Early Detection Of Cognitive Impairment Research Articles

Abstract 152: Bridging the Gap in Stroke Care: Enhancing Early Detection of Cognitive Impairments Through Standardized Screening Improves Therapy Referrals

Abstract 152: Bridging the Gap in Stroke Care: Enhancing Early Detection of Cognitive Impairments Through Standardized Screening Improves Therapy Referrals

Montreal Cognitive Assessment (MoCA) use in general practice for the early detection of cognitive impairment: a feasibility study.

GPs can detect cognitive impairment (CI) at a very early stage, allowing early support for people and their caregivers. The early onset of CI is between 50 years and 60 years. Currently, in France, the Mini-Mental State Examination (MMSE) remains the most used screening test, although it has a lower sensitivity and specificity than the Montreal Cognitive Assessment (MoCA) for detecting mild CI, taking an average of 15 minutes to complete. To investigate the feasibility of the MoCA during routine consultations in general practice for the early detection of CI and to determine prevalence of CI in a primary care setting. A quantitative, prospective feasibility study was carried out in real-life working conditions during routine GP consultations in France. GPs performed MoCA on adults aged ≥50 years, without suspected or confirmed CI. Sixty-one GPs performed 221 MoCA with a mean duration of 8 minutes and detected mild neurocognitive impairment in 62% of patients. The MoCA is feasible and easy to perform during routine consultations in general practice by trained and experienced physicians.

Cognitive assessment in patients with acute stroke: Exploring early intervention for dementia.

Early detection of cognitive impairment in patients with acute stroke could improve dementia treatment; however, such testing is uncommon. This study aimed to assess cognitive testing feasibility in patients with acute stroke and identify patient characteristics associated with testing ability. 291 patients with suspected acute stroke were admitted to our hospital between December 2016 and May 2017. Of these, 280 were diagnosed with stroke (median [interquartile range] age 73 [63-81] years; 70 % male; 11 % with prior cognitive decline) and included in the study. Patients able or unable to complete three cognitive function tests were classified as the testable or untestable group, respectively. Among the 280 patients, 72 % completed all three cognitive tests a mean of 4 [3-5] days after onset. Significant differences were found between the testable and untestable groups, particularly in age (70 [61-80] vs. 77 [69-84], p < 0.0001), Informant Questionnaire on Cognitive Decline in the Elderly score (78 [78-83] vs. 84 [78-96], p = 0.012), and National Institutes of Health Stroke Scale (NIHSS) score at onset (3 [1-6] vs. 13 [4-18], p < 0.0001). Advanced age, a higher NIHSS score, and pre-stroke cognitive decline were independent risk factors for the inability to undergo testing. Cognitive testing was feasible in 72 % of patients with acute stroke, particularly those who were younger, had normal pre-stroke cognitive function, and experienced mild strokes. Early cognitive assessment may enable timely dementia detection and treatment.

Usage patterns of cognitive assessments during Medicare's Annual Wellness Visit: A national survey.

Medicare's Annual Wellness Visit (AWV) is a logical opportunity for early detection of cognitive impairment, but recent data for uptake and cognitive assessments during it are lacking. We surveyed Medicare beneficiaries of a nationally representative panel about use of AWV and cognitive assessments and analyzed associations between uptake and beneficiaries' characteristics. Of 1871 participants, 80% had an AWV, among whom 31% underwent formal cognitive testing, 35% were asked about memory problems, including 15% having both. Males, rural residents, non-Hispanic Black beneficiaries, and those having subjective memory problems or a usual source of care were more likely to be probed, but no characteristics were associated with the probability of undergoing cognitive testing. Use of structured cognitive assessments did not increase with higher AWV uptake. Concerningly, individuals at higher risk of cognitive impairment were not more likely to be assessed, calling for policy interventions to increase assessment rates. Data are lacking on use of cognitive test during Medicare's Annual Wellness Visit (AWV). Use of structured cognitive tests did not increase with the uptake of the AWV. Individuals at higher risk of cognitive impairment did not receive more testing. More high-risk patients were asked about memory problems despite lack of testing.

