Digitised hand movement and plasma NfL are complementary biomarkers of the dementia continuum

Digitised hand movement and plasma NfL are complementary biomarkers of the dementia continuum | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Digitised hand movement and plasma NfL are complementary biomarkers of the dementia continuum Kaylee D Rudd, Michele L Callisaya, Katherine Lawler, Renjie Li, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7209767/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract BACKGROUND Emerging research suggests hand motor biomarkers help identify dementia, but it is unclear which test is best, or whether combining with a blood-based biomarker may further improve classification. We evaluated combinations of hand motor measures and plasma neurofilament light (NfL) to distinguish dementia, Mild Cognitive Impairment (MCI) and Subjective Cognitive Impairment (SCI) from cognitively-healthy controls (HC). METHODS Three-hundred and seventeen participants (71 dementia, 105 MCI, 59 SCI, 82 HC) completed key-tapping, finger-to-thumb-tapping, grip strength, and NfL analysis in a cognitive clinic. Age-, sex-, and education-adjusted Receiver-Operating-Characteristic curves measured classification accuracy. RESULTS Lower-frequency and higher-variability of key-tapping associated with weaker grip strength and higher NfL levels. Models combining key-tapping and NfL best classified dementia (97%), MCI (97%) and SCI (81%) from HC, outperforming individual measures. CONCLUSION Integration of a brief hand motor test and NfL could aid with distinguishing cognitive diagnosis groups. Findings support multimodal approaches for early dementia detection. Cognitive Neuroscience Geriatrics & Gerontology Upper Limb Motor Function MCI Subjective Cognitive Impairment Biomarkers ISLAND Early Diagnosis Cognitive Impairment Finger tapping Figures Figure 1 Figure 2 1 INTRODUCTION Dementia is one of the world’s leading causes of disability and mortality in older adults, and it is estimated that prevalence will triple by 2050. 1 Early detection of dementia risk is a priority as this facilitates early intervention. 2 The “dementia continuum” begins with the preclinical stage, when dementia-related changes start in the brain, but there are no cognitive symptoms, followed by Subjective Cognitive Impairment (SCI) when the person reports cognitive impairment, but standard clinical assessments remain normal. 3 – 4 The next stages along the continuum are Mild Cognitive Impairment (MCI) followed by early-stage dementia. 5 New blood-based biomarkers can aid early detection of dementia risk, which are considerably less invasive and costly than specialist brain imaging, such as Positron Emission Tomography scans (PET) scans, or cerebrospinal fluid (CSF) analysis, but are still limited by relative expense and lack of clinical availability. 4 , 6 – 8 An accumulating body of evidence shows that upper limb motor function changes in the earliest stages of the continuum and that these subtle alterations can be detected and analysed using easily accessible equipment such as webcams and computer keyboards. 9 – 15 For example, our group’s prior work found computer key-tapping frequency and speed generally decline across the continuum while variability of rhythm increases. 11 , 16 Weaker grip strength is associated with cognitive impairment, although it may be preserved in the early stages of the continuum. 13 , 15 , 17 However, no previous study has evaluated how different upper limb motor functions are associated with all symptomatic stages of the dementia continuum or with blood-based biomarkers of neurodegeneration. One such blood-based biomarker of particular interest is neurofilament light (NfL), a measure of neuroaxonal injury; NfL levels have been found to increase in CSF and blood across a range of neurological disorders including different types of dementia, which makes it ideal as a non-specific marker of dementia risk. 7 – 8 Furthermore, there is strong rationale for combining functional and biological measures, as so-called ‘multimodal’ approaches (combining different ‘modes’ of data such as brain imaging with cognitive function) improve classification of dementia and MCI from cognitively healthy controls (HC). 18 – 19 Considering the upper limb motor function changes across the dementia continuum, including in the preclinical and SCI, using combined models of upper limb motor measures and NfL may well improve discrimination of the cognitive groups. This multimodal approach holds promise to better understand the motor signatures across the continuum and hence aid with developing more accessible and affordable tests to detect dementia early. We aimed to: 1. examine the associations between finger-tapping (frequency and variability), key-tapping (frequency and variability), grip strength and NfL, adjusting for age, sex and education; 2. determine the classification accuracy of variables of finger-tapping, key-tapping, grip strength and NfL for cognitive diagnosis of dementia, MCI and SCI compared to the Null Model comprising age, sex and education, and 3. examine the best combination of hand motor tests and NfL to optimize classification accuracy for cognitive diagnosis groups compared to the Null Model. We hypothesized that (i) impaired finger- and key-tapping performance (e.g., lower frequency and greater variability) would be associated with lower grip strength and increased levels of NfL, and that (ii) combining hand motor and NfL would improve classification accuracy of cognitive diagnoses compared to models with an individual measure. 2 METHODS 2.1 Ethics and consent This study is a part of the ISLAND Cognitive Clinic (hereon referred to as the Clinic) study and approved by the Human Research Ethics Committee of the University of Tasmania (ref. H0018639). All participants’ capacity to consent to research was assessed by a clinician before giving written informed consent to participate in this study. 20 Healthy controls gave written informed consent after receiving study information and were given the opportunity to have their questions answered. 2.2 Participants and recruitment We recruited consecutive participants from adults attending the Clinic at the University of Tasmania, Australia. The Clinic protocol has been published in detail elsewhere. 20 In brief, participants received an interdisciplinary consensus diagnosis according to international diagnostic guidelines after a structured medical history, neurological examination, global cognitive screening test (Addenbrookes Cognitive Examinaton-3; ACE-III), a one-hour neuropsychological battery, a range of blood tests, and magnetic resonance imaging (MRI) brain scan with morphometry. 21 Participants with a diagnosis of dementia (specifically AD, vascular dementia (VaD), or mixed dementia (AD and VaD)), MCI (grouped as amnestic (aMCI) and non-amnestic (nMCI)) and SCI were included in this study. HC with no subjective or objective cognitive impairment were recruited from members of the ISLAND Project, of which the detailed protocol has previously been published. 22 In short, this project is a 10-year population-based health initiative to reduce Tasmanians’ risk of dementia through education and personalized feedback on modifiable risk factors. The participants in the ISLAND Project generally have a high educational attainment, with 21.5% completing a bachelor’s degree, compared to 11.6% Tasmanian average for adults aged ≥ 50 years. 22 Recognizing that this high cognitive reserve may buffer any functional decline with dementia-related brain pathology, 23 we purposefully set the ACE-III eligibility cutoff for HC to ACE-III ≥ 95/100. Scores below 95 have previously been shown to have a high sensitivity (83%) and specificity (73%) for MCI. 24 2.3 Data collection We collected data on age (years), sex (male/female/prefer not to say), self-reported hand dominance (hand used for writing (right/left)), full-time equivalent years of education, total score of a mood questionnaire - Geriatric Depression Scale (GDS) and the Hospital Anxiety and Depression Scale (HADS) for the Clinic and HC groups respectively, height/weight (measured in-person) and presence of osteoarthritis and pain in hands (recorded as yes/no). Additionally, we recorded the total and sub-scores of ACE-III, representing the cognitive domains of memory, attention, language, verbal fluency and visuo-spatial function. We also recorded the time (in seconds) taken to complete the Trail Making Test parts A and B, representing domains of motor sequencing and executive function respectively. 25 In terms of the motor tasks, all participants completed the grip strength measures first, then key-tapping and lastly finger-tapping test, supervised by the first author who was blinded to the cognitive diagnosis. NfL levels were calculated and used in the data analysis for a subset of participants who had provided blood samples. 2.3.1 Finger-tapping assessment Participants completed a finger-tapping test seated in a chair in a well-lit Clinic room. The task was recorded at 60 frames/second using a digital camera (Canon G9X Mark II; 1920 x 1080- pixel resolution), mounted on a tripod placed at 1.2m from the assessment chair. The task was demonstrated by a researcher, and participants had a chance to practice and seek clarification before recording. Participants were instructed to keep their hands at shoulder level and simultaneously tap both index fingers against their thumbs “as big and as fast” as they could for 15s. To alleviate the effect of a delayed start and/or fatigue, motor variables from the end of the first second to the end of the 11th second of the task (10s in total) were extracted for analysis. To extract finger-tapping features from recorded videos, we used a Computer Vision system called Rapid-Motion-Track that has been validated against wearable sensors. 26 – 27 Two finger-tapping variables were calculated for each hand and analysed: frequency (number of finger-taps in 30s) and variability (coefficient of variance - CV - of finger-tapping cycle time). These variables have been used in previous studies of hand motor-cognitive associations. 10 , 28 – 34 2.3.2 Computer key-tapping assessment The computer keyboard tapping test was completed using an online alternate tapping task called the BRadykinesia Akinesia INcoordination (BRAIN) test. In this test, participants tapped two computer keys 15cm apart “as quickly and accurately as possible” for 30s using the index finger of the left hand and then repeated using the right hand as per BRAIN test protocol. 11 , 29 Two key-tapping variables were selected: frequency (number of key-taps in 30s) and variability (CV of the time interval between key-taps). These variables are also conceptually equivalent to selected finger-tapping variables as finger-tapping and key-taping are not comparable like-for-like. 2.3.3 Grip strength We used a JAMAR Plus Digital Hand Grip dynamometer that has been shown to have good to excellent test-retest reliability in older adults. 35 Three trials of grip strength were measured with participants’ elbows at a 90° angle following the Southampton protocol. 