Routine cognitive screening in the elderly may facilitate earlier diagnosis of neurodegenerative diseases and access to care and resources for patients and families. However, despite growing rates of Alzheimer's and related disorders (ADRD), the availability and implementation of cognitive screening for older adults in the US remains quite limited. Remote cognitive assessment via smartphone app may reduce several barriers to more widespread screening. We examined the validity of a remote app-based cognitive screening protocol in healthy older adults by examining remote task convergence with standard-person assessments and cerebral amyloid (Aß) status as an AD biomarker.

Participants (N =117) were cognitively unimpaired adults aged 60-80 years (67.5% female, 88% White, 75% education > 16 years). A portion had Aß PET imaging results available from prior research participation [(Aß positive (Aß+) n =26, and Aß negative (Aß-) n = 44]. A modified Telephone Interview for Cognitive Status (TICSm) cutoff score of >34 was used to establish unimpaired cognition. Participants completed 8 consecutive assessment days using Mobile Monitoring of Cognitive Change (M2C2), a smartphone app-based testing platform developed as part of the National Institute of Aging's Mobile Toolbox initiative. Brief (i.e., 3-4 minute) M2C2 sessions were assigned daily within morning, afternoon, and evening time windows. Tasks included measures of visual working memory (WM), processing speed (PS), and episodic memory (EM) (see Thompson et al., 2022). Participants then completed a battery of standard neuropsychological assessments in-person at a follow-up visit.

Participants completed 22.6 (SD = 2.6) out of 24 assigned sessions (3 sessions x 8 days) on average. Performance on all M2C2 tasks decreased significantly with age. Women performed significantly better on WM and EM tasks relative to men. There were no detectable significant differences in performance by race or education. Shorter mean reaction time on M2C2 PS trials predicted faster Trails A and B completion (ß = .26, p < .01, 95% CI [3.8, 23.3] and ß = .20, p < .05, 95% CI [.23, 6.8], respectively). Greater mean M2C2 WM accuracy predicted longer maximum backward digital span (ß = .24, p = .01, 95% CI [.02, .16]). Greater mean M2C2 EM accuracy predicted stronger Logical Memory delayed recall (ß = .33, p < .001, 95% CI [.004, .012]) and total immediate recall on the Free and Cued Selective Reminding Test (ß = .19, p < .05, 95% CI [.000, .006]). Moreover, EM significantly distinguished Aß- and Aß+ individuals (t (68) = 3.0, p < .01) with fair accuracy (AUC = .72).

Mean performance across 8-days on each M2C2 task predicted same-domain cognitive task performance on a standard assessment battery, with medium effect sizes. Performance on the EM task was also sensitive to cerebral Aß status, consistent with subtle memory changes implicated in the preclinical stage of AD. These findings support the validity of this remote testing protocol in healthy older adults, with implications for future efforts to facilitate accessible and sensitive cognitive screening for early detection of ADRD. Limitations include the restricted generalizability of this primarily white and college educated sample.

Vascular complications, including elevated body mass index (BMI), are known risk factors for cognitive impairment. Obtaining a cognitive baseline is commonplace in pre-surgical protocols, including for Parkinson’s disease and epilepsy. Currently, routine evaluations for bariatric surgery candidates do not include neuropsychological assessment. This setting provides a unique opportunity to identify cognitive profiles of younger individuals at risk for cognitive impairment. Here, we argue for the standard implementation of a brief, online cognitive battery via telemedicine, to enhance existing protocols of bariatric pre-surgical evaluations.

Nineteen bariatric surgery candidates were referred to a private neuropsychological assessment practice for pre-surgical cognitive/psychological evaluation. Assessments were conducted by a neuropsychologist and a psychology graduate student, via remote video conferencing, between April 2020 and June 2022. Candidates completed a clinical interview, intake form, and the Behavior Rating Inventory of Executive Function for Adults (BRIEF-A), and were administered a battery of cognitive measures: Wechsler Test of Adult Reading (WTAR), Hopkins Verbal Learning Test-Revised (HVLT-R); and select subtests from the TestMyBrain Digital Neuropsychology Toolkit: Trail Making Tests A and B (TMT-A/B), Matrix Reasoning (MR), Digit Span Forward and Backward (DSF, DSB), Gradual Onset Continuous Performance Test (CPT), and Simple and Choice Reaction Time (SRT, CRT). Descriptive statistics were conducted to analyze sample demographics. Raw scores on cognitive measures were converted to z-scores and averaged across the sample.

