate (LMI) and delayed (LMD) recall tests, utilizing OpenSMILE for feature extraction. Multiple regression analyses were performed with a global composite score for PET amyloid, or PET tau for five regions (entorhinal, amygdala, parahippocampal, inferiortemporal, putamen) that were examined individually as outcomes. LMI, LMD, acoustic features derived from LMI and from LMD recorded voice segments were related to AD PET biomarkers in each regression model. Age, sex, education level, and APOE status were used as covariates for all analysis. Models using tau as outcome were adjusted for global amyloid. After adjusting for multiple comparisons using FDR (significance set at 0.5), among the 65 acoustic features, the Mel-frequency cepstral coefficient (p=0.01) and audio spectral features of the acoustic features derived from LMD (p=0.02) were most strongly associated with inferiotemporal tau region (see Table 2). No traditional measure of LMI or LMD was significantly related to any AD PET biomarker. In a group that was clinically asymptomatic for dementia/AD, only digital measures were associated with AD tau biomarker. These findings suggest that digital voice features have increased sensitivity to detect those at AD risk compared to traditional cognitive tests and have potential to serve as a biomarker of AD. Future studies in diverse populations are needed to test the robustness of these findings.

Widely used neuropsychological test instruments are notoriously biased across the demographics of age, sex/gender, education, language and culture. This includes verbal memory tests that elicit speech such as the paragraph recall or list-learning memory tests. Language tests are similarly biased, including the Boston Diagnostic Aphasia Examination Cookie Theft Test (CTT) that has been used to elicit both written and spoken responses for decades. Biased tests generate biased results, and this problem is exacerbated by the widespread practice of repurposing tests generated for those in largely high income, Westernized countries for use in other countries that are demographically and culturally distinct. We tested the use of large language model (LLM) tools to automatically generate verbal test stimuli that were more locally relevant. Starting with the CTT, that involves describing a picture that contains 14 core content units, we used ChatGPT to aid in generating customized pictures with the same number of core content units that were more universally relevant, (e.g., open park area, beach, etc.) by suggesting parameters and integrating resulting features into new pictures. We also tested capabilities to generate pictorial scenes that are more representative of regional settings, such as the Asia Pacific or Middle East. We are testing LLM capacity to generate similarly regionally customized list learning stimuli that maintain the original list learning test word frequency and difficulty across languages. Using an iterative approach, we were able to use automated artificial intelligence (AI)-driven tools to rapidly generate regionally relevant test stimuli that retained the original test characteristics to preserve generation of comparable results. Descriptive comparisons found that aided by automated AI tool, we could generate new neuropsychological test stimuli in a matter of minutes compared to months long manual methods. The rapid emergence of AI has generated significant concerns about its appropriate use for health and health care applications. Given that AI is here to stay, we are testing the capacity of this technological advance to address long-standing, persistent problems in AD research and usher in a new era of precision brain health that has global relevance in a rapidly aging world population.

Observe whether practice effects occur among cognitively unimpaired adults in digital, self-administered cognitive tasks at three-month intervals across four timepoints. Participants included 448 cognitively unimpaired adults (average age 63.9 years, 33.5% male, 90.9% white, 67.4% college graduates) enrolled in the Framingham Heart Study. Inclusion criteria included smartphone ownership, Wi-Fi access, and English fluency. Participants self-administered three subtests from the Digital Automated Neurobehavioral Assessment (DANA) battery —Simple Reaction Time (SRT), Go/No-Go (GNG), and Code Substitution (CS)—at four tri-monthly timepoints. Outcomes included response time (RT), accuracy (ACC), and cognitive efficiency (CE: ACC divided by RT, scaled). Repeated measures ANOVA and η2 were used to evaluate changes in each outcome over time. All three tests showed significant improvements (p < .001) in response time and cognitive efficiency across timepoints. When considering both significance and small effect size (η2 ≥0.01), SRT showed effects only for accuracy, while GNG and CS showed effects for response time and cognitive efficiency, but not accuracy. The largest changes for all tests occurred between T1-T2; GNG and CS continued to improve across subsequent timepoints, while SRT plateaued after T2. Distinct speed-accuracy dynamics appeared to influence practice effects based on task complexity. Effects were larger on more complex tasks (CS, GNG) and driven primarily by improved response speed, not accuracy. Meanwhile, improvements on the simpler task (SRT) were smaller and more driven by accuracy. Findings demonstrate practice effects specific to the present tasks, sample characteristics, and test-retest intervals. Speed-accuracy dynamics and task complexity shape longitudinal cognitive performance.