Designing an intervention to improve cognitive evaluations in primary care

BackgroundEarly diagnosis is crucial to the optimal management of patients with cognitive impairment due to Alzheimer’s disease (AD) or AD-related dementias. For some patients, early detection of cognitive impairment enables access to disease-modifying therapies. For all patients, it allows access to psychosocial supports. Patients typically first present their concerns about their cognition to a primary care provider, but in this setting, cognitive impairment is commonly underdiagnosed. There is also high variability in how cognitive evaluations are performed. We sought to understand barriers to and facilitators of cognitive evaluations in primary care, map barriers to implementation strategies, and gain consensus from stakeholders on possible strategies to improve dementia diagnosis in primary care.MethodsSemi-structured interviews conducted with primary care providers (PCPs). We used the Consolidated Framework for Implementation Research to inform our question guide and analysis, and incorporated chart-stimulated recall – using actual patients who had cognitive complaints who had presented to these providers – to understand clinicians’ medical decision-making processes. These data were used to map identified barriers and facilitators to targeted implementation strategies. Then, this candidate list of strategies was presented to an expert stakeholder panel including clinicians and clinical operations specialists. Through a modified Delphi process, the list was narrowed to select the most promising strategies to incorporate in an intervention to improve cognitive evaluations in primary care.ResultsTwenty PCPs were interviewed and mentioned barriers included lack of expertise to perform or interpret an assessment, time pressures, lack of incentives, competing priorities, lack of decision-making supports, and limited access to dementia specialists. Facilitators included the presence of an informant or caregiver and having additional staff to conduct cognitive testing. Implementation mapping resulted in a list of 15 candidate strategies. Using the modified Delphi process, these were narrowed to six.ConclusionsWe used a rigorous process to identify barriers to and facilitators of cognitive assessments in primary care, identify promising implementation strategies to address these barriers, and obtain the feedback of front-line users on these strategies. This holds substantial promise for improving cognitive assessments in primary care in future implementation trials.

Identification of adipose tissue-derived exosomal microRNA as a novel causal biomarker for cognitive impairment in type 2 diabetes mellitus: Triangulating evidence from Mendelian randomization and multicentre population studies.

To explore serum exosomal microRNAs (miRNAs) as risk biomarkers for early detection of cognitive impairment in type 2 diabetes mellitus (T2DM) patients. This study included two phases: a discovery phase and a validation phase. To detect adipose tissue exosomal biomarkers for T2DM patients, small RNA sequencing was conducted on a discovery population consisting of six T2DM patients and five subjects with normal glucose tolerance. To identify miRNAs with causal effects on cognitive impairment, Mendelian randomization (MR) analysis using publicly available genome wide association studies (GWAS) datasets was performed. Relationships between serum exosomal miRNAs and cognitive impairment were evaluated in a training population of 207 T2DM patients, and further validated in an external population of 101 T2DM patients from multiple centres. In the discovery phase, 13 exosomal miRNAs were significantly upregulated in adipose tissue of T2DM patients. MR analyses identified that increased miR-125a-5p was causally associated with increased Alzheimer's disease (AD) risk (OR = 1.231, 95% CI 1.062-1.426). In the validation phase, higher serum exosomal miR-125a-5p levels were related to increased amnestic mild cognitive impairment (aMCI) risk (OR = 1.066, 95% CI 1.030-1.103) and reduced left hippocampal body volume (r = -0.189, p < 0.05), achieving an area under the curve (AUC) of 0.728 for identifying aMCI in T2DM patients. External validation confirmed a diagnostic AUC of 0.738. Serum exosomal miR-125a-5p derived from adipose tissue can serve as a causal biomarker for cognitive impairment in T2DM patients.

Wearable Frontal EEG for Screening Cognitive Impairment

AbstractBackgroundAlzheimer's disease (AD) is a neurodegenerative disorder that is characterized by progressive cognitive decline. Early detection of AD is essential for timely intervention and treatment. The objective of this study is to explore the potential usefulness of virtual reality (VR) tasks combined with frontal EEG recordings for early detection of cognitive impairment in individuals with SCD (subject cognitive decline), MCI (mild cognitive impairment), and AD (Alzheimer's disease). We aimed to analyze the results and compared them with those of a control group with normal cognition.MethodResting‐state EEG was recorded in the awake‐resting state with closed eyes, and task‐induced EEG recordings were obtained during VR tasks from 108 participants. The participants included those with SCD, MCI, and AD, and they were compared with 18 subjects with normal cognition. Alpha reactivity was determined by measuring the decrease in relative alpha power over frontal electrodes.ResultThis study observed a significant decrease in alpha reactivity in the order of SCD, MCI, and AD compared to the control group with normal cognition in EEG recordings (P‐value &lt; 0.05). Additionally, the decibel power ratio of the EEG recorded during cognitive tasks exhibited a reversed order of decrease compared to the resting state across the same diagnostic groups.ConclusionThis study shows that when performing cognitive tasks with a virtual reality (VR) device on the forehead, distinct features for Alzheimer's disease are observed compared to the healthy control group. While larger and longer‐term studies are necessary, these findings suggest that cost‐effective digital biomarkers using measurable EEG could be valuable for early Alzheimer's diagnosis.