36 – 37 The maximum score from each hand was recorded in kilograms. 38 2.3.4 Blood-based biomarker The detailed method of blood sample collection, processing and storage prior to analysis has been published previously. 39 In brief, non-fasting venous blood samples were processed within two hours of collection. To prepare serum, blood was clotted at room temperature for 30 minutes and then centrifuged (2,000 g) for 10 minutes at 4°C. 39 Aliquoted serum samples were stored in polypropylene, screw-top cryostorage tubes at − 80°C. Plasma NfL was measured using the Neurology 2-Plex B assay (Simoa®) from Quanterix according to the manufacturer’s protocols. Samples were diluted x4 by adding 25µL of serum to 75µL of sample diluent. 39 Results were run against high- and low-quality control samples provided by Quanterix and were excluded from analysis if the intra-assay CV between duplicates was > 20%. 2.4 Statistical analysis The skewness of the distribution of each variable was evaluated using the P-P plot. Finger-tapping frequency per second was multiplied by 30 to represent the number of finger-taps in 30s. Mood scores were derived from GDS (out of 15) for the Clinic groups and the HADS (out of 21) HC and so, algebraic rearrangement was used to transform the proportion of HADS scores before analysing the differences between groups. Associations between variables of finger-tapping, key-tapping, grip strength and NfL were analysed using partial correlation adjusted for age, sex and education. The area under the Receiver Operating Characteristic curves (AUC) was calculated to estimate classification accuracy of cognitive diagnoses. For all AUC calculations, we controlled for independent variables of age, sex and education that are shown to affect motor and cognitive functions and have been adjusted for in other studies. 15 , 40 – 41 In Model One we examined classification accuracy of the Null Model (comprising age, sex, and education) plus each hand motor variable and NfL separately. We rejected the null hypothesis, that hand motor/NfL measures do not improve classification over and above covariates, if there was a significant difference between AUC of Models Null and One. In Model Two, we started with the motor variables from Model One with the highest AUC and then added each hand motor variable one at a time in the order of strongest to weakest AUC and p-value to assess whether they improved the AUC. If they did not increase the AUC, the variable was taken out and the next variable was added. This process was repeated until the best hand motor model for each diagnostic group was identified. Finally, and in the subset of participants with both hand motor and NfL measures, we added NfL to the model to determine if it provided any improvement in the AUC over and above the hand motor variables. For all statistical analysis, and considering the exploratory design of this study, the significance was set at P < 0.05 as utilized by previous studies with similar design. 41 – 43 Statistical analysis was completed using STATA 18.0. Bar charts were generated using SPSS version 28.8.1.0. 3 RESULTS We recruited 317 participants: 71 with dementia, 105 with MCI, 59 with SCI and 82 HC. Table 1 shows the characteristics and the mean (Standard Deviation) of upper limb motor and NfL measures for diagnostic groups. The dementia group comprised 42 AD, 8 VaD and 21 mixed AD/VaD dementia, and the MCI group comprised 69 participants with aMCI and 36 with nMCI. Supplementary Table 1 shows the characteristics of dementia and MCI subtypes. Overall, for finger-tapping, the HC group had the slowest frequency, followed by all-cause dementia, MCI (all subtypes) and the SCI group that had the highest finger-tapping frequency. There were no differences in finger-tapping variability between groups. Supplementary Table 2 presents the between-group associations for demographic and study variables. Across the continuum, HC had the highest key-tapping frequency, followed by SCI and MCI (all subtypes) and dementia (all types). The dementia group had the greatest key-tapping variability, followed by MCI, compared to SCI or HC. Figure 1 visualizes the average values of upper limb motor (dominant hand) and NfL across the continuum. Table 1 Characteristics of participants in each cognitive group. Dementia MCI SCI HC n: 71 n: 105 n: 59 n: 82 Age (years) mean (SD) 75.63 (7.79) 70.00 (9.56) 65.12 (8.73) 65.09 (6.98) Sex (Female) n (%) 39 (55) 64 (61) 40 (68) 67 (82) Education (years) mean (SD) 11.89 (2.88) 13.03 (3.29) 14.85 (3.16) 15.12 (2.53) Height (m) mean (SD) 1.65 (.10) 1.66 (.10) 1.66 (.09) 1.64 (.08) Weight (kg) mean (SD) 71.80 (15.29) 77.06 (19.44) 77.96 (17.73) 73.87 (15.29) Right-handed n (%) 67 (94) 93 (89) 55 (93) 78 (95) Hand osteoarthritis n (%) 43 (61) 52 (50) 18 (31) 32 (39) Pain in hands n (%) 23 (32) 37 (35) 12 (21) 24 (29) ACE-III Scores n: 66 n: 92 n: 49 n: 82 Total (/100) mean (SD) 77.96 (10.95) 88.92 (7.57) 93.63 (5.22) 97.67 (1.61) Attention (/18) mean (SD) 14.70 (3.03) 16.46 (1.78) 17.14 (1.63) 17.72 (.59) Memory (/26) mean (SD) 16.96 (4.60) 21.29 (4.25) 23.12 (2.37) 25.66 (.63) Fluency (/14) mean (SD) 8.49 (3.16) 11.05 (2.12) 12.35 (1.97) 12.92 (.98) Language (/26) mean (SD) 23.97 (2.14) 25.19 (1.34) 25.55 (.91) 25.68 (.65) Visuo-spatial (/13) mean (SD) 13.85 (2.26) 14.89 (1.42) 15.71 (1.81) 15.70 (.68) TMT - A n: 54 n: 100 n: 56 n: 82 Time (s) mean (SD) 46.70 36.82 (11.48) 31.13 (10.29) 29.93 (8.13) TMT - B n: 48 n: 94 n: 56 n: 82 Time (s) mean (SD) 158.12 (83.75) 112.77 73.39 (31.16) 58.25 (15.49) Mood Score n: 71 n: 105 n: 56 n: 82 (/15) mean (SD) 4.65 4.88 (3.24) 4.59 (3.12) 4.71 (3.33) Grip strength (kg) n: 69 n: 104 n: 58 n: 82 Non-dominant hand mean (SD) 26.00 (9.20) 27.81 (11.00) 29.42 (8.95) 30.62 (8.11) Dominant hand mean (SD) 27.64 (9.33) 29.43 (10.50) 31.96 (9.59) 32.48 (8.78) Finger-tapping n: 64 n: 100 n: 57 n: 45 Frequency (N) mean (SD) 50.68 (21.08) 53.94 (18.27) 57.91 (21.83) 42.45 (13.77) Frequency (D) mean (SD) 47.90 (18.13) 54.17 (19.41) 60.97 (21.83) 43.36 (12.86) Variability (N) mean (SD) 32.46 (26.46) 23.84 (18.93) 21.03 (13.98) 25.40 (19.59) Variability (D) mean (SD) 29.85 (28.04) 25.75 (20.80) 21.00 (15.73) 22.98 (16.31) Key-tapping n: 45 n: 51 n: 40 n: 57 Frequency (N) mean (SD) 43.02 (12.32) 50.98 (10.15) 59.75 (10.94) 66.21 (8.55) Frequency (D) mean (SD) 47.69 (14.05) 56.28 (9.72) 66.73 (11.00) 71.70 (9.18) Variability (N) mean (SD) 142.01 (129.59) 86.68 (35.47) 66.04 (40.61) 45.26 (19.51) Variability (D) mean (SD) 139.22 (136.38) 94.91 (63.13) 58.63 (27.11) 45.67 (29.08) NfL (pg/mL) n: 40 n: 39 n: 24 n: 39 mean (SD) 32.83 (14.04) 23.51 (12.92) 19.01 (8.74) 13.43 (4.88) Table 1 Legend : When adjusted for age, there were no differences in the reported pain or osteoarthritis between groups. There were no differences between groups in height/weight or mood scores. HADS Mean (SD) before transformation of proportion was 6.60 (4.66)/21. Abbreviations: MCI, Mild Cognitive Impairment; SCI, Subjective Cognitive Impairment; HC, cognitively healthy controls; AD, Alzheimer’s Disease; VaD, Vascular dementia; n, number; SD, standard deviation; Y, answering “Yes” to having osteoarthritis or pain in hands; N, non-dominant hand; D, dominant hand; ACE-III, Addenbrooke’s Cognitive Examination-version 3; TMT, Trail Making Test; NfL, neurofilament light; pg/mL, pictograms per millilitre. 3.1 Associations between hand motor variables and NfL Observed correlations between measures were weak to moderate; higher key-tapping frequency was associated with stronger grip measures and lower NfL levels. Greater finger-tapping variability was associated with greater key-tapping variability, weaker grip and higher NfL levels. Weaker grip measures were associated with higher NfL levels. Supplementary Table 3 presents the partial correlation (corr.) between variables adjusted for age, sex and education. In adjusted models, finger-tapping and key-tapping variability of non-dominant hand were associated (corr.: 0.254, P < 0.001). Key-tapping frequency of each hand was correlated with grip strength of the same hand (non-dominant corr.: 0.228, P : 0.002; dominant corr.: 0.245, P < 0.001). Key-tapping variability (dominant hand) was associated with grip strength of the same hand (corr.: -0.252, P < 0.001). Key-tapping frequency and variability (non-dominant hand only) were associated with NfL (frequency corr.: -0.253, P : 0.008, variability corr.: 0.426, P < 0.001). Grip strength of either hand was associated with NfL (non-dominant corr.: -0.248, P : 0.003 and dominant corr.: -0.240, P : 0.004). 3.2 Classification accuracy of motor variables and NfL Supplementary Table 4 presents the accuracy of individual upper limb motor variables and NfL for classifying cognitive diagnoses. Finger-tapping frequency of either hand improved classification of the SCI group (but not dementia or MCI) from HC (AUC: ≥ 0.79, P ≤ 0.006). Compared to the Null Model, key-tapping frequency and variability of either hand improved classification of dementia (all AUCs ≥ 0.94; P ≤ 0.035), and MCI (all AUCs ≥ 0.83; P ≤ 0.008) from HC, but not SCI. Grip strength did not aid with classification of any diagnostic group from HC. NfL improved discrimination of only the dementia group from HC. Compared to the Null Model, no measure improved classification of dementia, or MCI from SCI, or aided with discriminating dementia from MCI. 3.3 The best combination of hand motor tests and NfL to optimize classification accuracy of cognitive diagnosis groups To differentiate dementia from HC, the best combination of key-tapping variables was key-tapping frequency of non-dominant hand (AUC: 0.95). Combining NfL with the non-dominant hand key-tapping frequency slightly improved the classification (AUC: 0.97, P : 0.021) of dementia from HC. The best combination of variables to discriminate MCI from HC was key-tapping frequency of both dominant and non-dominant hands and variability of the non-dominant hand (AUC: 0.96). The addition of NfL slightly improved accuracy (AUC: 0.97). The best combination of variables to discriminate SCI from HC was all four key-tapping variables, e.g., frequency of both dominant and non-dominant hands and variability of both dominant and non-dominant hands, (AUC: 0.76). Adding NfL to the model improved classification accuracy (AUC: .81). Figure 2 and Table 2 present classification accuracy of combined hand motor and NfL measures for cognitive diagnosis groups compared to the Null Model and the model comprising covariates and hand motor measures in the subset of participants with both hand motor and NfL measures. Table 2 Classification accuracy of combined upper limb motor and NfL measures for cognitive diagnosis groups in an adjusted model. AUC 95% CI P HC v Dementia n: 59 Null 0.85 0.75; 0.95 Null + Key-tapping frequency (N) 0.95 0.88; 1.00 0.073 Null + Key-tapping frequency (N) + NfL 0.97 0.93; 1.00 0.021 HC v MCI n: 60 Null 0.73 0.60; 0.87 Null + Key-tapping frequency (N + D) + Key-tapping variability (N) 0.96 0.92; 1.00 0.001 Null + Key-tapping frequency (N + D) + Key-tapping variability (N) + NfL 0.97 0.94; 1.00 < 0.001 HC v SCI n: 46 Null 0.54 0.36; 0.72 Null + Key-tapping frequency (N + D) + Key-tapping variability (N + D) 0.76 0.61; 0.90 0.067 Null + Key-tapping frequency (N + D) + Key-tapping variability (N + D) + NfL 0.81 0.68; 0.93 0.018 Table 2 Legend : Model Two was developed using the variables with the strongest AUC and p-value from Model One. P < 0.05 indicates added variable improves prediction of diagnosis over and above the Null Model. Abbreviations: HC, cognitively healthy controls; MCI, Mild Cognitive Impairment; SCI, Subjective Cognitive Impairment; AUC, area under the Receiver Operating Characteristic curve; CI, Confidence Interval; P , p-value. N, nondominant hand; D, dominant hand; NfL, neurofilament light. 