The average age at evaluation was 38.2 (9.6) years and average pre-surgical BMI was 46.6k/m2 (9.3), indicating morbid obesity (BMI>40k/ m2). Ten (52.6%) candidates identified as female, 10 (52.6%) identified as White, 8 (42.1%) had 12 years of education or less, 4 (21.1%) were unemployed, and 9 (47.4%) had comorbid psychiatric diagnoses. BRIEF-A sub-scales were within the average range (T’s= 47.9 - 52.9, SD’s [10.1 - 12.8]). Estimated premorbid IQ was average at 102.7 (11.4). Neuropsychological data revealed group performance within the average range on DSF and DSB (z’s= 0.00), TMT-A (z= -0.16), MR (z= -0.53), CPT (z= -0.39), and HVLT-R False Positives (z= 0.05) and Recognition Discriminability (z= -0.44). SRT (z= -0.70), CRT Accuracy (z= -1.37), TMT-B (z= -0.79), HVLT-R Total Recall and Percent Retained (z’s= -0.88), and Delayed Recall and True Positives (z’s= -1.27) were low average.

In this sample of pre-surgical bariatric candidates with average intelligence, baseline evaluations revealed mild deficits in reaction time accuracy, visual motoric set-shifting, and verbal learning/memory. These deficits may be the result of microvascular changes in the brain secondary to physical compromise. Results provide additional insight into potential early-onset executive dysfunction, psychomotor slowing, and verbal learning/memory difficulties. In addition to these relative areas of neuropsychological weakness, candidates demonstrated relative strengths in attention, working memory, and visuospatial functioning. These insights provide pre-surgical evaluators with additional information to tailor recommendations and treatment approaches that foster surgical success. With a remote, concise, easy-to-administer battery of tests, routine neuropsychological assessment for bariatric surgery candidates is both a feasible and a useful tool for identifying areas of cognitive strengths and weaknesses. Documenting a patients’ cognitive baseline can assist with monitoring long-term vascular risk-factors and potential cognitive impairment.

Neuropsychological (NP) tests are increasingly computerized, which automates testing, scoring, and administration. These innovations are well-suited for use in resource-limited settings, such as low- to middle- income countries (LMICs), which often lack specialized testing resources (e.g., trained staff, forms, norms, equipment). Despite this, there is a dearth of research on their acceptability and usability which could affect performance, particularly in LMICs with varying levels of access to computer technology. NeuroScreen is a tablet-based battery of tests assessing learning, memory, working memory, processing speed, executive functions, and motor speed. This study evaluated the acceptability and usability of NeuroScreen among two groups of LMIC adolescents with and without HIV from Cape Town, South Africa and Kampala, Uganda.

Adolescents in Cape Town (n=131) and Kampala (n=80) completed NeuroScreen and questions about their use and ownership of, as well as comfort with computer technology and their experiences completing NeuroScreen. Participants rated their technology use -comfort with and ease-of-use of computers, tablets, smartphones, and NeuroScreen on a Likert-type scale: (1) Very Easy/Very Comfortable to (6) Very Difficult/Very Uncomfortable. For analyses, responses of Somewhat Easy/Comfortable to Very Easy/Comfortable were collapsed to codify comfort and ease. Descriptive statistics assessed technology use and experiences of using the NeuroScreen tool. A qualitative question asked how participants would feel receiving NeuroScreen routinely in the future; responses were coded as positive, negative, or neutral (e.g., “I would enjoy it”). Chi-squares assessed for group differences.

South African adolescents were 15.42 years on average, 50.3% male, and 49% were HIV-positive. Ugandan adolescents were 15.64 years on average, 50.6% male, and 54% HIVpositive. South African participants were more likely than Ugandan participants to have ever used a computer (71% vs. 49%; p<.005), or tablet (58% vs. 40%; p<.05), whereas smartphone use was similar (94% vs 87%). South African participants reported higher rates of comfort using a computer (86% vs. 46%; p<.001) and smartphone (96% vs. 88%; p<.05) compared to Ugandan participants. Ugandan adolescents rated using NeuroScreen as easier than South African adolescents (96% vs. 87%; p<.05).). Regarding within-sample differences by HIV status, Ugandan participants with HIV were less likely to have used a computer than participants without HIV (70% vs. 57%; p<.05, respectively).The Finger Tapping test was rated as the easiest by both South African (73%) and Ugandan (64%) participants. Trail Making was rated as the most difficult test among Ugandan participants (37%); 75% of South African participants reported no tasks as difficult followed by Finger Tapping as most difficult (8%). When asked about completing NeuroScreen at routine doctor’s visits, most South Africans (85%) and Ugandans (72%) responded positively.

