Abstract
Affordable, noninvasive methods of predicting functional decline are needed for individuals at risk for Alzheimer’s disease. This study tested whether a timed upper-extremity motor task predicted functional decline over one year in 79 adults diagnosed with amnestic mild cognitive impairment. Participants completed subjective and objective measures of daily functioning at baseline and one year later. Motor task performance and delayed memory were also evaluated at baseline. Motor task performance was a significant predictor of one-year follow-up daily functioning, improving model fits by 18– 35%. Thus, motor behavior has potential to be an affordable enrichment strategy that is sensitive to functional decline.
INTRODUCTION
There is a strong need to develop cost-effective, non-invasive enrichment strategies for Alzheimer’s disease (AD) clinical trials. The methods for acquiring current biomarkers such as amyloid and tau are expensive (e.g., positron emission tomography) or invasive (e.g., lumbar puncture) (for review see [1]). While blood-based biomarkers are more affordable, these still require extensive laboratory resources (i.e., blood sample handling infrastructure) and are currently only in the experimental stage [2].
In light of the longstanding need to develop widely accessible AD biomarkers [3], goal-directed movement has potential as a low-cost, non-invasive enrichment strategy [4]. Specifically, voluntary movements involving both multilimb coordination and manual dexterity have been associated with mild cognitive impairment (MCI) and AD [5–7], likely due to executive dysfunction [8–11], but have not been explored as predictors of disease progression. Traditionally, these studies require demanding technology (e.g., movement sensors, motion capture technology, electromyography). Most simpler motor measures that do not require extensive technology, like gait speed or grip strength, show mixed prognostic effects of dementia at baseline (e.g., [12–15]). Consequently, we have developed a timed coordination task that does not require any hardware or software, yet can also differentiate between cognitively intact and cognitively impaired individuals [16] and is feasible for amnestic MCI cohorts [17]. We now test whether this simple motor task has prognostic value for AD progression.
While cognitive decline is often the primary outcome in clinical trials, functional decline (i.e., decline in activities of daily living [ADLs]) is also sensitive to disease progression [18–20]. Given that many ADLs require goal-directed movement, the purpose of this brief report was to test whether our timed motor task can predict self-reported functional decline over one year in adults diagnosed with amnestic MCI.
MATERIALS AND METHODS
Participants
Data from 79 participants with amnestic MCI were included in this exploratory study from a larger clinical trial sample (ClinicalTrials.gov Identifier: NCT02301546; completed). Inclusion criteria were 65 years old or older, had a collateral source available to answer questions about thinking abilities and ADLs, had access and the ability to use a computer and the internet, spoke English, and demonstrated that they had single- or multi-domain amnestic MCI. Exclusion criteria were history of major neurological (e.g., stroke, Parkinson’s disease) or psychiatric illnesses (e.g., schizophrenia, bipolar disorder) or substance abuse, current major depression (>7 on the 15-item Geriatric Depression Scale), or cognitive impairment suggestive of dementia. Relevant group characteristics are provided in Table 1. This study was approved by the University of Utah Institutional Review Board. All participants provided informed consent as self or by proxy prior to enrollment.
Timed motor task
A full visual description of the timed motor task can be viewed on Open Science Framework (https://osf.io/phs57/wiki/Functional_reaching_task/), and its methods have been published previously [16, 21]. To summarize, participants use a standard plastic spoon to acquire two raw kidney beans at a time from a central cup (all cups 9.5 cm diameter and 5.8 cm deep) to one of three distal cups arranged at a radius of 16 cm at –40°, 0°, and 40° relative to the central cup. All cups were the same size. Participants in this study were tested using their dominant hand, and started by moving to the cup ipsilateral of the hand used. They then returned to the central cup to acquire two more beans at a time to transport to the middle cup, then the contralateral cup, and then repeated this sequence four more times for a total of 15 out-and-back movements. Task performance was measured as trial time (in seconds), i.e., how long it took to complete 15 movements, such that lower values indicate better performance. Errors such as transporting the wrong number of beans, dropping beans, or reaching in the wrong direction were recorded; however, no participants (0%) made any errors with their dominant hand. Participants first completed 10 trials with their nondominant hand for practice. Motor task performance was evaluated at baseline.