The Relationship between Subjective Visual Impairment and Domain‐Specific Cognitive Performance in Cognitively Normal Older Adults and MCI Patients

AbstractBackgroundAge is the greatest risk factor for self‐reported visual impairment (VI) and Alzheimer’s Disease (AD). While prior studies demonstrate associations between VI and cognitive screening measures in older adults, the relationship between VI and specific cognitive domains remains unexplored. We examined the relationship between subjective VI and executive function, attention, visuospatial construction, episodic memory, and processing speed.MethodParticipants were older adults aged 55‐80: 58 cognitively unimpaired (CU) (MoCA&gt;=26, CDR=0) and 19 MCI patients (MoCA&gt;19, &lt;24, CDR=.5). Subjective VI was measured with the National Eye Institute Visual Function Questionnaire (NEI‐VFQ). A neuropsychological battery including the MoCA, CDR, Digit Symbol Substitution Test (DSST), Repeatable Battery for the Assessment of Neuropsychological Status (RBANS), and Free and Cued Selective Reminding Test (FCSRT) was administered to assess cognition. Partial correlations controlling for age and years of education compared subjective VI scores to cognitive scores. A multivariate, repeated measures ANCOVA was run on a subset (N=23) of participants with 1 year follow‐up data to determine whether group performance differed over time on NEI‐VFQ and DSST.ResultThere were significant correlations between scores on the composite NEI‐VFQ and both the DSST and visuospatial sub‐scale of the RBANS in the CU and patient groups (p&lt;.05). CU adults showed a greater practice effect over time on the DSST than the patient group, who declined in performance. Specifically, CU APOE E4 carriers had a larger increase in DSST score over 12 months than non‐carriers, along with a larger increase in subjective VI on the NEI‐VFQ.ConclusionAttention, processing speed, visuospatial construction, and executive function performance was related to a standardized measure of subjective VI in CU older adults and MCI patients. After 12 months, CU APOE E4 carriers showed a stronger practice effect than CU non‐ carriers on the DSST, concomitant with increase in subjective VI. Subjective measures of VI may serve as a tool in early detection of cognitive impairment, although increases in VI over time co‐occurred with increased practice effects in CU APOE E4 carriers. Longitudinal studies are ongoing to examine the predictive value of subjective VI on cognitive impairment.

Lessons Learned from a Global, Multisite Early Detection Program in Primary Care Implementation Program

AbstractDavos Alzheimer’s Collaborative Healthcare System Preparedness (DAC‐SP) aims to catalyze global healthcare system transformation, providing patients with quicker access to life‐changing innovations and therapies. Utilizing implementation science, the DAC‐SP Early Detection flagship program launched in 2021, engaging seven healthcare systems across six countries (Brazil, Jamaica, Japan, Mexico, Scotland, and the United States). The program's primary aim was to increase the rate of early detection of cognitive impairment by integrating commercially available digital cognitive assessments (DCAs) into primary care and other non‐specialty care settings. Patients with suspected cognitive decline based on the DCA were then offered a blood‐based biomarker (BBM) to determine the likelihood of amyloid presence in addition to customary follow‐up testing.Over the past two years, health system leaders from the early detection program were convened into a Community of Practice (CoP) that met monthly to discuss operational and clinical challenges, collaborate on solutions, and exchange learnings. As part of the CoP meetings, site leaders identified common implementation challenges for DCAs and BBMs in real‐world primary care settings. These challenges span various levels of health systems, including policy (e.g., reimbursement and procurement mechanisms, lack of guidance and consensus on workflow), the outer setting (e.g., availability/capacity of brain health expertise, lab infrastructure, fit with administrative systems), the inner setting (e.g., training and equipment required for blood draws, shipping logistics, administrative support) and understanding clinician knowledge, attitudes and behavior (e.g., level of comfort or experience with cognitive assessments, interpretation and disclosure of BBM results, access to training, task shifting resources, support from peers). Education about BBM specifications, use cases, and integration into clinical routines was essential in understanding the adoptability and scalability of BBMs in primary care settings.