This study found that even with low prior tablet use and varying levels of comfort in using technology, South African and Ugandan adolescents rated NeuroScreen with high acceptability and usability. These data suggest that scaling up NeuroScreen in LMICs, where technology use might be limited, may be appropriate for adolescent populations. Further research should examine prior experience and comfort with tablets as predictors NeuroScreen test performance.

Virtual testing can reduce cost and burdens, as well as increase access to clinical care. Few studies have examined the equivalency of virtual and in-person administration of standardized measures of executive functioning in children. During the COVID-19 pandemic, we utilized virtual administration of the Delis-Kaplan Executive Function System, Color-Word Interference Test (DKEFS-CW) in our ongoing longitudinal research study exploring outcomes in children clinically recovered from concussion compared to never-concussed peers. In the current study, we explore the equivalence of scores obtained via in-person and virtual administration of the DKEFS-CW in youth recovered from concussion and never-concussed controls.

Participants included 112 youth ages 10-18 (Mage=14.05 years, SD=2.296; 53.5 % Male) who completed the DKEFS-CW in-person (n=63) or virtually (n=49) as part of their involvement in the parent study. Of these, 38 were recovered from concussion (Mdays since injury— 91.21, SD=88.91), and 74 were never-injured controls. Virtual administration was done via Zoom by presenting digital scans of the DKEFS stimulus book using the screen-sharing function. Participants set up and joined the Zoom call from a secondary device (cell phone) that was set in a stable position to provide a view of their screen, mouse and keyboard setup. Group (in-person vs remote) differences in DKEFS-CW scores were examined using independent-samples t-tests for all subtest conditions (color naming, word reading, inhibition, and inhibition/switching). T-tests/chi-square tests were used to examine between-group differences in demographic variables (i.e., age, sex maternal education, IQ, concussion history). Demographic variables that were significantly different by group were then included as covariates in ANCOVA models examining the effect of administration context on performance.

There were no significant differences in DKEFS-CW scaled scores between those who were administered the measure in-person or virtually (Color Naming: Min-person=10.78, Mvirtual=10.08, t(110)=1.634, p=.105; Word Reading: Min-person=11.25, Mvirtual=10.92, t(110)=.877, p=.382; Inhibition: M in-person= 11.70, Mvirtual=11.24, t(110)=1.182, p=.240; Inhibition/Switching: Mi n-person= 11.29, Mvirtual=10.82, t(110)=1.114, p=.268). There were no significant between-group differences in concussion history, sex, maternal education or IQ. However, those who were administered the DKEFS-CW in-person (Mage=13.55) were significantly younger than those who were administered the measure virtually (Mage=14.69), t(110)=-2.777, p=.006. After controlling for age, there remained no significant relationship between administration context (in-person vs. virtual) and DKEFS-CW performance for any subtest condition (Color Naming: F(1,30)=.016, p=.889; Word Reading: F(1,76)=.655, p=.421; Inhibition: F(1,30)=.038, p=.847; Inhibition/Switching: F(1,30)=.015, p=.902).

The recommended practice for remote administration of DKEFS-CW is to have test stimuli presented flat on a table by a trained facilitator present with the examinees. Here, we provide preliminary evidence of equivalence between DKEFS-CW scores from tests completed in-person and those completed virtually with stimuli presented on a computer screen. Future studies are needed to replicate these findings in clinical populations with greater variability in executive function. Some clinical populations may also require more in-person support. Likewise, future studies may examine the role of trained facilitators or caregivers in the virtual testing process.

The COVID-19 pandemic created barriers to healthcare that necessitated changes in services to meet needs of individuals. With these changes, technological advances in computerized cognitive testing became critical. As researchers and clinicians accelerated adaptation of computerized testing formats, considerations for development and interpretation of such tools have proved imperative. One such computerized tool, RC21X, utilizes performance measurement software comprising 15 modules to evaluate an individual’s processing speed, memory, executive functions, and neuromotor coordination. Although initial data has revealed strong psychometric properties (Saganis et al., 2020), a need to explore various attributes of this web-based tool has emerged. The current study examined impact of dominant handedness on an RC21X neuromotor task.