4 DISCUSSION We examined three digitised upper limb motor function tests and a blood-based biomarker of neurodegeneration in large groups of clinically diagnosed participants with subjective and objective cognitive impairment, as well as healthy controls. In models adjusted for age, sex and education we found associations between slower, less rhythmic key-tapping, weaker grip strength and higher NfL levels. We also found that individually, classification accuracy of upper limb motor function and NfL measures for diagnostic groups were different. Our main finding was that combining key-tapping with NfL measures significantly improved classification of all cognitive diagnosis groups from HC with adjusted accuracies of 97%, 97% and 81% for dementia, MCI and SCI respectively. Considering the best results only required data from a short test of motor function on a standard computer keyboard, our findings are significant in developing an accessible clinical tool. 4.1 Associations between motor variables and NfL We report a novel finding: that impaired hand function, indicated by lower key-tapping frequency, greater key-tapping variability and lower grip strength, was associated with increased NfL levels. These correlations have not been examined in groups with a cognitive diagnosis and require further investigation; Two previous studies investigated associations between grip strength and NfL in community-dwelling adults without a cognitive diagnosis but had contradictory findings; the first, a cross-sectional study of 1925 adults aged 20–75 years, reported those with higher NfL levels had lower grip strength. 44 However, another study of 507 adults aged 71–81 years examining grip strength and NfL cross-sectionally and at 2-year follow-up found no associations at either analysis. 45 Associations found between each of the hand motor tasks were weaker than we expected considering available evidence from neuroimaging studies. Looking at finger-tapping and key-tapping tasks first, sparse brain MRI studies in healthy individuals show activation of similar brain areas including the primary and supplementary motor cortices, parietal lobe, and basal ganglia during both tasks. 46 – 47 This would imply similarities in these tasks’ neural pathways, but two recent studies examined the associations between tapping tests and brain structure in people with cognitive impairment present a slightly different picture. One investigated the age/sex-adjusted associations between computer key-tapping measures and hippocampal volume on MRI scans in 26 persons with AD, 27 aMCI and 47 HC. 48 They found lower speed and more variability on key-tapping associated with smaller hippocampal volumes. On the other hand, another study investigated the age/sex-adjusted associations between finger-tapping measures and whole brain MRI scan in 71 individuals with AD and 65 MCI and reported lower tapping frequency associated with reduced grey matter volume in the primary motor cortex. 49 While both tapping tasks activate the motor cortex, the degree and distribution of activation may vary. 50 Key-tapping requires more precision and speed control when alternately tapping small keys; hence, as a more complex task may activate larger areas of the brain compared to finger-tapping. With no need for precision or following visual input, finger-tapping can be considered a simpler task that uses established/learned motor patterns in the motor cortex. 50 The complexity of these tasks, therefore, might explain the generally weak associations found between finger-tapping and key-tapping measures. Correlations between tapping tasks and grip strength were also weak although brain MRI analysis during grip strength test shows activation of brain regions similar to key-tapping and finger-tapping tests. 51 – 52 This finding might also be related to differences in tasks’ complexity and cognitive requirements; compared to tapping tasks, grip strength measure requires attention and controlled activation of muscles. 52 This test might also rely more on the palm/digits’ tactile feedback than on visual input, hence using slightly different neural pathways than used in tapping tasks. 51 Further research is necessary to understand the associations between upper limb motor function and brain structure. 4.2 Classification accuracy of hand motor variables and NfL Considering the associations found between finger-tapping frequency and cognitive groups, we expected this variable to classify cognitive diagnosis groups, especially dementia and MCI. However, we found finger-tapping frequency only accurately classified SCI from HC. Finger-tapping frequency of our SCI group was higher than all other groups, including HC. The exact reason for this finding is unclear, but it may be related to prioritization of an aspect of the task by different groups. Given the same instruction was given to all participants, the HC group appears to have prioritized amplitude and tap “big”, hence having a lower frequency while cognitive diagnosis groups put speed first and tapped faster. To our knowledge, no other study has tested finger-tapping performance in SCI, hence further research is required. SCI is a heterogeneous and less explored stage of the continuum; contradictory findings are reported by the few studies investigating upper limb motor-cognitive associations in SCI. Our previous study showed greater key-tapping variability was associated with SCI as well as MCI and dementia 11 , but another study reported no differences in key-tapping frequency or variability across 11 individuals with AD, 19 aMCI, 23 SCI and 12 HC. 53 However, they had small sample sizes and used a different key-tapping test, involving 10s tapping of a lever with the index finger that might be less demanding on neural pathways than an alternate key-tapping test. Different methods of key-tapping tests make comparison of findings complicated, underlining careful consideration of the methods used to test motor function before drawing conclusions. In our study, grip strength did not improve classification of any diagnostic group, although it has been reported in longitudinal studies as a predictor of future conversion to dementia. 43 , 54 Our finding, however, aligns with recent evidence suggesting grip strength is often preserved in the early stages of dementia and that testing upper limb functions such as key-tapping provides more information on motor-cognitive associations. 17 , 42 We found that NfL only improved classification accuracy of dementia from HC. NfL-cognition associations need further exploration as current evidence is contradictory. For example, a four-year longitudinal study reported no association between NfL and cognition 39 , and another study with 15–30 months follow-up found significant associations. 40 In our study, NfL levels were associated with cognitive diagnosis groups and yet, NfL did not improve classification of MCI, or SCI, from HC. This highlights the importance of future research with larger sample sizes. 4.3 The best combination of hand motor tests and NfL to optimize classification accuracy of cognitive diagnosis groups Combining key-tapping measures with NfL significantly improved classification accuracy of all cognitive diagnoses including SCI. Our findings are important especially considering cognitive tests cannot discriminate SCI from HC. Using a combination of measures from different sources is common practice in clinical investigation of health disorders including dementia. 55 Advances in identification of dementia-related pathology in the brain and blood has led to development of tests such as PET scans and blood-based biomarkers. These tests have made the investigation of novel multimodal models possible. In recent years, various combinations of biomarkers, including cognitive, motor, genetic, brain imaging and blood-based, have been investigated for classification of dementia and MCI and there are also studies currently under way. 56 – 59 However, to our knowledge, no study has examined a combination of hand motor function and NfL. Current evidence shows that combining different biomarkers generally improves classification accuracy of dementia and MCI. 15 , 41 , 43 For example, one study reported combining MRI and CSF biomarkers classified 93% of AD and 76% of MCI compared to 87% and 72% individually. 60 Still, models using brain imaging and CSF are limited by costs and low accessibility. Emerging evidence on digital biomarkers of dementia has provided new opportunities in developing combined models. Our study used digitised tests, accessible using everyday devices, and a plasma NfL measure that are considerably less costly and invasive. 4.4 Strengths and limitations We examined three upper limb motor tests in large samples of consecutively recruited participants from a cognitive clinic and included NfL. Diagnosis of our cognitive groups was based on a robust interdisciplinary consensus diagnosis after comprehensive gold-standard assessment, rather than a global cognitive screening test. We used a well-established protocol to measure grip strength and a simple, short and validated key-tapping test. We also tested finger-tapping; commonly used in clinics and research. 26 , 61 Additionally, we adjusted all models for age, sex and education. Limitations are also acknowledged; although participants had rests in-between tests, the finger-tapping results could have been affected by fatigue as they were completed after grip strength and key-tapping tests. Future investigators could consider alternating tests’ order to alleviate fatigue. We acknowledge that the durations of finger-tapping and key-tapping tasks were not the same in this study and that it may have affected our findings. Additionally, we did not investigate other measures of neurodegeneration such as hippocampal volume in this cross-sectional study. Future research could consider investigating these associations longitudinally, consider using the same duration for tapping tasks and include other neurodegeneration measures. Finally, future studies should consider adjusting NfL levels for renal function and other comorbidities that may impact NfL levels. 