There is growing evidence that epigenetic age acceleration may predict late life cognitive decline and dementia, but it is unknown whether this is due to accelerated neurodegeneration or reduction in cognitive resilience. We examined the relationship between epigenetic clocks and domain specific neuropsychological (NP) factor scores, mild cognitive impairment (MCI), Alzheimer’s Disease (AD), and all-cause dementia, before and after accounting for plasma total tau (t-tau), a marker of neurodegeneration. DNA methylation and plasma t-tau (Simoa assay; Quanterix) data from 2091 Framingham Heart Study Offspring cohort participants were generated from blood at the same Exam 8 visit (2005-2008). Three epigenetic clock measures: DunedinPACE, PC PhenoAge, and PC GrimAge were estimated from the DNA methylation data. Longitudinal NP factor scores were previously derived for memory, language, and executive function using confirmatory factor analysis. We tested the association of epigenetic age acceleration with cognitive trajectories using linear mixed effects models and with time to MCI, all-cause dementia and AD using Cox-proportional hazard models. Models were run with and without adjustment for plasma t-tau. All models included APOE ε4-carrier status, education, smoking, age, and sex as covariates. Epigenetic measures were standardized in all models. At Exam 8, the sample was, on average, 66.3 (SD = 9.0) years of age, 54.8% female, and had 16.4 (SD = 2.7) years of education. DundeinPACE was significantly associated with faster decline in executive function (β timeXepi_age = -0.005, 95% CI:[-0.009,-0.002], p = 0.0020), but not with baseline executive function. Older PhenoAge (β epi_age = -0.041, 95% CI:[-0.067,-0.014], p = 0.0028) and GrimAge (β epi_age = -0.042, 95% CI:[-0.073,-0.011], p = 0.0084) were significantly associated with worse baseline executive function, but not with rate of decline. Older PhenoAge also was significantly associated with worse baseline memory (β epi_age = -0.037, 95% CI:[-0.061,-0.012], p = 0.0036). DunedinPACE was significantly associated with time to MCI (HR = 1.20, 95% CI:[1.06,1.35], p = 0.0034), AD (HR = 1.30, 95% CI:[1.07,1.57], p = 0.0068) and all-cause dementia (HR = 1.30, 95% CI:[1.10,1.53], p = 0.0017). Results remained similar after adjustment for plasma t-tau. Epigenetic age acceleration may be a marker of cognitive resilience, particularly in executive function. Of the three epigenetic clocks examined, DundedinPACE showed the most robust associations with cognitive resilience, with lower DunedinPACE associated with greater cognitive resilience.