Digital Cognitive Screening in Primary Care: Feasibility, Acceptability, and Usability of the MyCog Measurement System

AbstractBackgroundAnnual cognitive screening in older adults is essential for early detection of cognitive impairment, yet less than half of cases are detected in primary care. We introduce two innovative app‐based screeners that help overcome barriers to routine cognitive screening. MyCog is a tablet app that is self‐administered in person during the rooming process for a primary care visit (Figure 1). MyCog Mobile is a smartphone app that is self‐administered remotely prior to a primary care visit (Figure 2), which can save clinicians even more time during an in‐person visit. We present evidence from two studies: 1) an evaluation of the adoption and usability of MyCog in primary care; and 2) a pilot study of the reliability and usability of MyCog Mobile.MethodIn Study 1, the implementation of MyCog in 10 primary care clinics was assessed as part of a pragmatic trial evaluating cognitive screening for adults aged 65+ during Annual Wellness Visits (AWVs). Using EHR data, we assessed adoption via the portion of patients who completed MyCog. In Study 2, 51 adults 65+ self‐administered MyCog Mobile remotely on their own smartphones twice about two weeks apart and rated its usability on the Simplified System Usability Scale (S‐SUS). We examined the reliability, correlations with education, and usability ratings.ResultIn Study 1, of the 3921 eligible patients, 1263 completed MyCog (32.2%), with individual clinic compliance ranging from 11% to 68% of patients. Feedback revealed several addressable clinician and patient barriers, including patient frustration on memory tests, unclear test progress, and need for a Spanish version, and minor design flaws. In Study 2, all participants successfully completed each measure, and reliability metrics were broadly acceptable (Table 1). The MyPictures and education correlation was lower than expected. Ratings on the S‐SUS indicated good usability of MyCog Mobile (M = 73.78, SD = 20.11).ConclusionMyCog demonstrated good adoption by primary care clinics in a pragmatic trial, which may be improved through design and implementation changes. Some limitations of MyCog may be addressed with MyCog Mobile, which demonstrated preliminary feasibility evidence. Practice implications and next steps are discussed.

Using Machine Learning to Explore Multimodal Digital Markers for Early Detection of Cognitive Impairment in Alzheimer’s Disease

AbstractBackgroundRecent technological advancements have revolutionized our approach to healthcare, enabling us to harness the potential of smartphones and wearables to collect data that can be used to characterize Alzheimer’s disease (AD) heterogeneity and to develop digital biomarkers. Our focus is to create comprehensive cross‐domain digital datasets and establish an infrastructure that allows for seamless data sharing. Central to accelerating the potential of digital biomarkers for more accurate and early detection is privacy‐protecting data access, which when combined with deep molecular phenotyping, will enhance our understanding of the biological mechanisms underlying clinical expression.MethodsIn the preliminary phase of this project, we analyzed data from 64 participants from the Boston University Alzheimer’s Disease Research Center and encompassing approximately 1480 variables. Our analysis approach leverages a novel machine learning (ML) technology, Attractor AI, that is capable of differentiating causal and non‐causal subpopulations within small patient or study populations and large volumes of measures, enhancing the efficacy of predictive models.ResultsWe were able to subcategorize 50% of the 27 cognitively impaired (CI) subjects. A notable discovery was a distinct subpopulation of 8 individuals, 7 of whom were CI, characterized significantly by higher sleep‐derived variables such as various desaturation thresholds and periodicity measures (p=0.008‐0.00007). Additionally, incorporating maximum heart rate, revealed another group of 8 subjects, 6 identified as CI, distinguished by elevated heart rate during one or more of their measuring instances (p=10‐10).ConclusionsWhile these results are preliminary, they signal a promising direction to cluster subgroups of people along similar dimensions, laying the groundwork for a precision medicine solution. Our future endeavors include expanding the scope of multimodal digital data to encompass aspects like vocalization and speech patterns derived from cognitive assessments, gait analysis, physical activity, and other cognitive tests. The integration of these diverse data streams, coupled with our preliminary sleep analysis findings, has the potential to result in a robust and accurate subgrouping system for CI to help identification of AD risk pathways that might be amenable to early intervention and either delay or prevent potential transition to AD.