62 – 63 In conclusion, a combination of digitised upper limb motor test variables and NfL classified all symptomatic stages of the dementia continuum, indicating potential benefits of further exploring this multimodal approach in detecting those in early stages of dementia. Declarations CONFLICT OF INTEREST STATEMENT The authors have no competing interests to declare. SOURCES OF FUNDING Royal Hobart Hospital Research Foundation RHHRF (20-003) grant supported plasma NfL analysis in this study. Grant title: Blood-based biomarkers for neurodegeneration dementia, Major Project (2022), University of Tasmania. KR is supported by an Australian Government Research Training Program living allowance and University of Tasmania tuition fee scholarship. The ISLAND Clinic and ISLAND Project were supported by grants from the Medical Research Future Fund, the National Health and Medical Research Council, Tasmanian Masonic Medical Research Foundation, St Lukes Health and the Hopkins Foundation. The funding bodies have no direct role in the study’s design, data collection, analysis, interpretation, or manuscript preparation. CONSENT STATEMENT All study participants provided written informed consent. This study was approved by the Human Research Ethics Committee of the University of Tasmania. References Nichols E, Vos T. The estimation of the global prevalence of dementia from 1990-2019 and forecasted prevalence through 2050: An analysis for the Global Burden of Disease (GBD) study 2019. Alzheimer's & Dementia . 2021;17(S10):e051496. doi:https://doi.org/10.1002/alz.051496 Jack Jr CR, Knopman DS, Jagust WJ, et al. Tracking pathophysiological processes in Alzheimer's disease: an updated hypothetical model of dynamic biomarkers. The Lancet Neurology . 2013;12(2):207-216. doi:https://doi.org/10.1016/s1474-4422(12)70291-0 Garcia-Ptacek S, Eriksdotter M, Jelic V, Porta-Etessam J, Kåreholt I, Manzano Palomo S. Subjective cognitive impairment: Towards early identification of Alzheimer disease. Neurologia (English Edition) . 2016/10/01/ 2016;31(8):562-571. doi:https://doi.org/10.1016/j.nrleng.2013.02.011 Dubois B, Villain N, Frisoni GB, et al. Clinical diagnosis of Alzheimer's disease: recommendations of the International Working Group. The Lancet Neurology . 2021;20(6):484-496. doi:https://doi.org/10.1002/alz.051167 Mitchell AJ, Beaumont H, Ferguson D, Yadegarfar M, Stubbs B. Risk of dementia and mild cognitive impairment in older people with subjective memory complaints: meta-analysis. Acta Psychiatrica Scandinavica . 2014;130(6):439-451. doi:https://doi.org/10.1111/acps.12336 Jack Jr CR, Bennett DA, Blennow K, et al. NIA‐AA research framework: toward a biological definition of Alzheimer's disease. Alzheimer's & dementia . 2018;14(4):535-562. doi:https://doi.org/10.1016/j.jalz.2018.02.018 Mattsson N, Andreasson U, Zetterberg H, Blennow K, Initiative AsDN. Association of plasma neurofilament light with neurodegeneration in patients with Alzheimer disease. JAMA neurology . 2017;74(5):557-566. doi:https://doi.org/10.1001/jamaneurol.2016.6117 Gaetani L, Blennow K, Calabresi P, Di Filippo M, Parnetti L, Zetterberg H. Neurofilament light chain as a biomarker in neurological disorders. Journal of Neurology, Neurosurgery & Psychiatry . 2019;90(8):870-881. doi:https://doi.org/10.1136/jnnp-2018-320106 Wang X, St George RJ, Bindoff AD, et al. Estimating presymptomatic episodic memory impairment using simple hand movement tests: A cross-sectional study of a large sample of older adults. Alzheimer's & Dementia . 2024;20(1):173-182. doi:https://doi.org/10.1002/alz.13401 Mollica MA, Tort-Merino A, Navarra J, et al. Early detection of subtle motor dysfunction in cognitively normal subjects with amyloid-β positivity. Cortex . 2019;121:117-124. doi:https://doi.org/10.1016/j.cortex.2019.07.021 Rudd KD, Lawler K, Callisaya ML, et al. Hand Motor Dysfunction Is Associated with Both Subjective and Objective Cognitive Impairment across the Dementia Continuum. Dementia and Geriatric Cognitive Disorders . Jul 29 2024:1-11. doi:https://doi.org/10.1159/000540412 Cui M, Zhang S, Liu Y, Gang X, Wang G. Grip strength and the risk of cognitive decline and dementia: a systematic review and meta-analysis of longitudinal cohort studies. Frontiers in aging neuroscience . 2021;13:625551. doi:https://doi.org/10.3389/fnagi.2021.625551 Buchman AS, Wilson RS, Boyle PA, Bienias JL, Bennett DA. Grip strength and the risk of incident Alzheimer’s disease. Neuroepidemiology . 2007;29(1-2):66-73. doi:https://doi.org/10.1159/000109498 Li R. Using artificial intelligence (AI)-based computer vision methods to investigate associations between hand movement and Alzheimer’s Disease (AD) risk . University of Tasmania; 2024. Koppelmans V, Ruitenberg MFL, Schaefer SY, et al. Classification of Mild Cognitive Impairment and Alzheimer’s Disease Using Manual Motor Measures. Neurodegenerative Diseases . 2024:54-70. doi:https://doi.org/10.1159/000539800 Li R, Wang X, Lawler K, et al. Brief webcam test of hand movements predicts episodic memory, executive function, and working memory in a community sample of cognitively asymptomatic older adults. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring . 2024/01/01 2024;16(1):e12520. doi:https://doi.org/10.1002/dad2.12520 Hooyman A, Malek‐Ahmadi M, Fauth EB, Schaefer SY. Challenging the relationship of grip strength with cognitive status in older adults. International Journal of Geriatric Psychiatry . 2021;36(3):433-442. doi:https://doi.org/10.1002/gps.5441 Rudd KD, Callisaya ML, Lawler K, Noyce AJ, Vickers JC, Alty J. Stepping and tapping: combining motor tasks improves cognitive classification. GeroScience . 2025/05/08 2025;doi:https://doi.org/10.1007/s11357-025-01678-7 Huang G, Li R, Bai Q, Alty J. Multimodal learning of clinically accessible tests to aid diagnosis of neurodegenerative disorders: a scoping review. Health Information Science and Systems . 2023/07/22 2023;11(1):32. doi:https://doi.org/10.1007/s13755-023-00231-0 Alty J, Lawler K, Salmon K, et al. A new one-stop interdisciplinary cognitive clinic model tackles rural health inequality and halves the time to diagnosis: Benchmarked against a national dementia registry. International Journal of Geriatric Psychiatry . 2023;38(8):e5988. doi:https://doi.org/10.1002/gps.5988 Sachdev PS, Blacker D, Blazer DG, et al. Classifying neurocognitive disorders: the DSM-5 approach. Nature Reviews Neurology . 2014/11/01 2014;10(11):634-642. doi:https://doi.org/10.1038/nrneurol.2014.181 Bartlett L, Doherty K, Farrow M, et al. Island study linking aging and neurodegenerative disease (ISLAND) targeting dementia risk reduction: protocol for a prospective web-based cohort study. JMIR Research Protocols . Mar 1 2022;11(3):e34688. doi:https://doi.org/10.2196/34688 Bruno D, Schurmann Vignaga S. Addenbrooke’s cognitive examination III in the diagnosis of dementia: a critical review. Neuropsychiatric disease and treatment . 2019:441-447. doi:https://doi.org/10.2147/NDT.S151253 Pendlebury ST, Mariz J, Bull L, Mehta Z, Rothwell PM. MoCA, ACE-R, and MMSE versus the National Institute of Neurological Disorders and Stroke–Canadian Stroke Network vascular cognitive impairment harmonization standards neuropsychological battery after tia and stroke. Stroke . 2012;43(2):464-469. doi:https://doi.org/10.1161/STROKEAHA.111.633586 Bowie CR, Harvey PD. Administration and interpretation of the Trail Making Test. Nature protocols . 2006;1(5):2277-2281. doi:https://doi.org/10.1038/nprot.2006.390 Li R, Lao C, St. George R, et al. Rapid-Motion-Track: Markerless Tracking of Fast Human Motion with Deeper Learning. Intelligence-Based Medicine . 2024;10(100162)doi:https://doi.org/10.1016/j.ibmed.2024.100162 Li R, Tran S, Garg S, Lawler K, Alty J, Bai Q. Parallel multi-scale networks with deep supervision for hand keypoint detection. arXiv preprint arXiv:211210275 . 2021;doi:https://doi.org/10.48550/arXiv.2112.10275 Bologna M, Guerra A, Colella D, et al. Bradykinesia in Alzheimer's disease and its neurophysiological substrates. Clin Neurophysiol . Apr 2020;131(4):850-858. doi:https://doi.org/10.1016/j.clinph.2019.12.413 Noyce AJ, Nagy A, Acharya S, et al. Bradykinesia-akinesia incoordination test: validating an online keyboard test of upper limb function. PLoS One . 2014;9(4):e96260. doi:https://doi.org/10.1371/journal.pone.0096260 Newby R, Muhamed S, Alty J, et al. Geste antagoniste effects on motor performance in dystonia—a kinematic study. Movement Disorders Clinical Practice . 2022;9(6):759-764. doi:https://doi.org/10.1002/mdc3.13505 Suzumura S, Kanada Y, Osawa A, et al. Assessment of finger motor function that reflects the severity of cognitive function. Fujita Med J . 2021;7(4):122-129. doi:https://doi.org/10.20407/fmj.2020-013 Tomita Y, Tanaka S, Takahashi S, Takeuchi N. Detecting cognitive decline in community-dwelling older adults using simple cognitive and motor performance tests. Geriatrics & Gerontology International . 2020;20(3):212-217. doi:https://doi.org/10.1111/ggi.13863 Williams S, Zhao Z, Hafeez A, et al. The discerning eye of computer vision: Can it measure Parkinson's finger tap bradykinesia? J Neurol Sci . Sep 15 2020;416:117003. doi:https://doi.org/10.1016/j.jns.2020.117003 Amo-Salas J, Olivares-Gil A, García-Bustillo Á, García-García D, Arnaiz-González Á, Cubo E. Computer Vision for Parkinson’s Disease Evaluation: A Survey on Finger Tapping. Healthcare . 2024;12(4):439. doi:https://doi.org/10.3390/healthcare12040439 Bohannon R. Test-Retest Reliability of Measurements of Hand-Grip Strength Obtained by Dynamometry from Older Adults: A Systematic Review of Research in the PubMed Database. The Journal of frailty & aging . 2017;6(2):83-87. doi:https://doi.org/10.14283/jfa.2017.8 Roberts HC, Denison HJ, Martin HJ, et al. A review of the measurement of grip strength in clinical and epidemiological studies: towards a standardised approach. Age and ageing . 2011;40(4):423-429. doi:https://doi.org/10.1093/ageing/afr051 Beaudart C, Rolland Y, Cruz-Jentoft AJ, et al. Assessment of muscle function and physical performance in daily clinical practice: a position paper endorsed by the European Society for Clinical and Economic Aspects of Osteoporosis, Osteoarthritis and Musculoskeletal Diseases (ESCEO). Calcified tissue international . 2019;105:1-14. doi:https://doi.org/10.1007/s00223-019-00545-w McGrath R, Vincent BM, Jurivich DA, et al. Handgrip strength asymmetry and weakness together are associated with functional disability in aging Americans. The Journals of Gerontology: Series A . 2021;76(2):291-296. doi:https://doi.org/10.1093/gerona/glaa100 Collins JM, Bindoff AD, Roccati E, Alty JE, Vickers JC, King AE. Does serum neurofilament light help predict accelerated cognitive ageing in unimpaired older adults? Frontiers in Neuroscience . 2023;17:1237284. doi:https://doi.org/10.3389/fnins.2023.1237284 Mielke MM, Syrjanen JA, Blennow K, et al. Plasma and CSF neurofilament light: relation to longitudinal neuroimaging and cognitive measures. Neurology . 2019;93(3):e252-e260. doi:https://doi.org/10.1212/WNL.0000000000007767 Yamada Y, Kobayashi M, Shinkawa K, et al. Characteristics of Drawing Process Differentiate Alzheimer’s Disease and Dementia with Lewy Bodies. Journal of Alzheimer’s Disease . 2022;90(2):693-704. doi:https://doi.org/10.3233/JAD-220546 Koppelmans V, Duff K, Schaefer SY. Using a Functional Upper Limb Motor Task to Predict Cognitive and Functional Decline in Mild Cognitive Impairment and Alzheimer’s Disease. Neurorehabilitation and Neural Repair . 