Repeated self-administered mobile-based cognitive assessments are increasingly being utilized to identify preclinical cognitive decline. Repetition of cognitive tests during a short time interval often leads to improved performance (i.e., practice effects) due to familiarity with test structure. Therefore, test-specific practice effects may delay the detection of subtle decline. Alternatively, the lack of practice effect may indicate a decline in cognitive function. This study investigated whether practice effects were present on three digital tests repeated over a 3-month interval in two cognitively unimpaired cohorts. 549 cognitively intact participants (average age 63.5 years, 40.6% male, 89.6% white) enrolled in the Framingham Heart Study and 26 cognitively unimpaired participants (average age 55 years, 15.4% male, 80.8% white) from the Bogalusa Heart Study completed three self-administered subtests from Digital Automated Neurobehavioral Assessment (DANA) on their own mobile device at baseline (T1) and three months later (T2). Cognitive scores were calculated using Cognitive Efficiency (CE), a derived measure of speed and accuracy, for Code Substitution (CS), Simple Reaction Time (SRT), and Go/No-Go (GNG). CS assesses visual scanning and attention, learning, and immediate recall; SRT measures pure reaction time; and GNG assesses sustained attention and impulsivity. For each cohort, separate paired sample t-tests examined mean differences in CE between timepoints for all three tests. On all three tests, similar trends in test performance were found in both cohorts. FHS participant’s CE scores significantly improved at T2 compared to T1 on CS and GNG, but not on SRT (Table 2). Similar trends were observed among BHS participants, with more relative improvements on CS and GNG compared to SRT. This study found similar trends in test-specific practice effects on mobile-based cognitive tests in two different cohorts. CS and GNG are inherently more complex than SRT, which may explain why test familiarity improved performance three months after baseline. This is congruent with traditional paper-and-pencil neuropsychological tests, wherein tests with complex stimuli show more practice effects relative to simple tasks. Additional research is needed to determine which tests are more sensitive to detecting earliest indicators of cognitive decline, and whether the presence/absence of practice effects itself are clinically meaningful.

Plasma amyloid and tau levels are known to be associated with the risk of Alzheimer’s disease (AD). In recent years, digital cognitive assessment has been increasingly recognized as a potential tool for long-term cognitive monitoring. This study aims to explore the potential of digital cognitive metrics as efficient and low-cost alternatives to plasma biomarkers. This study included participants from the Framingham Heart Study who had either their spoken responses to neuropsychological examinations digitally recorded or had completed the digital Clock Test (dCDT), along with plasma biomarker measurements taken within five years. Simoa™ Tau 2.0 Kit and INNO-BIA plasma Aβ were used for tau, and Aβ40 and Aβ42 measurement, respectively. The study analyzed digital cognitive metrics, comprising 65 acoustic features extracted via OpenSMILE software and 101 features derived from the dCDT. Linear regression models, adjusted for age and sex, were employed to examine the association between these digital cognitive metrics and plasma total tau, Aβ40, and Aβ42 levels. This study included 4274 participants (mean age 57±14 years old, 53.7% women). Table 1 shows the top digital cognitive metrics associated with plasma biomarkers. Notably, plasma total tau was associated with 38 acoustic features. The most significant association was observed between audspec_lengthL1norm (sum of the auditory spectrum) and plasma tau (beta=-0.09, SE = 0.03, P<0.001). Among the dCDT features, COMTotInkLengthCC, which represents the total drawing length of pen strokes in the command clock drawing, showed the most significant association with the plasma total tau level (beta=-0.13, SE = 0.06, P=0.023). Our results demonstrate an association between acoustic and digital clock drawing with plasma total tau, which is consistent with previous research that finds that cognitive performance is linked to AD tau pathology. These results highlight the promise of these digital cognitive metrics as an easily scalable, low-cost screening tool for those at high risk for AD. Future investigations to test the robustness of digital cognitive metrics as a surrogate screening measure will be critical in tackling AD pathology.