The sample consisted of 602 participants: 553 (91.86%) were right-hand dominant and 49 (8.14%) were left-hand dominant. Of participants who identified their sex, 81.2% were male, 18.3% were female; 0.5% chose not to identify. Age ranged from 7-95 years (M = 41.21, SD = 18.81). This study focused on the RC21X Eye-Hand Coordination subtest. Using a Fitts’ Law paradigm, the module provided instruction for participants to alternately press the “A” and “L” keys on a keyboard as quickly and accurately as possible using only one upper extremity (UE) at a time (tested separately for right then left UE). We computed a one-way between groups multivariate analysis of variance (MANOVA) to investigate handedness differences on task performance. Dependent variables were individuals’ performances on right- and left-UE tasks; the independent variable was dominant handedness. We conducted preliminary assumption testing with no serious violations noted. We also separated the sample by dominant handedness to compare right versus left-hand performance using paired samples t-tests within each group. There were no significant differences between the two groups on either age or sex.

There was a statistically significant difference between right-hand dominant and left-hand dominant participants on the dependent variables, F (2, 599) = 8.84, p < .001, Wilks’ Lambda = .971. Mean scores indicated that right-hand dominant participants (M = 52.87, SD = 20.42) outperformed their left-hand dominant counterparts (M = 46.30, SD = 12.79) when using their right UE, though both groups performed similarly when using their left UE (right-hand dominant M = 48.55, SD = 17.81; left-hand dominant M = 49.70, SD = 14.13). These findings were present despite expected results from paired samples t-tests that revealed individuals performed best with their dominant hand.

Results revealed that handedness is necessary to consider in design and utilization of computerized neuropsychological tests. The large proportion of right-hand dominant individuals may have affected our results; however, our sample is representative of handedness distribution in the general population. Although our paired samples t-tests support validity of RC21X, continued investigation of computerized performance measurement tools is necessary. Future research must explore the possibility of an ordering effect (i.e., right-handed participants starting with their dominant UE, but left-handed participants starting with their nondominant UE) or due to construction of everyday items (e.g., computer keyboards) primarily for right-hand dominant people.

There are numerous adverse health outcomes associated with dementia caregiving, including increased stress and depression. Caregivers often face time-related, socioeconomic, geographic, and pandemic-related barriers to treatment. Thus, implementing mobile health (mHealth) interventions is one way of increasing caregivers’ access to supportive care. The objective of the current study was to collect data from a 3-month feasibility trial of a multicomponent mHealth intervention for dementia caregivers.

40 community-dwelling dementia caregivers were randomized to receive the CARE-Well (Caregiver Assessment, Resources, and Education) App or internet links connected to caregiver education, support, and resources. Caregivers were encouraged to use the App or links at least 4 times per week for 3 months. The App consisted of self-assessments, caregiver and stress reduction education, behavior problem management, calendar reminders, and online social support. Caregivers completed measures of burden, depression, and desire to institutionalize at baseline and post-intervention. Feasibility data included App usage, retention and adherence rates, and treatment satisfaction. Data were analyzed via descriptive statistics.

Caregivers were mostly white (95%), female (68%), in their mid-60s, (M= 66.38, SD= 10.64), and well-educated (M= 15.52 years, SD= 2.26). Caregivers were mainly spouses (68%) or adult children (30%). Care recipients were diagnosed with mild (60%) or moderate (40%) dementia, with 80% diagnosed as having Alzheimer’s disease. Overall, the study had an 85% retention rate (80% for App group; 90% for links group). 58% of caregivers in the App group were considered high users, using the App >120 minutes over the course of 3 months (M= 362.42, SD= 432.68), and an average of 16.44 days (SD= 15.51). 15% of the sample was non-adherent due to time constraints, disinterest, and/or technology issues. Most participants (75%) using the App were mostly or very satisfied, about 87% would be likely or very likely to seek similar programs in the future, and 93% found the App mostly or very understandable. Groups did not significantly differ on clinical outcomes, although the study was not powered for an efficacy analysis. Within groups analysis revealed significant increases in depressive symptoms at post-treatment for caregivers in both groups.

This study demonstrated initial feasibility of the CARE-Well App for dementia caregivers. App use was lower than expected, however, high satisfaction, ease of use, and willingness to use similar programs in the future were endorsed. Some caregivers did not complete the intervention due to caregiving responsibilities, general disinterest, and/or technology issues. Although the study was not designed to assess clinical outcomes, we found that both groups reported higher depressive symptoms at post-treatment. This finding was unexpected and might reflect pandemic-related stress, which has been shown to particularly impact dementia caregivers. Future studies should address the efficacy of multicomponent mHealth interventions for dementia caregivers and the effects of increased dose on clinical outcomes. mHealth interventions should be refined to cater to varying levels of technology literacy among caregivers, and further research should aim to better integrate interventions into caregivers’ routines to enhance treatment engagement.