Subjective and objective measures of daily functioning
The primary measure of ADL function was the score on the self-report portion of the 18-item Alzheimer’s Disease Cooperative Study-Activities of Daily Living scale adapted for MCI (ADCS-ADL-MCI) with a possible score range of 0–57, with higher scores indicating better functional abilities [22]. This questionnaire assesses participants’ competence (ranging from full independence to full dependence) in six basic ADLs (e.g., eating) and 16 instrumental ADLs (e.g., using a telephone) over the previous four weeks. Secondary ADL measures included the informant portion of the ADCS-ADL-MCI (same questions but answered by a collateral) and three subtests of the Independent Living Scale (i.e., ILS Health and Safety, Managing Money, and Managing Home and Transportation). The Health and Safety subtest asks participants about health and health emergencies (e.g., explain how to safely cross a busy street or practice correct personal hygiene). The Managing Money subtest asks practical questions about money, including math skills (e.g., counting out exact change, performing simple arithmetic by hand, writing checks). The Managing Home and Transportation subtest asks questions regarding home issues and traveling into the community (e.g., how to take public transportation, looking someone up in a telephone book and calling that number). Using normative data from the standardization sample of cognitively-intact individuals, raw scores were converted to T-scores, with higher scores indicating better functional abilities. No demographic corrections (e.g., age, education, sex) were provided in the normative data. All ADL measures were administered at baseline and again one year later (mean±SD = 1.32±0.13 years). Participants underwent extensive neuropsychological assessment at baseline; however, only the Delayed Memory Index from the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) was examined here given that it 1) was the only RBANS index with a mean score <1 SD below normal; 2) captures the hallmark symptom of the amnestic subtype of MCI and was redundant with other delayed memory tests in the neuropsychological assessment (i.e., Hopkins Verbal Learning Test and Brief Visuospatial Memory Test); and 3) was used as a diagnostic criterion for participants enrolled in the corresponding clinical trial. The RBANS Delayed Memory Index includes recall of a learned list of words (and recognition of), a story, and a figure.
Statistical analysis
Multivariate linear regression was conducted to predict one-year follow-up scores on the ADCS-ADL-MCI Survey, as well as the Health and Safety, Managing Money, and Managing Home and Transportation subtests of the ILS. This regression included participants’ baseline dominant hand motor task performance (i.e., trial time) as a predictor while controlling for baseline ADL scores, age, gender, years of education, and RBANS Delayed Memory Index score. Assumptions for regression were inspected visually using Q-Q plots and all analysis were performed in R (v3.5.1). Statistical models with and without motor task performance as a dependent variable were compared by analysis of variance to determine if the contribution of motor task performance to prediction accuracy was statistically significant.
RESULTS
Mean ADCS-ADL-MCI scores at baseline and one-year follow-up are provided in Table 1, along with those for the ILS subtests. Values for the primary outcome (ADCS-ADL-MCI, self-report) were consistent with previously reported scores in MCI [23, 24]. Wilcoxon signed-ranked test indicated significant decline overall in the primary variable of interest, the ADCS-ADL-MCI (self-report) (p < 0.05), as well as the informant portion (p < 0.01), and Managing Home and Transportation ILS subtest (p < 0.005). Correlation matrices of all ADL measures at baseline (Table 2) and at one-year follow-up (Table 3) indicated minimal change over time in how the measures themselves do (or do not) relate to each other.
Group characteristics (n = 79)
†Standard score (M = 100, SD = 15). ‡T-score (M = 50, SD = 10). Mean (SD) unless otherwise noted. Paired tests were performed with Wilcoxon signed-ranked test: $p < 0.05, ¶p < 0.01; #p < 0.005.
Correlation matrix (Pearson correlation coefficient and 95% CI) of ADL measures at baseline
Correlation matrix (Pearson correlation coefficient and 95% CI) of ADL measures at one-year follow-up
Regression analyses revealed that motor task performance at baseline was a significant predictor of self-reported one-year ADCS-ADL-MCI score (β= –0.12; 95% CI = [–0.24, –0.005]; p < 0.05), even when controlling for baseline ADCS-ADL-MCI score (β= 0.65; 95% CI = [0.36, 0.93]; p < 0.001), age (p > 0.05), years of education (p > 0.05), gender (p > 0.05), and RBANS Delayed Memory Index score (p > 0.05). It is noted that age, years of education, gender, and Delayed Memory Index were not significant in this model. The full model yielded an R2 = 0.26 (F = 5.57; p < 0.0001; df = 72). Comparison of regression models with and without baseline motor task performance (R2 = 0.22) through analysis of variance demonstrated that the inclusion of motor task performance significantly improved prediction (p < 0.05) of one-year follow-up by about 18%. Motor task performance was not a significant predictor of the informant portion of the ADCS-ADL-MCI (p > 0.05).