Greater occupational complexity may be protective against dementia in later life, but it is unclear if it contributes to cognitive resilience and whether different aspects of occupational complexity are associated with resilience. We examined relationships between occupational complexity related to data, people, and things, and cognitive resilience to neurodegeneration. 1,699 participants from the Framingham Heart Study Offspring cohort who were aged ≥60 years, had a plasma total tau (t-tau) measure (a marker of neurodegeneration), and a neuropsychological (NP) exam visit within five years of the plasma t-tau measurement were included. Plasma t-tau was measured using the Simoa assay (Quanterix) on samples collected at Exam 8 (2005-2008). NP factor scores were previously derived for memory, language, and executive function using confirmatory factor analysis. Occupational data were collected at the NP exam, from which occupational complexity was disaggregated into data complexity, people complexity, and things complexity according to the 1970 US Census Dictionary of Occupational Titles. Cognitive resilience was operationalized using a residual approach by regressing each NP factor score on the plasma t-tau measure, adjusting for age, sex, education, time from blood draw, and APOE ε4 status. The adjusted residuals were then regressed on each type of occupational complexity, dichotomized into higher complexity versus lower complexity. The sample was, on average, 70 years of age, 53% female, and had 15 years of education. Higher data (β = 0.20, 95% confidence interval (CI) = 0.15-0.25, p<0.001), people (β = 0.11, 95% CI = 0.07-0.15, p<0.001), and things (β = 0.05, 95% CI = 0.01-0.09, p = 0.015) occupational complexity were most strongly associated with resilience in executive function. Higher data (β = 0.10, 95% CI = 0.05-0.15, p<0.001) and people (β = 0.07, 95% CI = 0.03-0.11, p = 0.001) occupational complexity were associated with resilience in memory. Higher data (β = 0.07, 95% CI = 0.01-0.12, p = 0.014) occupational complexity was associated with resilience in language. Specific types of occupational complexity contribute to resilience to neurodegeneration in specific cognitive domains differently. Occupational complexity may offer the most resilience in executive function and occupations with high data complexity may offer the most cognitive resilience.

Remote, self-administered digital assessments provide scalable approaches for monitoring cognition in older adults. Practice effects (PE), once viewed as measurement noise, are increasingly recognized as potential markers of cognitive decline. As disease-modifying therapies for neurodegenerative conditions emerge, longitudinal digital assessments may capture subtle cognitive changes. Although the majority of PE studies have focused on memory tasks, it is critical to determine whether PE occur in other domains. Clarifying the potential presence of PE across multiple timepoints on the FDA-cleared Defense Automated Neurocognitive Assessment (DANA) platform remains critical for determining whether PE can serve as a measure of cognitive decline longitudinally. To examine whether PE occur across quarterly self-administrations of digital cognitive tasks across different cognitive domains. Demographic characteristics were examined as potential moderators. Cognitively intact older adults from the Framingham Heart Study self-administered DANA tasks at baseline and every three months for one year (four assessments, T1-T4). Tasks included Simple Reaction Time (SRT), Go/No-Go (GNG), and Code Substitution (CS). These tasks varied in complexity, from simple motor responding (SRT) to inhibitory control (GNG) and associative learning with working memory demands (CS). The primary outcome was Cognitive Efficiency, a derived measure of speed and accuracy, which are also recorded as secondary measures. PE were quantified using linear mixed-effects models and change in mean score across timepoints expressed in pooled SD units expressed as Cohen d. A total of 717 participants completed all four timepoints (mean age: 64.1±9.5 years; 65.4% females; 67.1% college educated). PE were observed on all tasks with small effect (d: SRT=0.16, GNG=0.32, CS=0.43). Demographics influenced baseline performance: older age predicted lower baseline scores on all tasks, females scored lower on SRT and GNG, and higher education predicted better performance on all tasks. Age was the only demographic moderator of PE, with smaller improvements on CS among participants >65 years (β=-0.39, P<.001). Quarterly self-administered digital cognitive assessments elicited measurable PE across processing speed, inhibitory control, and visual scanning, with stronger effects on more complex tasks. Demographics influenced baseline performance, while age moderated PE on the most demanding task. These findings underscore the importance of characterizing PE in digital assessments and highlight their potential clinical utility in detecting early cognitive impairment.