Telecommunication-assisted neuropsychological assessment (teleNP) has become more widespread, particularly in response to the COVID-19 pandemic. However, comparatively few studies have evaluated in-home teleNP testing and none, to our knowledge, have evaluated the National Alzheimer’s Coordinating Center’s (NACC) Uniform Data Set version 3 tele-adapted test battery (UDS v3.0 t-cog). The current study compares in-home teleNP administration of the UDS v3.0, acquired while in-person activities were suspended due to COVID-19, with a prior in-person UDS v3.0 evaluation.

210 participants from the Michigan Alzheimer’s Disease Research Center’s longitudinal study of memory and aging completed both an in-person UDS v3.0 and a subsequent teleNP UDS v3.0 evaluation. The teleNP UDS v3.0 was administered either via video conference (n = 131), telephone (n = 75), or hybrid format (n = 4) with approximately 16 months between evaluations (mean = 484.7 days; SD = 122.4 days; range = 320-986 days). The following clinical phenotypes were represented at the initial assessment period (i.e., the most recent in-person UDS v3.0 evaluation prior to the teleNP UDS v3.0): cognitively healthy (n = 138), mild cognitive impairment (MCI; n = 60), dementia (n = 11), and impaired not MCI (n = 1). Tests included both the in-person and teleNP UDS v3.0 measures, as well as the Hopkins Verbal Learning Test-Revised (HVLT-R) and Letter “C” Fluency.

We calculated intraclass correlation coefficients (ICC) with raw scores from each time point for the entire sample. Sub-analyses were conducted for each phenotype among participants with an unchanged consensus research diagnosis: cognitively healthy (n = 122), MCI (n = 47), or cognitively impaired (i.e., MCI, dementia, and impaired not MCI) (n = 66). Test-retest reliability across modalities and clinical phenotypes was, in general, moderate. The poorest agreement was associated with the Trail Making Test (TMT) - A (ICC = 0.00; r = 0.027), TMT - B (ICC = 0.26; r = 0.44), and Number Span Backward (ICC = 0.49). The HVLT-R demonstrated moderate reliability overall (ICC = 0.51-0.68) but had notably weak reliability for cognitively healthy participants (ICC = 0.12-0.36). The most favorable reliability was observed in Craft Story 21 Recall - Delayed (ICC = 0.77), Letter Fluency (C, F, and L) (ICC = 0.74), Multilingual Naming Test (MINT) (ICC = 0.75), and Benson Complex Figure – Delayed (ICC = 0.79).

Even after accounting for the inherent limitations of this study (e.g., significant lapse of time between testing intervals), our findings suggest that the UDS v3.0 teleNP battery shows only modest relationships with its in-person counterpart. Particular caution should be used when interpreting measures showing questionable reliability, though we encourage further investigation of remote vs. in-person testing under more controlled conditions.

Although remote neuropsychological assessments have become increasingly common, current research on the reliability and validity of scores obtained from remote at-home assessments are sparse. No studies have examined remote at-home administration of the National Alzheimer’s Coordinating Center (NACC) Uniform Data Set (UDS) even though this battery is being used to track over 45,000 participants over time. This study aimed to determine whether remote UDS scores can be combined with in-person data by assessing whether rates of score changes over time (i.e., reliability) differed by modality and whether remote and in-person scores converge (i.e., validity).

Data for UDS visits conducted from 09/2005 to 12/2021 from 43 Alzheimer’s Disease Research Centers were examined. We identified 311 participants (254 cognitively unimpaired, 7 impaired - not mild cognitive impairment, 25 mild cognitive impairment, 25 dementia) who completed 2 remote UDS visits 0.868 years apart (SD = 0.200 years). First, initial remote scores were correlated with most recent in-person scores. Second, we examined whether rates of change differed between remote and in-person assessments. Repeated-measure one-way ANOVA were used to compare rates calculated from the same individual from remote versus inperson assessments. We additionally identified a demographically- and visit-number-matched group of 311 participants with in-person UDS visits given that all remote visits occurred after in-person visits; one-way ANOVAs were used to compare remote rates to rates from in-person assessments from the matched in-person group. Finally, accuracy of remote scores were assessed by quantifying the difference between the actual remote scores and predicted scores based on repeated in-person assessments. These residual values were then divided by the maximum score to form error rates.