Diffusion Kurtosis Imaging provide complementary information over structural MRI in the assessment of hippocampal subfields in patients with cognitive decline

AbstractBackgroundHippocampal atrophy is an established biomarker of neurodegeneration in Alzheimer's disease, affecting specific subfields (De Flores, La Joie and Chételat, 2015). In this study, we used 7T MRI and advanced diffusion MRI (dMRI) to investigate the relationship between hippocampal subfield volumes and microstructure and assess their sensitivity to cognitive impairment.MethodSeventeen cognitively impaired (CI; age: 69±8, M/F: 12/5, MMSE: 28) and 22 cognitively unimpaired subjects (CU; age: 62±10, M/F: 6/16) were recruited in the context of the COSCODE project (Ribaldi et al., 2021). MRI was performed at 7 Tesla, including a high resolution T2w FSE sequence for hippocampal segmentation and multi‐shell diffusion MRI sequence for kurtosis tensor imaging (DKI, Jensen and Helpern, 2010) fitting. Hippocampal subfields were segmented using Automated Segmentation of Hippocampal Subfields (ASHS) package. Microstructure analysis included DTI and DKI (Figure 1). Mean values for each diffusion metric were calculated in subfields of the ASHS atlas (Xie et al. 2016). Subfield volumes were normalised by the total intracranial volume while dMRI metrics were residualized by age via a linear regression trained only on CU and then correlated to the volumetric data across within each group (statistical significance: p&lt;0.05). Lasso logistic regression was used to identify the metrics that better discriminated CI from CU.ResultFew hippocampal subfields’ volumes correlated with dMRI metrics (Figure 2). In the CU group, DG volume showed negative associations with diffusion metrics, primarily RD (r:‐0.59, p=0.007), MD(r:‐0.56, p=0.012), MK(r:‐0.46, p=0.043) and AK(r:‐0.62, p=0.005). In the CI group, negative correlations between BA35 (r:‐0.70, p=0.008) and CA3 (r:‐0.57, p=0.041) volumes were found with AK. When discriminating CI from CU, the logistic regression found the ERC volume as the most important metric, followed by the MK of the DG, and AK of CA2 (Figure 3).ConclusionThe disconnection between dMRI metrics and volumetric measures and the selection of DKI metrics in discriminating CI from CU participants suggest that DKI metrics may act as complementary biomarkers to subfield volumes for early detection of cognitive impairment.

Digital Clock Assessment in South Asian Setting: Insights from the P‐CARRS study

AbstractBackgroundDetection of presymptomatic individuals or those with subtle cognitive changes in midlife may prevent or slow the course of Alzheimer's Disease by identifying candidates for disease‐modifying treatments. Utilizing newer delivery approaches involving digital measures shows promise for cognitive phenotyping, early detection, ease of administration, and scoring, particularly in low‐resource settings. However, the feasibility of these approaches, along with their association with demographics and their effectiveness in detecting fine‐grained aspects of cognitive performance in low‐resource settings, remains unclear. We conducted a pilot study evaluating digital cognitive screening using an iPad‐administered clock drawing test in a diverse South Asian population residing in India.MethodsIndividuals ≥ 50 years old (median: 61.0 years; range: 50‐89 years; female: 56%; education: up to primary schooling: 8%; high school to secondary: 65%; graduation and above: 27.4%) nested within the ongoing population‐based longitudinal Precision‐CARRS study representing socio‐demographically and linguistically diverse adults from Delhi and Chennai in India (n=300; 150/site) were studied. Participants were administered the FDA‐approved, iPad‐based clock drawing test using a commercially available Linus Health Digital Clock (DCTclock) screening tool. The DCTclock summary score, ranging from 0 to 100, was derived through AI analytics and provided various features, including drawing efficiency, information processing, simple motor skills, and visuospatial reasoning. Performance on the traditionally administered paper/pencil version of the Mini‐Cog test incorporating clock drawing and three‐item recall was used to classify participants as cognitively impaired (0‐2 points) or intact (3‐5 points).ResultsDCTclock was administered in ∼3‐4 minutes, and only 0.7% of clocks were unanalyzable due to missing components. Older age and lower education (p&lt;0.0001) were associated with worse DCTclock summary scores. There were no sex differences in performance (p=0.90). Compared to those cognitively intact based on the Mini‐Cog scores (n=252), the impaired group (n=48) performed significantly worse on the DCTclock summary score and spatial reasoning components (Table).ConclusionScreening for cognitive status using digital clock measurement is highly feasible and takes a short amount of time to administer in low‐resource settings. The DCTclock provides fine‐grained measures of performance not available on traditional paper/pencil measures, which may prove sensitive to early detection of cognitive impairment.