2024;0(0):15459683241309591. doi:https://doi.org/10.1177/15459683241309591 Orchard SG, Polekhina G, Ryan J, et al. Combination of gait speed and grip strength to predict cognitive decline and dementia. Alzheimer's & dementia (Amsterdam, Netherlands) . 2022;14(1):e12353. doi:https://doi.org/10.1002/dad2.12353 Dilixiati D, Kadier K, Qiao B, et al. Association between serum neurofilament light chain levels and grip strength among US adults: a cross-sectional study using National Health and Nutrition Examination Survey data from 2013 to 2014. BMJ open . 2025;15(4):e084766. doi:https://doi.org/10.1136/bmjopen-2024-084766 He L, de Souto Barreto P, Giudici KV, et al. Cross-Sectional and Longitudinal Associations Between Plasma Neurodegenerative Biomarkers and Physical Performance Among Community-Dwelling Older Adults. The Journals of Gerontology: Series A . 2020;76(10):1874-1881. doi:https://doi.org/10.1093/gerona/glaa284 Giancardo L, Sánchez-Ferro A, Butterworth I, Mendoza CS, Hooker JM. Psychomotor Impairment Detection via Finger Interactions with a Computer Keyboard During Natural Typing. Scientific Reports . 2015/04/16 2015;5(1):9678. doi:https://doi.org/10.1038/srep09678 Buijink AWG, Broersma M, van der Stouwe AMM, et al. Rhythmic finger tapping reveals cerebellar dysfunction in essential tremor. Parkinsonism & Related Disorders . 2015/04/01/ 2015;21(4):383-388. doi:https://doi.org/10.1016/j.parkreldis.2015.02.003 Koppelmans V, Ruitenberg MFL, Schaefer SY, et al. Delayed and More Variable Unimanual and Bimanual Finger Tapping in Alzheimer’s Disease: Associations with Biomarkers and Applications for Classification. Journal of Alzheimer’s Disease . 2023;95(3):1233-1252. doi:https://doi.org/10.3233/JAD-221297 Suzumura S, Sugioka J, Sakurai K, et al. Finger motor skills and related brain regions in patients with cognitive disorder. Journal of Alzheimer’s Disease . 2024:13872877251344223. doi:https://doi.org/10.1177/13872877251344223 Witt ST, Laird AR, Meyerand ME. Functional neuroimaging correlates of finger-tapping task variations: an ALE meta-analysis. Neuroimage . 2008;42(1):343-356. doi:https://doi.org/10.1016/j.neuroimage.2008.04.025 Ward N, Frackowiak R. Age‐related changes in the neural correlates of motor performance. Brain . 2003;126(4):873-888. doi:https://doi.org/10.1093/brain/awg071 Jiang R, Westwater ML, Noble S, et al. Associations between grip strength, brain structure, and mental health in > 40,000 participants from the UK Biobank. BMC Medicine . 2022/09/09 2022;20(1):286. doi:https://doi.org/10.1186/s12916-022-02490-2 Prigatano GP, Braga LW, McElvogue M, et al. Motor correlates of finger tapping variability in subjective memory complaints, mild cognitive impairment and probable Alzheimer's disease. Journal of Alzheimer’s Disease . 2025:13872877241312984. doi:https://doi.org/10.1177/13872877241312984 Montero-Odasso M, Pieruccini-Faria F, Ismail Z, et al. CCCDTD5 recommendations on early non cognitive markers of dementia: A Canadian consensus. Alzheimer's & Dementia: Translational Research & Clinical Interventions . 2020;6(1):e12068. doi:https://doi.org/10.1002/trc2.12068 Kumar S, Rani S, Sharma S, Min H. Multimodality Fusion Aspects of Medical Diagnosis: A Comprehensive Review. Bioengineering . 2024;11(12):1233. doi:https://doi.org/10.3390/bioengineering11121233 Jonell P, Moëll B, Håkansson K, et al. Multimodal capture of patient behaviour for improved detection of early dementia: clinical feasibility and preliminary results. Frontiers in Computer Science . 2021;3:642633. doi:https://doi.org/10.3389/fcomp.2021.642633 Alty J, Bai Q, Li R, et al. The TAS Test project: a prospective longitudinal validation of new online motor-cognitive tests to detect preclinical Alzheimer's disease and estimate 5-year risks of cognitive decline and dementia. BMC Neurol . Jul 18 2022;22(1):266. doi:https://doi.org/10.1186/s12883-022-02772-5 Park JE, Gunasekaran TI, Cho YH, et al. Diagnostic blood biomarkers in Alzheimer’s disease. Biomedicines . 2022;10(1):169. doi:https://doi.org/10.3390/biomedicines10010169 Bi X-a, Cai R, Wang Y, Liu Y. Effective diagnosis of Alzheimer’s disease via multimodal fusion analysis framework. Frontiers in genetics . 2019;10:976. doi:https://doi.org/10.3389/fgene.2019.00976 Zhang D, Wang Y, Zhou L, Yuan H, Shen D, Initiative AsDN. Multimodal classification of Alzheimer's disease and mild cognitive impairment. Neuroimage . 2011;55(3):856-867. doi:https://doi.org/10.1016/j.neuroimage.2011.01.008 Rudd KD, Lawler K, Callisaya ML, Alty J. Investigating the associations between upper limb motor function and cognitive impairment: a scoping review. GeroScience . 2023;45:3449–3473. doi:https://doi.org/10.1007/s11357-023-00844-z Akamine S, Marutani N, Kanayama D, et al. Renal function is associated with blood neurofilament light chain level in older adults. Scientific reports . 2020;10(1):20350. doi:https://doi.org/10.1038/s41598-020-76990-7 Wu J, Xiao Z, Wang M, et al. The impact of kidney function on plasma neurofilament light and phospho-tau 181 in a community-based cohort: the Shanghai Aging Study. Alzheimer's Research & Therapy . 2024;16(1):32. doi:https://doi.org/10.1186/s13195-024-01401-2 Additional Declarations The authors declare no competing interests. Supplementary Files SupplementaryTable1.docx Demographic characteristics of dementia and MCI subtypes. SupplementaryTable2.docx Associations between measures of hand motor tests and NfL and cognitive diagnosis groups. SupplementaryTable3.docx Associations between variables of finger-tapping, key-tapping, grip strength and NfL in the model adjusted for age, sex and education. SupplementaryTable4.docx Classification accuracy of individual upper limb motor variables and NfL for cognitive diagnosis of dementia, MCI and SCI in an adjusted model. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7209767","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":490605968,"identity":"922e952f-030f-4ec4-a480-c855f38691fb","order_by":0,"name":"Kaylee D Rudd","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAklEQVRIiWNgGAWjYBACCQYGxgMQZoLhA6igASEtDDAtxgYMCSRqMZMgSotk+xmDAx8qGPINjidvq+b9YWfPwN68TYKh5jBOLdI8OQYHZ5xhsNxw5lnZbZ6E5MQGnmNlEgzHcGuRY8gxOMzbxmBgcCPHDKiFOYFBIgfoQjY8WvjfGBz++w+ipZgnod6eQf4NUMs/PA6TANrC2ADRwsyTAGRL8JhJMLbh1iI541nBwZ5jEgaSZ54VS85JO57YxpNWbJHYl45Ti8T55I0PftTYGPAdT9744Y1NtT0/++GNNz58s8aphYGBAxQLEgg+G4hIwKOBgYH9AV7pUTAKRsEoGAUMALY8UZX+Q+baAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-8073-8691","institution":"Wicking Dementia Research and Education Centre, University of Tasmania","correspondingAuthor":true,"prefix":"","firstName":"Kaylee","middleName":"D","lastName":"Rudd","suffix":""},{"id":490605969,"identity":"83f83192-2f97-443f-b7bb-6f5d1ebd4225","order_by":1,"name":"Michele L Callisaya","email":"","orcid":"https://orcid.org/0000-0003-2122-1622","institution":"Menzies Institute for Medical Research, University of Tasmania","correspondingAuthor":false,"prefix":"","firstName":"Michele","middleName":"L","lastName":"Callisaya","suffix":""},{"id":490605970,"identity":"4d97541f-e9a6-43c4-96e6-be8bc4a4a636","order_by":2,"name":"Katherine Lawler","email":"","orcid":"https://orcid.org/0000-0002-1484-1113","institution":"School of Allied Health, Human Services and Sport, La Trobe University","correspondingAuthor":false,"prefix":"","firstName":"Katherine","middleName":"","lastName":"Lawler","suffix":""},{"id":490605971,"identity":"73599a38-c116-406c-846e-98f4cf7042cf","order_by":3,"name":"Renjie Li","email":"","orcid":"https://orcid.org/0000-0003-2374-5067","institution":"Wicking Dementia Research and Education Centre, University of Tasmania","correspondingAuthor":false,"prefix":"","firstName":"Renjie","middleName":"","lastName":"Li","suffix":""},{"id":490605972,"identity":"c2e7c4da-c0de-4b50-a2a3-db6cba08abb6","order_by":4,"name":"Jessica M Collins","email":"","orcid":"https://orcid.org/0000-0002-1938-2470","institution":"Wicking Dementia Research and Education Centre, University of Tasmania","correspondingAuthor":false,"prefix":"","firstName":"Jessica","middleName":"M","lastName":"Collins","suffix":""},{"id":490605973,"identity":"771c94e0-001d-4488-ae29-ff069724fe2f","order_by":5,"name":"Rebecca St George","email":"","orcid":"https://orcid.org/0000-0001-6206-1069","institution":"School of Psychological Sciences, University of Tasmania","correspondingAuthor":false,"prefix":"","firstName":"Rebecca","middleName":"St","lastName":"George","suffix":""},{"id":490606186,"identity":"1fdd4c85-bf2a-4e73-897a-0942a98a3961","order_by":6,"name":"Alastair J Noyce","email":"","orcid":"https://orcid.org/0000-0003-3027-5497","institution":"Wolfson Institute of Population Health, Queen Mary University of London","correspondingAuthor":false,"prefix":"","firstName":"Alastair","middleName":"J","lastName":"Noyce","suffix":""},{"id":490606187,"identity":"5eb9ca6b-feab-4b5c-9257-ee12a28475f0","order_by":7,"name":"James C Vickers","email":"","orcid":"https://orcid.org/0000-0001-5671-4879","institution":"Wicking Dementia Research and Education Centre, University of Tasmania","correspondingAuthor":false,"prefix":"","firstName":"James","middleName":"C","lastName":"Vickers","suffix":""},{"id":490606188,"identity":"5e0b8724-79a2-4f76-94e0-06a9693ba5d0","order_by":8,"name":"Anna E King","email":"","orcid":"https://orcid.org/0000-0003-1792-0965","institution":"Wicking Dementia Research and Education Centre, University of Tasmania","correspondingAuthor":false,"prefix":"","firstName":"Anna","middleName":"E","lastName":"King","suffix":""},{"id":490606189,"identity":"7e06e067-7a23-4431-8ba7-49e79c491379","order_by":9,"name":"Jane Alty","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA90lEQVRIiWNgGAWjYBACxgYwZcHAwN4AFTpAnBYJCQaeA1A2IS1QANQikUCkFub2swcfMNRI1BncfGP+4McfBjm+GwmMn3nwOawnL9mA4ZiEhMHtHMPG3jYGY8kbCczSeLU05JhJMLBBtDTwNjAkbriRwIBfS/8b8x8M/4Babp4xbPzzh6EeqIX5N14tM3LMGBjbgFpu8Bg287AxJBjcSGDDb8uMN8YSiX0SkjPPpBXOlm2TMJx55mGb5Rw8Wgz7cww/fPhmw893/PCGj2/+2MjzHU8+fOMNPi0NQCIBiBUOgPkSDPDoxQXk4Qz86kbBKBgFo2AkAwDnDkz5B52ziwAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-5456-8676","institution":"Wicking Dementia Research and Education Centre, University of Tasmania","correspondingAuthor":true,"prefix":"","firstName":"Jane","middleName":"","lastName":"Alty","suffix":""}],"badges":[],"createdAt":"2025-07-25 03:03:33","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-7209767/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7209767/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":87691257,"identity":"3ded1e61-51d5-4e22-bd7a-cc329f6bc581","added_by":"auto","created_at":"2025-07-28 04:27:18","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":812376,"visible":true,"origin":"","legend":"\u003cp\u003eBar charts of dominant hand finger-tapping and key-tapping and grip strength measures and NfL for diagnostic groups of dementia (all types), MCI (all subtypes), SCI and HC. NfL, neurofilament light; HC, cognitively healthy controls; SCI, Subjective Cognitive Impairment; MCI, Mild Cognitive Impairment. Graphs are made in SPSS version 28.8.1.0.