The relationship between sex-specific blood biomarkers and memory changes in middle-aged adults remains unclear. We aimed to investigate this relationship using the data from the Framingham Heart Study (FHS). We conducted association analysis, partial correlation analysis, and causal dose-response curves using blood biomarkers and other data from 793 middle-aged participants (<= 60 years) from the FHS Offspring Cohort. The results revealed associations of adiponectin and fasting blood glucose with midlife memory change, along with a U-shaped relationship of high-density lipoprotein cholesterol with memory change. No significant associations were found for the other blood biomarkers (e.g., amyloid beta protein 42) with memory change. To our knowledge, this is the first sex-specific network analysis of blood biomarkers related to midlife memory change in a prospective cohort study. Our findings highlight the importance of targeting cardiometabolic risks and the need to validate midlife-specific biomarkers that can accelerate the development of primary preventive strategies.

Background Smartphone‐based cognitive assessments have emerged as promising tools, bridging gaps in accessibility and reducing bias in Alzheimer disease and related dementia research. However, their congruence with traditional neuropsychological tests and usefulness in diverse cohorts remain underexplored. Methods and Results A total of 406 FHS (Framingham Heart Study) and 59 BHS (Bogalusa Heart Study) participants with traditional neuropsychological tests and digital assessments using the Defense Automated Neurocognitive Assessment (DANA) smartphone protocol were included. Regression models investigated associations between DANA task digital measures and a neuropsychological global cognitive Z score (Global Cognitive Score [GCS]), and neuropsychological domain‐specific Z scores. FHS participants’ mean age was 57 (SD, 9.75) years, and 44% (179) were men. BHS participants' mean age was 49 (4.4) years, and 28% (16) were men. Participants in both cohorts with the lowest neuropsychological performance (lowest quartile, GCS1) demonstrated lower DANA digital scores. In the FHS, GCS1 participants had slower average response times and decreased cognitive efficiency scores in all DANA tasks (P<0.05). In BHS, participants in GCS1 had slower average response times and decreased cognitive efficiency scores for DANA Code Substitution and Go/No‐Go tasks, although this was not statistically significant. In both cohorts, GCS was significantly associated with DANA tasks, such that higher GCS correlated with faster average response times (P<0.05) and increased cognitive efficiency (all P<0.05) in the DANA Code Substitution task. Conclusions Our findings demonstrate that smartphone‐based cognitive assessments exhibit concurrent validity with a composite measure of traditional neuropsychological tests. This supports the potential of using smartphone‐based assessments in cognitive screening across diverse populations and the scalability of digital assessments to community‐dwelling individuals.

Abstract Objective Research demonstrates adequate consistency between in-person and tele-np assessments with the latter typically omitting visually based measures and limiting capture of process-oriented scores. This study presents a novel tele-np protocol designed at the Framingham Heart Study (FHS) allowing for capture of visual and Boston Process Approach (BPA) data. Methods Beginning in 2021, FHS participants completed a tele-np protocol via HIPAA-compliant Zoom that includes tests similar to those administered in-person assessing all cognitive domains including visual memory and visuospatial functioning. Records forms, digital pen for BPA capture, and a label for return of materials, were mailed to participants. Regression models were used to compare in-person and tele-np results. Results The tele-np protocol consisting of 15 tests was administered to 169 participants across 30 states and three countries. Participants were 71.8 years on average (SD = 9.98), well-educated (84.6% ≥ college-educated), Caucasian (86.4%), and 53% male. Mean completion time of the tele-np protocol is 77.7 minutes (SD = 13.2). 76% of participants were deemed to be cognitively intact and 24% cognitively impaired based on their tele-np and previous in-person data. With the exception of a test of processing speed (β = −5.24 [−7.58,-2.89], p < 0.001), raw scores across most tests did not significantly differ based on mode of administration (tele-np vs. in-person) after adjusting for age, sex, education, and years between in-person and tele-np evaluations. Conclusion Preliminary findings indicate that the current tele-np protocol yields reliable results. Tele-np offers a cost-effective approach to improve accessibility and retention of participants, especially those out of state who may have otherwise been excluded from data collection.