Implementation of Digital Cognitive Assessment for Early Detection of Cognitive Impairment: A Davos Alzheimer’s Collaborative Flagship Site

AbstractBackgroundIn Japan, the number of patients with dementia is increasing and there is a shortage of human resources who can conduct detailed psychological examinations, and the development and implementation of new tools is essential for improving dementia care. Japan is one of seven sites across the globe focusing on healthcare system preparedness as part of the Davos Alzheimer’s Collaborative Early Detection flagship program. As part of this program, Japan implemented new tools including a digital cognitive assessment (DCA) and blood‐based biomarker (BBM) to increase rates of early detection of cognitive impairment.MethodsPatients 60+ years old were given a Mini‐Mental State Examination (MMSE), a common pen‐and‐paper test of cognitive function, and based on the score, primary care providers recorded a pre‐referral diagnosis. Patients who scored abnormal or borderline on the MMSE were then referred to a specialist center and completed a full work‐up including a clinical and psychological examination, a DCA (Cogstate Cognigram), blood test for reversible causes, imaging (i.e., MRI, perfusion SPECT), and a BBM (C2N PrecivityAD). Based on the results from the full clinical work‐up, the specialist provided a post‐referral diagnosis.ResultsIn the early detection program, the MMSE detected 46 total patients with cognitive impairment (MMSE scores ≥10 and &lt;28), with 33 (72%) cognitively abnormal and 13 (28%) borderline scores. The number and type of pre‐referral diagnostic assessment made based on the MMSE scores and from the post‐referral diagnosis following the full clinical work‐up are provided in Table 1. There was good agreement between the initial and final diagnosis, with 8 total changes in diagnosis. The most notable changes came in the Alzheimer Disease diagnosis category where there was a net increase in 4 diagnoses of AD (2 were reversed and 6 new diagnoses of AD were added).ConclusionComprehensive assessments, including implementing new tools like a DCA and BBM, along with traditional clinical examination are necessary to make an accurate diagnosis. Initial findings from the Japan flagship site contribute to the global learnings to help speed and scale future innovations and developments in order to reach the right patients at the right time.

Glutaric acid associates with early detection of cognitive impairment deciphering the association with depression

AbstractBackgroundThe initiation of cognitive impairment is triggered by a myriad of pathological events occurring decades before clinical symptoms manifest. Perturbed glucose and fatty acid metabolism notably contribute to the development of cognitive impairment, progressing further into clinical dementia. These metabolic alterations are evident in plasma through changes in specific metabolites. Notably, these changes are characteristic features in the pathophysiology of depression, a significant risk factor for cognitive impairment. This project aims to establish a blood‐based biomarker signature for early identification of cognitive changes in individuals with depression. Moreover, it aims to assist in diagnosing and understanding disease progression by quantifying plasma metabolite levels linked to fatty acid metabolism in samples obtained from participants in the Sydney Memory and Ageing Study.MethodPlasma samples underwent analysis to quantify fatty acids and carnitines using chemical derivatization through UPLC‐MRM/MS coupled with a 4000 QTRAP mass spectrometer which was equipped with an electrospray ion (ESI) source and operated in the multiple‐reaction monitoring (MRM) mode with negative‐ion (‐) detection.10‐μL aliquots of each of the resultant calibration solutions and each of the sample solutions were injected for analysis.ResultANCOVA, revealed significant changes in the levels of several biomarkers including glycolic acid (p‐value: 0.033), C181 carnitine (p‐value: 0.013), glutaric acid (p‐value: 0.08) levels in groups with Mild Cognitive Impairment (MCI) and MCI with comorbid depression compared to healthy groups. ROC curve analysis demonstrated that glutaric acid effectively distinguished between MCI and non‐MCI participants. The optimal threshold for classifying participants into MCI and non‐MCI groups was determined using the Youden Index to dichotomize them based on high and low biomarker levels. Logistic regression analysis, adjusting for age, sex, APOE ε4 genotype, and comorbid depression, revealed a significant association between glutaric acid levels and cognitive impairment in the context of depression, with a p‐value of &lt;0.001 and an odds ratio of 2.49.ConclusionGlutaric acid proved to be a reliable discriminator between participants with MCI and those without MCI. The aforementioned findings revealed a strong correlation between glutaric acid and cognitive impairment associated with depression which could be used as biomarker for early detection of cognitice impairment.