\u003c/p\u003e","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7209767/v1/0c138482d98fd8fcb826e412.jpg"},{"id":87691262,"identity":"be495ff2-52ff-4ea4-82cd-bf70357921fd","added_by":"auto","created_at":"2025-07-28 04:27:18","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":802479,"visible":true,"origin":"","legend":"\u003cp\u003eThis figure summarises the final model for diagnostics groups of dementia, MCI and SCI compared to HC.\u003cstrong\u003e \u003c/strong\u003e\u0026nbsp;Area under the receiver Operating Characteristic curve (AUC)\u003cstrong\u003e \u003c/strong\u003eof the final models for classification of dementia, MCI and SCI groups from HC. Each panel also shows the AUC of the Null Model and the AUC of the Model comprising the best combination of key-tapping variables. MCI, Mild Cognitive Impairment; SCI, Subjective Cognitive Impairment; HC, cognitively healthy controls.\u003c/p\u003e","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7209767/v1/d970f979d01520cad7940f57.jpg"},{"id":87691664,"identity":"0659639d-d5a3-4ee8-aac6-32850084a06b","added_by":"auto","created_at":"2025-07-28 04:43:19","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2554022,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7209767/v1/62815bf3-4d1f-478f-aaf0-105c0dfb24f5.pdf"},{"id":87691253,"identity":"d5aaa59e-7f57-47e8-b8f4-24d71953668a","added_by":"auto","created_at":"2025-07-28 04:27:18","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":37790,"visible":true,"origin":"","legend":"\u003cp\u003eDemographic characteristics of dementia and MCI subtypes.\u003c/p\u003e","description":"","filename":"SupplementaryTable1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7209767/v1/4953d43f58ea023c96f1780e.docx"},{"id":87691256,"identity":"56a62422-b271-4e3c-8bb1-bc22b9f41cb4","added_by":"auto","created_at":"2025-07-28 04:27:18","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":32477,"visible":true,"origin":"","legend":"\u003cp\u003eAssociations between measures of hand motor tests and NfL and cognitive diagnosis groups.\u003c/p\u003e","description":"","filename":"SupplementaryTable2.docx","url":"https://assets-eu.researchsquare.com/files/rs-7209767/v1/8692c7ba884aece81a837466.docx"},{"id":87691376,"identity":"709e16a3-a8fc-4ee2-b028-8bb4747242f3","added_by":"auto","created_at":"2025-07-28 04:35:18","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":29013,"visible":true,"origin":"","legend":"\u003cp\u003eAssociations between variables of finger-tapping, key-tapping, grip strength and NfL in the model adjusted for age, sex and education.\u0026nbsp;\u003c/p\u003e","description":"","filename":"SupplementaryTable3.docx","url":"https://assets-eu.researchsquare.com/files/rs-7209767/v1/ae8923deb55820e123a6b54f.docx"},{"id":87691258,"identity":"ab236543-68f7-4f40-9117-8072fbdbd0dc","added_by":"auto","created_at":"2025-07-28 04:27:18","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":49772,"visible":true,"origin":"","legend":"\u003cp\u003eClassification accuracy of individual upper limb motor variables and NfL for cognitive diagnosis of dementia, MCI and SCI in an adjusted model.\u003c/p\u003e","description":"","filename":"SupplementaryTable4.docx","url":"https://assets-eu.researchsquare.com/files/rs-7209767/v1/192feeb8f78e1a1faec95cf0.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eDigitised hand movement and plasma NfL are complementary biomarkers of the dementia continuum\u003c/p\u003e","fulltext":[{"header":"1 INTRODUCTION","content":"\u003cp\u003eDementia is one of the world\u0026rsquo;s leading causes of disability and mortality in older adults, and it is estimated that prevalence will triple by 2050.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e Early detection of dementia risk is a priority as this facilitates early intervention.\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e The \u0026ldquo;dementia continuum\u0026rdquo; begins with the preclinical stage, when dementia-related changes start in the brain, but there are no cognitive symptoms, followed by Subjective Cognitive Impairment (SCI) when the person reports cognitive impairment, but standard clinical assessments remain normal.\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e The next stages along the continuum are Mild Cognitive Impairment (MCI) followed by early-stage dementia.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e New blood-based biomarkers can aid early detection of dementia risk, which are considerably less invasive and costly than specialist brain imaging, such as Positron Emission Tomography scans (PET) scans, or cerebrospinal fluid (CSF) analysis, but are still limited by relative expense and lack of clinical availability.\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eAn accumulating body of evidence shows that upper limb motor function changes in the earliest stages of the continuum and that these subtle alterations can be detected and analysed using easily accessible equipment such as webcams and computer keyboards.\u003csup\u003e\u003cspan additionalcitationids=\"CR10 CR11 CR12 CR13 CR14\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e For example, our group\u0026rsquo;s prior work found computer key-tapping frequency and speed generally decline across the continuum while variability of rhythm increases.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e Weaker grip strength is associated with cognitive impairment, although it may be preserved in the early stages of the continuum.\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e However, no previous study has evaluated how different upper limb motor functions are associated with all symptomatic stages of the dementia continuum or with blood-based biomarkers of neurodegeneration. One such blood-based biomarker of particular interest is neurofilament light (NfL), a measure of neuroaxonal injury; NfL levels have been found to increase in CSF and blood across a range of neurological disorders including different types of dementia, which makes it ideal as a non-specific marker of dementia risk.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e Furthermore, there is strong rationale for combining functional and biological measures, as so-called \u0026lsquo;multimodal\u0026rsquo; approaches (combining different \u0026lsquo;modes\u0026rsquo; of data such as brain imaging with cognitive function) improve classification of dementia and MCI from cognitively healthy controls (HC).\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e Considering the upper limb motor function changes across the dementia continuum, including in the preclinical and SCI, using combined models of upper limb motor measures and NfL may well improve discrimination of the cognitive groups. This multimodal approach holds promise to better understand the motor signatures across the continuum and hence aid with developing more accessible and affordable tests to detect dementia early.\u003c/p\u003e\u003cp\u003eWe aimed to: 1. examine the associations between finger-tapping (frequency and variability), key-tapping (frequency and variability), grip strength and NfL, adjusting for age, sex and education; 2. determine the classification accuracy of variables of finger-tapping, key-tapping, grip strength and NfL for cognitive diagnosis of dementia, MCI and SCI compared to the Null Model comprising age, sex and education, and 3. examine the best combination of hand motor tests and NfL to optimize classification accuracy for cognitive diagnosis groups compared to the Null Model.\u003c/p\u003e\u003cp\u003eWe hypothesized that (i) impaired finger- and key-tapping performance (e.g., lower frequency and greater variability) would be associated with lower grip strength and increased levels of NfL, and that (ii) combining hand motor and NfL would improve classification accuracy of cognitive diagnoses compared to models with an individual measure.\u003c/p\u003e"},{"header":"2 METHODS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Ethics and consent\u003c/h2\u003e\u003cp\u003eThis study is a part of the ISLAND Cognitive Clinic (hereon referred to as the Clinic) study and approved by the Human Research Ethics Committee of the University of Tasmania (ref. H0018639). All participants\u0026rsquo; capacity to consent to research was assessed by a clinician before giving written informed consent to participate in this study.\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e Healthy controls gave written informed consent after receiving study information and were given the opportunity to have their questions answered.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Participants and recruitment\u003c/h2\u003e\u003cp\u003eWe recruited consecutive participants from adults attending the Clinic at the University of Tasmania, Australia. The Clinic protocol has been published in detail elsewhere.\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e In brief, participants received an interdisciplinary consensus diagnosis according to international diagnostic guidelines after a structured medical history, neurological examination, global cognitive screening test (Addenbrookes Cognitive Examinaton-3; ACE-III), a one-hour neuropsychological battery, a range of blood tests, and magnetic resonance imaging (MRI) brain scan with morphometry.\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e Participants with a diagnosis of dementia (specifically AD, vascular dementia (VaD), or mixed dementia (AD and VaD)), MCI (grouped as amnestic (aMCI) and non-amnestic (nMCI)) and SCI were included in this study.\u003c/p\u003e\u003cp\u003eHC with no subjective or objective cognitive impairment were recruited from members of the ISLAND Project, of which the detailed protocol has previously been published.\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e In short, this project is a 10-year population-based health initiative to reduce Tasmanians\u0026rsquo; risk of dementia through education and personalized feedback on modifiable risk factors. The participants in the ISLAND Project generally have a high educational attainment, with 21.5% completing a bachelor\u0026rsquo;s degree, compared to 11.6% Tasmanian average for adults aged\u0026thinsp;\u0026ge;\u0026thinsp;50 years.\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e Recognizing that this high cognitive reserve may buffer any functional decline with dementia-related brain pathology,\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e we purposefully set the ACE-III eligibility cutoff for HC to ACE-III\u0026thinsp;\u0026ge;\u0026thinsp;95/100. Scores below 95 have previously been shown to have a high sensitivity (83%) and specificity (73%) for MCI.\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3 Data collection\u003c/h2\u003e\u003cp\u003eWe collected data on age (years), sex (male/female/prefer not to say), self-reported hand dominance (hand used for writing (right/left)), full-time equivalent years of education, total score of a mood questionnaire - Geriatric Depression Scale (GDS) and the Hospital Anxiety and Depression Scale (HADS) for the Clinic and HC groups respectively, height/weight (measured in-person) and presence of osteoarthritis and pain in hands (recorded as yes/no). Additionally, we recorded the total and sub-scores of ACE-III, representing the cognitive domains of memory, attention, language, verbal fluency and visuo-spatial function. We also recorded the time (in seconds) taken to complete the Trail Making Test parts A and B, representing domains of motor sequencing and executive function respectively.