Remote UDS score on MoCA-blind, Craft story immediate and delayed recall, digits forward, digits backward, phonemic fluency (F, L, F + L), and semantic fluency (animals, vegetables, animals + vegetables) were all highly correlated (all ps < 0.001) with scores obtained from preceding in-person assessments. At the group level, within-subject comparisons between remote and in-person rates of change were not significantly different for 7/11 tests; between-subject comparisons were not significantly different for 10/11 tests. Vegetable fluency had slightly reduced rates of change with remote assessment compared to inperson assessments. Critically, remote scores were consistent with predicted scores based on the trajectory of each subject’s in-person assessments with group mean error rates ranging from 0.7% (Craft Delayed Recall) to 3.9% (Phonemic fluency - F).

Our results demonstrate adequate reliability and convergent validity for remotely administered verbally based tests from the NACC UDS battery. Importantly, our findings provide some support for combining remote and in-person scores for studies that transitioned to remote testing due to COVID-19. However, future research is needed for tests with visual stimuli that assess visual memory, visuospatial function, and aspects of executive function.

Measures of clinical significance are critical for meaningful interpretation of treatment outcome research on Alzheimer’s disease. A common method of quantifying clinical significance is to calculate a minimal clinically important difference (MCID), which represents the smallest numerical change on an outcome measure that corresponds to an added benefit in a patient’s life. Often the MCID is calculated based on an anchor response. Individuals who report a meaningful change serve as the “anchors”, and the mean level of change for this group serves as the MCID. In research on Alzheimer’s disease, there are several possible raters to provide anchors, including patients, family observers, and clinicians, who may or may not agree on whether there has been a meaningful change in outcome. The goal of this study was to examine the extent to which agreement among anchors impacts MCID estimation and whether this relationship is moderated by cognitive severity status.

Analyses were completed on a longitudinal sample of 2,247 adults, age 50-103, from the Uniform Data Set 3.0. Outcome measures included the Clinical Dementia Rating - Sum of Boxes (CDR-SB), Functional Activities Questionnaire, and Montreal Cognitive Assessment.

For all of the outcomes, the MCID estimate was significantly higher when meaningful decline was endorsed by all of the raters compared to situations in which there was disagreement among the raters. For example, on the CDR-SB, agreement significantly impacted MCID estimates (F(1, 2241)=168.80, p<0.001; partial h2 = 0.07), such that the agreement group had greater CDR-SB change score (mean=1.29, SD1.98) than the no agreement group (mean=0.37, SD=1.38; Tukey HSD: p<0.001). In addition, the MCID estimate increased with increasing levels of cognitive impairment. For instance, on the CDR-SB, MCID estimates were significantly different across the severity groups (F(2, 2241)=138.27, p<0.001; partial h2 = 0.11), such that increase in CDR-SB was highest for the mild dementia group (mean=1.84, SD=2.42), moderate in the MCI group (mean=0.71, SD=1.30), and lowest for the cognitively normal group (mean=0.07, SD=0.55; Tukey HSD; all p’s < 0.001). Finally, cognitive severity status moderated the influence of agreement among raters on MCID estimation for the CDR-SB and FAQ, such that rater agreement demonstrated less influence on the MCID as disease severity increased. For example, on the CDR-SB, post-hoc tests revealed that there was a significant difference across agreement groups in the cognitively normal (p<0.001; Cohen’s d = 0.96) and MCI groups (p<0.001; Cohen’s d = 0.49), but agreement did not impact MCID estimates for the mild dementia group (p=0.065).

MCID estimates based on one anchor may underestimate meaningful change, and researchers should consider the viewpoints of multiple raters in constructing MCIDs. Consideration of agreement appears most important in the early stages of cognitive decline, which are the focus of most modern clinical trials.

Spontaneous speech involves tight coordination of a constellation of cognitive mechanisms (including planning, lexical selection, grammatical encoding, internal & external monitoring). Recent years brought a flurry of interest in detailed analysis of spontaneous speech in search of markers of prodromal Alzheimer’s disease. This work dates back to the nun studies by Snowdon et al (1996) and reveals promise for early detection through identification of subtle but significant changes in the nature of speech output years prior to diagnosis of dementia. A major challenge for neuropsychology is to develop methods to harness the potential sensitivity of language to subtle cognitive changes when testing individuals in clinical settings. In this talk I will present two lines of research that illustrate how reading aloud can be used to engage the cognitive mechanisms of spontaneous speech production in a manner that provides an easily accessible measure of Alzheimer’s disease/risk.