One-class classification with confound control for cognitive screening in older adults using gait, fingertapping, cognitive, and dual tasks

Background and Objectives:Early detection of cognitive impairment is crucial for timely clinical interventions aimed at delaying progression to dementia. However, existing screening tools are not ideal for wide population screening. This study explores the potential of combining machine learning, specifically, one-class classification, with simpler and quicker motor-cognitive tasks to improve the early detection of cognitive impairment. Methods:We gathered data on gait, fingertapping, cognitive, and dual tasks from older adults with mild cognitive impairment and healthy controls. Using one-class classification, we modeled the behavior of the majority group (healthy controls), identifying deviations from this behavior as abnormal. To account for confounding effects, we integrated confound regression into the classification pipeline. We evaluated the performance of individual tasks, as well as the combination of features (early fusion) and models (late fusion). Additionally, we compared the results with those from two-class classification and a standard cognitive screening test. Results:We analyzed data from 37 healthy controls and 16 individuals with mild cognitive impairment. Results revealed that one-class classification had higher predictive accuracy for mild cognitive impairment, whereas two-class classification performed better in identifying healthy controls. Gait features yielded the best results for one-class classification. Combining individual models led to better performance than combining features from the different tasks. Notably, the one-class majority voting approach exhibited a sensitivity of 87.5% and a specificity of 75.7%, suggesting it may serve as a potential alternative to the standard cognitive screening test. In contrast, the two-class majority voting failed to improve the low sensitivities achieved by the individual models due to the underrepresentation of the impaired group. Conclusion:Our preliminary results support the use of one-class classification with confound control to detect abnormal patterns of gait, fingertapping, cognitive, and dual tasks, to improve the early detection of cognitive impairment. Further research is necessary to substantiate the method’s effectiveness in broader clinical settings.

Cognitive function and its determinants in elderly Indonesians residing in long-term care: Insights from a cross-sectional study

Background Multiple medical conditions arising from reduced physical and physiological functioning, including cognitive decline, manifest in older persons. This study aims to examine the relationship between cognitive function and associated risk factors in older persons living in long-term care facilities in Indonesia. Methods This study involved 350 elderly individuals residing in long-term care institutions. A cross-sectional design utilizing an analytical survey methodology was implemented. Data were gathered via interviews employing a demographic questionnaire and the Montreal Cognitive Assessment (MoCA). Statistical analysis was conducted using SPSS (version 23). Results Univariate analysis demonstrated significant correlations between cognitive performance and gender, ethnicity, level of education, medical history, subjective memory issues, smoking habits, alcohol consumption, dietary intake of fruits and vegetables, and employment history (p &lt; 0.05). Higher education (OR = 0.69, 95% CI: 0.56–0.84) and reduced subjective memory complaints (OR = 0.29, 95% CI: 0.20–0.44) correlated positively with enhanced cognitive function, but alcohol intake (OR = 6.79, 95% CI: 2.42–19.1) correlated with impaired cognitive function. Conclusions the level of education, subjective memory complaints, and alcohol intake are substantially correlated with cognitive performance in older persons residing in long-term care facilities. Evaluating demographic characteristics in elderly individuals can assist healthcare professionals in the early detection of cognitive impairment, facilitating prompt interventions in long-term care environments.

Early Identification of Cognitive Impairment in Community Environments Through Modeling Subtle Inconsistencies in Questionnaire Responses: Machine Learning Model Development and Validation.

The underdiagnosis of cognitive impairment hinders timely intervention of dementia. Health professionals working in the community play a critical role in the early detection of cognitive impairment, yet still face several challenges such as a lack of suitable tools, necessary training, and potential stigmatization. This study explored a novel application integrating psychometric methods with data science techniques to model subtle inconsistencies in questionnaire response data for early identification of cognitive impairment in community environments. This study analyzed questionnaire response data from participants aged 50 years and older in the Health and Retirement Study (waves 8-9, n=12,942). Predictors included low-quality response indices generated using the graded response model from four brief questionnaires (optimism, hopelessness, purpose in life, and life satisfaction) assessing aspects of overall well-being, a focus of health professionals in communities. The primary and supplemental predicted outcomes were current cognitive impairment derived from a validated criterion and dementia or mortality in the next ten years. Seven predictive models were trained, and the performance of these models was evaluated and compared. The multilayer perceptron exhibited the best performance in predicting current cognitive impairment. In the selected four questionnaires, the area under curve values for identifying current cognitive impairment ranged from 0.63 to 0.66 and was improved to 0.71 to 0.74 when combining the low-quality response indices with age and gender for prediction. We set the threshold for assessing cognitive impairment risk in the tool based on the ratio of underdiagnosis costs to overdiagnosis costs, and a ratio of 4 was used as the default choice. Furthermore, the tool outperformed the efficiency of age or health-based screening strategies for identifying individuals at high risk for cognitive impairment, particularly in the 50- to 59-year and 60- to 69-year age groups. The tool is available on a portal website for the public to access freely. We developed a novel prediction tool that integrates psychometric methods with data science to facilitate "passive or backend" cognitive impairment assessments in community settings, aiming to promote early cognitive impairment detection. This tool simplifies the cognitive impairment assessment process, making it more adaptable and reducing burdens. Our approach also presents a new perspective for using questionnaire data: leveraging, rather than dismissing, low-quality data.