As for the Independent Living Scale subtests, baseline motor task performance was also a significant predictor of one-year follow-up score on Health and Safety (β= –0.3.; 95% CI = [–0.53, –0.06]; p < 0.05) when controlling for baseline Health and Safety score (p > 0.05), age (p > 0.05), years of education (p > 0.05), gender (p > 0.05), and RBANS Delayed Memory Index score (p > 0.05). It is again noted that age, years of education, gender, and Delayed Memory Index were not significant in this model. The full model yielded an R2 = 0.23 (F = 4.8; p < 0.001; df = 72). Comparison of regression models with and without baseline motor task performance (R2 = 0.17) through analysis of variance demonstrated that the inclusion of motor task performance significantly improved prediction (p < 0.05) of one-year follow-up by about 35%. It is noted that motor task performance was not a significant predictor of the remaining ILS subtests (both Managing Money and Managing Home and Transportation, p > 0.05).
DISCUSSION
This brief report tested whether a timed motor task can predict functional decline over one year in adults diagnosed with amnestic MCI. Results showed that motor task performance was a significant predictor of ADCS-ADL-MCI (self-reported) and Health and Safety ILS subtest scores at one-year follow-up, even after controlling for baseline performance on these functional measures, and significantly improved the amount of variance explained. Moreover, delayed memory was not a significant predictor for any of the one-year follow-up scores, which is consistent with other studies showing the inability to predict change in ADL with memory tests [25, 26]. It is noted that this simple motor task takes <5 minutes to administer, costs less than $10 to fabricate from household items, and does not show any gender differences. We acknowledge high education levels and lack of racial/ethnic diversity within the sample (see Table 1), however, which warrants future research in more diverse cohorts to better estimate the potential of motor behavior as an affordable enrichment strategy for AD clinical trials.
It is noted that motor task performance predicted one-year functional change for only one of the three ILS subtests (i.e., Health and Safety). Previous work has demonstrated that the Health and Safety subtest is related to cognitive function, including executive function, but unrelated to self- or informant-reported instrumental ADLs (IADLs) [27]; thus, it is plausible that more of the questions within this subtest (compared to Managing Money or Transportation) map onto executive function and basic ADLs rather than IADLs, as suggested by validity studies [28]. Since declines in basic ADLs tend to not occur until later dementia stages [29, 30], our motor task may actually be identifying individuals who are progressing faster, irrespective of the severity of their memory deficits (thereby illustrating its potential value as an enrichment strategy). Furthermore, the fact that motor task performance predicted one-year functional change for only the self-reported portion of the ADCS-ADL-MCI (not the informant portion) is consistent with previous studies showing that individuals with MCI self-endorse functional decline earlier than their informants do, possibly due to a self-awareness of change in ADLs before observers detect any signs of decline [31].
Future research is needed to determine the mechanisms underlying the predictive nature of this motor task. Because this task involves spatial navigation and spatial working memory for remembering and reaching to target locations, it likely recruits hippocampal and parietal regions that are particularly susceptible to early stages of dementia [32, 33]. Furthermore, the complex and dexterous nature of this motor task likely involves executive function (more so than grip strength) for planning and executing individual task components in a specific order [8–11, 34]. Taken together, it is possible that this motor task recruits relevant structures and cognitive processes that have been shown to be prognostic of functional decline [25, 35–37]. Thus, an important next step is to validate this and other motor tasks against more established biomarkers (e.g., hippocampal volume, amyloid and tau deposition) to better establish its clinical utility. So, even though there are important avenues to expand research on this motor task, it has promise as a cost-effective and non-invasive screening tool that could be utilized early and broadly in a multi-tiered diagnostic/treatment process for AD (see [38]) as well as for enriching samples in AD clinical trials.
Footnotes
ACKNOWLEDGMENTS
The ADCS-ADL-MCI survey was used with permission from the NIA Alzheimer’s Disease Cooperative Study (NIA Grant AG10483). This work was supported by the National Institutes of Health [grant numbers R01AG045163 and K01AG047926]. This sponsor had no involvement in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.