In the first study, Spanish-English bilinguals with mild-to-moderate Alzheimer’s disease (n=20) and proficiency matched controls (n=29) read aloud mixed-language paragraphs with a small number of language-switched words, and we recorded the number of times they automatically translated switch words by accident (e.g., saying pero instead of but; effectively autocorrecting language switches to avoid producing switches overtly). In the second study, cognitively normal monolinguals at risk for AD based on CSF biomarkers (n=14) and controls (n=50) read aloud short paragraphs in which ten critical target words were replaced with autocorrect targets (e.g., The player who scored that final [paint] for the local team reported [him] experience). Participants were instructed to avoid autocorrecting (e.g., avoid saying point instead of paint or his instead of him), and we recorded the number of times they autocorrected by accident.

Bilinguals with AD translated switch words more often than controls, and ROC curves revealed good-to-excellent discrimination between patients and controls based solely on the number of errors produced during reading aloud (AUC or Area Under the Curve values ranged from .71-.92). In the second study, cognitively normal monolinguals with high CSF Tau/Aß42 (i.e., an AD-like biomarker profile) produced more autocorrect errors (e.g., saying point instead of paint)than those below the biomarker threshold, and autocorrection errors showed potential for discriminating individuals with higher AD risk from controls (AUC=.76; 95%CI .62-.90).

Difficulty stopping automatic translation of language switch words and autocorrection during reading aloud reveals promise as a diagnostic tool. Reading aloud elicits rapid production of hundreds of words while maintaining tight experimental control over the content of speech and harnessing the power and complexity of language to enable detection of very subtle cognitive changes through simple analysis of critical targets. I will discuss the theoretical implications of this work for understanding how bilinguals choose a single language for production, the nature of cognitive impairments in early AD and areas of need for further research to maximize the potential utility of reading aloud for detection of cognitive impairment.

Alzheimer’s disease (AD) clinical trials lack diverse representation, limiting their generalizability. In addition, the clinical meaningfulness of observed changes during treatment may vary as a function of participant characteristics. Defining meaningful change in AD within diverse ethnoracial groups is therefore greatly needed. Meaningful change in AD trials can be assessed by three different anchors: participants, informants, or clinicians. Previous research has suggested that estimations of the minimal clinically important difference (MCID) vary by disease severity, choice of anchor, and anchor agreement. These relationships have been studied primarily within non-Hispanic white (NHW) samples. This project evaluates anchor-based MCID within and across the three most prevalent ethnoracial groups in the United States, non-Hispanic White (NHW), Hispanic/Latino (H/L), and Black/African-American (B/AA).

Data from the National Alzheimer’s Coordinating Center Uniform Dataset (NACC UDS) were used to investigate MCID within older adults (ages 50+) diagnosed as cognitively normal or cognitively impaired due to suspected AD. Data were taken from all versions of the UDS and consisted of all available participants with two consecutive annual visits. The identified sample (N=22,043) is approximately 83.6% NHW, 4.7% H/L, and 11.7% B/AA. Participant, informant, and clinician anchor variables were utilized to compare proportions of anchor agreement within and across ethnoracial groups. MCID on the Mini-Mental State Exam (MMSE) was estimated within each ethnoracial group and compared across the independent variables of anchor agreement and disease severity (cognitively normal (CN), mild cognitive impairment (MCI), and dementia) in 2x3 ANOVAs.

Participant age (M = 71.56, SD = 9.03) did not significantly differ across ethnoracial groups; years of education significantly differed across groups, p < .001, with NHW (M=15.83 SD=3.05), H/L (M=12.49, SD=5.01), and B/AA (M=14.42, SD=3.22). Across all three anchors (participant, informant, clinician), unanimous agreement about the presence or absence of a decline in functioning was present in about 75.1% of the full sample. To further explore agreement differences across groups, anchor agreement was classified into a 3-level variable: 1) agreement that the participant remained stable over time, 2) agreement that the participant declined, and 3) disagreement. The proportion of each level within each ethnoracial group was significantly different, (x2(4, n = 22,043) = 179.16, p < .001, phi = .09, NHW (34.5% agreement-stable, 41.4% agreement-declined, 24.1% disagreement), H/L (30.5%, 42.6%, 26.9%, respectively), and B/AA (42.2%, 28.1%, 29.7%, respectively). MCID estimates on the MMSE followed similar trends within each ethnoracial group. There was a significant main effect of disease severity, such that MCID estimates increased in magnitude with increasing disease severity. There were no significant main effects of anchor agreement for any ethnoracial group. Within the NHW sample only, an interaction effect between diagnostic severity and anchor agreement was significant (p = .007).