Improving Early Detection of Cognitive Impairment in Older Adults in Primary Care Clinics: Recommendations From an Interdisciplinary Geriatrics Summit.

As the population ages, the prevalence of cognitive impairment due to neurodegenerative diseases such as Alzheimer disease (AD) is expected to double in the United States to nearly 14 million over the next 40 years. AD and related dementias (ADRD) are a leading cause of morbidity and mortality and among the costliest to society. Although emerging biomedical interventions for ADRD focus on early stages and are currently limited to AD, care management can benefit patients with ADRD across the disease course. Moreover, some causes of cognitive impairment are modifiable, and optimal overall management may slow or prevent additional decline. Nevertheless, a sizable proportion of cases of cognitive impairment among older adults remain undiagnosed. Primary care practitioners are often the first health care professionals to encounter cognitive concerns or to be able to observe changes in function resulting from cognitive impairment; hence, they have much to contribute to population health solutions for detecting cognitive impairment among older adults. In this report, we present key points and gaps in knowledge about methods for detecting cognitive impairment in primary care clinics. These were developed via an interdisciplinary Geriatrics Summit hosted by the National Academy of Neuropsychology in 2022, attended by representatives of national organizations engaged in work to improve care of older adults. We propose a novel workflow to facilitate detecting cognitive impairment during routine primary care, focusing on opportunities provided by the annual wellness visit, a preventive visit available to Medicare beneficiaries, along with additional recommendations and opportunities for clinical practice and research.

Prevalence of Unrecognized Cognitive Impairment in Federally Qualified Health Centers

There is a need for early and equitable detection of cognitive impairment among older adults. To examine the prevalence of unrecognized cognitive impairment among older adults receiving primary care from federally qualified health centers (FQHCs). This cross-sectional study was conducted at 5 FQHCs providing primary care in Indianapolis, Indiana, between 2021 and 2023. Participants were adults aged 65 years and older, without a diagnosis of mild cognitive impairment (MCI), dementia, or severe mental illness. Data analysis was performed from September 2023 to April 2024. The primary outcome was a diagnosis of dementia or MCI, as determined by an interdisciplinary clinical team using data from structured patient and study partner interviews, medical record reviews, and a detailed cognitive assessment, including neuropsychological testing. Differences between participants determined to have normal cognition, MCI, and dementia were assessed statistically using analysis of variance for continuous variables, χ2 or Fisher exact tests for categorical variables, or Fisher exact test alone when expected cell counts were 5 or less. A total of 844 eligible individuals were consecutively approached, 294 consented to participate, and 204 completed the study (mean [SD] age, 70.0 [5.1] years; 127 women [62.3%]). One hundred eight participants (52.9%) were African American, 5 (2.5%) were Hispanic, 199 (97.5%) were not Hispanic, and 90 (44.1%) were White. The mean (SD) duration of education was 13.1 (2.6) years, and the mean (SD) Area Deprivation Index score was 78.3 (19.9), indicating a high level of neighborhood disadvantage. In total, 127 patients (62.3%) met the diagnostic criteria for MCI, 25 (12.3%) had dementia, and 52 (25.5%) had no cognitive impairment. Compared with non-Hispanic White individuals and after adjusting for age, sex, and education level, African American individuals were more than twice as likely to have MCI or dementia (odds ratio, 2.73; 95% CI, 1.38-5.53; P = .02). This cross-sectional study found that unrecognized cognitive impairment is ubiquitous among older adults from underrepresented, minoritized racial and ethnic groups and those who are socially vulnerable receiving primary care from FQHCs. To overcome the disparity in early detection of cognitive impairment, timely, equitable, scalable, and sustainable detection approaches need to be developed.