Across ethnoracial groups, MCID estimates on the MMSE are reliably influenced by the severity of disease. However, the benefit of anchor-based MCID estimates may vary by ethnoracial group with respect to both anchor choice and use of anchor agreement. The origins of anchor disagreement and perceived stability in functioning warrant further exploration.

Inequity in Alzheimer’s disease and related dementias (ADRD) clinical research is severely hindering our progress towards a cure for all, while inflating national costs. ADRD alone is currently costing United States 321 billion dollars in 2022, projected to increase to 1 trillion by 2050. Alzheimer’s disease disproportionately impacts Black, Hispanic, Asian or Native Americans. Yet, ADRD clinical research to date has not included equitable number of participants from communities of color to be representative of the U.S. population. Hispanic/Latinos currently represent 1% of ADRD clinical trials’ samples despite representing 18% of the US population.

In our previous outreach and recruitment study with the Human Connectome Project - Aging, we attained a 11.35% recruitment success rate of Hispanics/Latinos living in Los Angeles County Districts. We implemented a comprehensive Spanish-English bilingual, multi-method, multi-setting community-academic engagement, outreach, and recruitment protocol with a focus on brain health literary and ADRD biomarker research literacy.

Whereas community educational engagement and outreach was the most efficient and highest interest turn-out recruitment strategy, 61% of engaged and interested Hispanic/Latinos (50 years old and older) were automatically excluded during the telephone screening due to English-language proficiency/fluency. Highest enrollment success rate was from UCLA mailing list, clinical registries, and referrals. Hispanics/Latinos successfully recruited were bilingual or multilingual, 83% identified white as their racial background, 85% attained higher education, and 70% resided in middle-to-high income levels areas.

Our results captured institutional barriers leading to a recruitment bias towards affluent Hispanics/Latinos with access to healthcare systems. Our applied science of recruitment and retention requires significant improvements in study design and methodology, tailored and targeted to communities’ socio-ecological context. It also requires the extrapolation of generalizable theoretically informed mechanisms of successful engagement, recruitment, and retention strategies for replication and/or modification in other settings/locations, and countries.

With participant recruitment being a top barrier to AD research progress, the rate of screen failure in Alzheimer’s disease (AD) clinical trials is unsustainable. Although steps have been undertaken to consider solutions to the continued recruitment shortage, there is unfortunately minimal emphasis on reducing screen failure rates based on study inclusion criteria. Here we present information attempting to understand the cognitive, emotional, and functional features of individuals who failed screening measures for AD trials.

The current study is a retrospective, cross-sectional analysis. Thirty-eight participants (aged 50-83) having (1) previously received a clinical diagnostic workup at a transdisciplinary cognitive specialty clinic and (2) previously screened for a specific industry-sponsored clinical trial of MCI/early AD (EMERGE) met inclusion criteria. Previously collected clinical data were analyzed to identify predictors of AD trial screen pass/fail status.

Of the 38 participants in the current study, 14 screen passed into this AD clinical trial, and 24 screen failed. Higher screen failure rates were significantly related to gender, with 83% of female participants screen failing this AD trial versus 45% of male participants. There was no difference in age or education between screen pass/fail groups, nor were differences present for performance on visual or verbal memory tasks, or the MOCA. Conversely, those participants screen failing this AD clinical trial performed significantly worse on nonmemory cognitive domains pertaining to general fund of knowledge, working memory, and executive functioning. Additionally, the screen fail group reported greater levels of anxiety, but not depression nor endorsements on a measure of functional status.

Worse performance on non-memory neuropsychological domains was related to screen failure status for the EMERGE AD clinical trial. This finding may be explained by the traditional recruitment pathway from clinic to trials, which beyond the diagnosis of interest is up to the opinion of the physician to determine “fit” for a trial. Higher screen failure rates may result from physicians erroneously viewing more globally-impaired patients as being more appropriate for an AD clinical trial, resulting in greater tendencies towards recruiting patients who are too severe to meet inclusion criteria for a trial. Recruiting patients into clinical trials earlier in their disease course - when disease severity is less - may result in reduced screen failure rates in AD trials. That we could not detect a relationship between memory-related tasks and screen fail/pass status may be explained by either (1) the measures used in the EMERGE trial were not as sensitive to subtle changes in memory, or (2) that memory dysfunction is necessary for a diagnosis of AD but not sufficient to distinguish who will be successfully screened into an AD clinical trial. Overall, these findings have the potential to advance the field by reducing screen failure rates in AD clinical trials by using information already available to clinical trial teams, which will enhance trial-recruitment infrastructure and encourage greater engagement of older adults in AD research.