\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e In terms of the motor tasks, all participants completed the grip strength measures first, then key-tapping and lastly finger-tapping test, supervised by the first author who was blinded to the cognitive diagnosis. NfL levels were calculated and used in the data analysis for a subset of participants who had provided blood samples.\u003c/p\u003e\u003cdiv id=\"Sec6\" class=\"Section3\"\u003e\u003ch2\u003e2.3.1 Finger-tapping assessment\u003c/h2\u003e\u003cp\u003eParticipants completed a finger-tapping test seated in a chair in a well-lit Clinic room. The task was recorded at 60 frames/second using a digital camera (Canon G9X Mark II; 1920 x 1080- pixel resolution), mounted on a tripod placed at 1.2m from the assessment chair. The task was demonstrated by a researcher, and participants had a chance to practice and seek clarification before recording. Participants were instructed to keep their hands at shoulder level and simultaneously tap both index fingers against their thumbs \u0026ldquo;as big and as fast\u0026rdquo; as they could for 15s. To alleviate the effect of a delayed start and/or fatigue, motor variables from the end of the first second to the end of the 11th second of the task (10s in total) were extracted for analysis.\u003c/p\u003e\u003cp\u003eTo extract finger-tapping features from recorded videos, we used a Computer Vision system called Rapid-Motion-Track that has been validated against wearable sensors.\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e Two finger-tapping variables were calculated for each hand and analysed: frequency (number of finger-taps in 30s) and variability (coefficient of variance - CV - of finger-tapping cycle time). These variables have been used in previous studies of hand motor-cognitive associations.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan additionalcitationids=\"CR29 CR30 CR31 CR32 CR33\" citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section3\"\u003e\u003ch2\u003e2.3.2 Computer key-tapping assessment\u003c/h2\u003e\u003cp\u003eThe computer keyboard tapping test was completed using an online alternate tapping task called the BRadykinesia Akinesia INcoordination (BRAIN) test. In this test, participants tapped two computer keys 15cm apart \u0026ldquo;as quickly and accurately as possible\u0026rdquo; for 30s using the index finger of the left hand and then repeated using the right hand as per BRAIN test protocol.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e Two key-tapping variables were selected: frequency (number of key-taps in 30s) and variability (CV of the time interval between key-taps). These variables are also conceptually equivalent to selected finger-tapping variables as finger-tapping and key-taping are not comparable like-for-like.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section3\"\u003e\u003ch2\u003e2.3.3 Grip strength\u003c/h2\u003e\u003cp\u003eWe used a JAMAR Plus Digital Hand Grip dynamometer that has been shown to have good to excellent test-retest reliability in older adults.\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e Three trials of grip strength were measured with participants\u0026rsquo; elbows at a 90\u0026deg; angle following the Southampton protocol.\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e The maximum score from each hand was recorded in kilograms.\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section3\"\u003e\u003ch2\u003e2.3.4 Blood-based biomarker\u003c/h2\u003e\u003cp\u003eThe detailed method of blood sample collection, processing and storage prior to analysis has been published previously.\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e In brief, non-fasting venous blood samples were processed within two hours of collection. To prepare serum, blood was clotted at room temperature for 30 minutes and then centrifuged (2,000 g) for 10 minutes at 4\u0026deg;C.\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e Aliquoted serum samples were stored in polypropylene, screw-top cryostorage tubes at \u0026minus;\u0026thinsp;80\u0026deg;C. Plasma NfL was measured using the Neurology 2-Plex B assay (Simoa\u0026reg;) from Quanterix according to the manufacturer\u0026rsquo;s protocols. Samples were diluted x4 by adding 25\u0026micro;L of serum to 75\u0026micro;L of sample diluent.\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e Results were run against high- and low-quality control samples provided by Quanterix and were excluded from analysis if the intra-assay CV between duplicates was \u0026gt;\u0026thinsp;20%.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e2.4 Statistical analysis\u003c/h2\u003e\u003cp\u003eThe skewness of the distribution of each variable was evaluated using the P-P plot. Finger-tapping frequency per second was multiplied by 30 to represent the number of finger-taps in 30s. Mood scores were derived from GDS (out of 15) for the Clinic groups and the HADS (out of 21) HC and so, algebraic rearrangement was used to transform the proportion of HADS scores before analysing the differences between groups.\u003c/p\u003e\u003cp\u003eAssociations between variables of finger-tapping, key-tapping, grip strength and NfL were analysed using partial correlation adjusted for age, sex and education. The area under the Receiver Operating Characteristic curves (AUC) was calculated to estimate classification accuracy of cognitive diagnoses. For all AUC calculations, we controlled for independent variables of age, sex and education that are shown to affect motor and cognitive functions and have been adjusted for in other studies.\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e In Model One we examined classification accuracy of the Null Model (comprising age, sex, and education) plus each hand motor variable and NfL separately. We rejected the null hypothesis, that hand motor/NfL measures do not improve classification over and above covariates, if there was a significant difference between AUC of Models Null and One.\u003c/p\u003e\u003cp\u003eIn Model Two, we started with the motor variables from Model One with the highest AUC and then added each hand motor variable one at a time in the order of strongest to weakest AUC and p-value to assess whether they improved the AUC. If they did not increase the AUC, the variable was taken out and the next variable was added. This process was repeated until the best hand motor model for each diagnostic group was identified. Finally, and in the subset of participants with both hand motor and NfL measures, we added NfL to the model to determine if it provided any improvement in the AUC over and above the hand motor variables.\u003c/p\u003e\u003cp\u003eFor all statistical analysis, and considering the exploratory design of this study, the significance was set at \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05 as utilized by previous studies with similar design.\u003csup\u003e\u003cspan additionalcitationids=\"CR42\" citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e Statistical analysis was completed using STATA 18.0. Bar charts were generated using SPSS version 28.8.1.0.\u003c/p\u003e\u003c/div\u003e"},{"header":"3 RESULTS","content":"\u003cp\u003e We recruited 317 participants: 71 with dementia, 105 with MCI, 59 with SCI and 82 HC. Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e shows the characteristics and the mean (Standard Deviation) of upper limb motor and NfL measures for diagnostic groups. The dementia group comprised 42 AD, 8 VaD and 21 mixed AD/VaD dementia, and the MCI group comprised 69 participants with aMCI and 36 with nMCI. Supplementary Table\u0026nbsp;1 shows the characteristics of dementia and MCI subtypes.\u003c/p\u003e\u003cp\u003eOverall, for finger-tapping, the HC group had the slowest frequency, followed by all-cause dementia, MCI (all subtypes) and the SCI group that had the highest finger-tapping frequency. There were no differences in finger-tapping variability between groups. Supplementary Table\u0026nbsp;2 presents the between-group associations for demographic and study variables.\u003c/p\u003e\u003cp\u003eAcross the continuum, HC had the highest key-tapping frequency, followed by SCI and MCI (all subtypes) and dementia (all types). The dementia group had the greatest key-tapping variability, followed by MCI, compared to SCI or HC. Figure\u0026nbsp;1 visualizes the average values of upper limb motor (dominant hand) and NfL across the continuum.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eCharacteristics of participants in each cognitive group.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDementia\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMCI\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eSCI\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eHC\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003en: 71\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003en: 105\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003en: 59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003en: 82\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge (years)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e75.63 (7.79)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e70.00 (9.56)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e65.12 (8.73)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e65.09 (6.98)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSex (Female)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003en (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e39 (55)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e64 (61)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e40 (68)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e67 (82)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEducation (years)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e11.89 (2.88)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e13.03 (3.29)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e14.85 (3.16)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e15.12 (2.53)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHeight (m)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.65 (.10)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.66 (.10)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.66 (.09)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.64 (.08)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWeight (kg)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e71.80 (15.29)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e77.06 (19.44)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e77.96 (17.73)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e73.87 (15.29)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight-handed\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003en (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e67 (94)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e93 (89)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e55 (93)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e78 (95)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHand osteoarthritis\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003en (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e43 (61)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e52 (50)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e18 (31)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e32 (39)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePain in hands\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003en (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e23 (32)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e37 (35)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e12 (21)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e24 (29)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eACE-III Scores\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003en: 66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003en: 92\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003en: 49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003en: 82\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal (/100)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e77.96 (10.95)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e88.92 (7.57)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e93.63 (5.22)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e97.67 (1.61)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAttention (/18)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e14.70 (3.03)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e16.46 (1.78)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e17.14 (1.63)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e17.72 (.59)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMemory (/26)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e16.96 (4.60)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e21.29 (4.25)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e23.12 (2.37)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e25.66 (.63)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFluency (/14)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8.49 (3.16)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e11.05 (2.12)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e12.35 (1.97)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e12.92 (.98)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLanguage (/26)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e23.97 (2.14)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e25.19 (1.34)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e25.55 (.91)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e25.68 (.65)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVisuo-spatial (/13)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e13.85 (2.26)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e14.89 (1.42)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e15.71 (1.81)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e15.70 (.68)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTMT - A\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003en: 54\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003en: 100\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003en: 56\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003en: 82\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime (s)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e46.70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e36.82 (11.48)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e31.13 (10.29)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e29.93 (8.13)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTMT - B\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003en: 48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003en: 94\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003en: 56\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003en: 82\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime (s)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e158.12 (83.75)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e112.77\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e73.39 (31.16)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e58.25 (15.49)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMood Score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003en: 71\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003en: 105\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003en: 56\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003en: 82\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(/15)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e4.88 (3.24)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e4.59 (3.12)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e4.71 (3.33)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGrip strength (kg)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003en: 69\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003en: 104\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003en: 58\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003en: 82\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNon-dominant hand\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e26.00 (9.20)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e27.81 (11.00)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e29.42 (8.95)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e30.62 (8.11)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDominant hand\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e27.64 (9.33)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e29.43 (10.50)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e31.96 (9.59)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e32.48 (8.78)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFinger-tapping\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003en: 64\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003en: 100\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003en: 57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003en: 45\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency (N)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e50.68 (21.08)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e53.94 (18.27)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e57.91 (21.83)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e42.45 (13.77)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency (D)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e47.90 (18.13)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e54.17 (19.41)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e60.97 (21.83)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e43.36 (12.86)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariability (N)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e32.46 (26.46)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e23.84 (18.93)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e21.03 (13.98)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e25.40 (19.59)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariability (D)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e29.85 (28.04)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e25.75 (20.80)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e21.00 (15.73)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e22.98 (16.31)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eKey-tapping\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003en: 45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003en: 51\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003en: 40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003en: 57\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency (N)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e43.02 (12.32)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e50.98 (10.15)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e59.75 (10.94)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e66.21 (8.55)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency (D)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e47.69 (14.05)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e56.28 (9.72)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e66.73 (11.00)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e71.70 (9.18)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariability (N)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e142.01 (129.59)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e86.68 (35.47)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e66.04 (40.61)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e45.26 (19.51)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariability (D)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e139.22 (136.38)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e94.91 (63.13)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e58.63 (27.11)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e45.67 (29.08)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNfL (pg/mL)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003en: 40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003en: 39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003en: 24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003en: 39\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e32.83 (14.04)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e23.51 (12.92)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e19.01 (8.74)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e13.43 (4.88)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e \u003cb\u003eLegend\u003c/b\u003e: When adjusted for age, there were no differences in the reported pain or osteoarthritis between groups. There were no differences between groups in height/weight or mood scores. HADS Mean (SD) before transformation of proportion was 6.60 (4.66)/21. Abbreviations: MCI, Mild Cognitive Impairment; SCI, Subjective Cognitive Impairment; HC, cognitively healthy controls; AD, Alzheimer\u0026rsquo;s Disease; VaD, Vascular dementia; n, number; SD, standard deviation; Y, answering \u0026ldquo;Yes\u0026rdquo; to having osteoarthritis or pain in hands; N, non-dominant hand; D, dominant hand; ACE-III, Addenbrooke\u0026rsquo;s Cognitive Examination-version 3; TMT, Trail Making Test; NfL, neurofilament light; pg/mL, pictograms per millilitre.\u003c/p\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e3.1 Associations between hand motor variables and NfL\u003c/h2\u003e\u003cp\u003eObserved correlations between measures were weak to moderate; higher key-tapping frequency was associated with stronger grip measures and lower NfL levels. Greater finger-tapping variability was associated with greater key-tapping variability, weaker grip and higher NfL levels. Weaker grip measures were associated with higher NfL levels.\u003c/p\u003e\u003cp\u003eSupplementary Table\u0026nbsp;3 presents the partial correlation (corr.) between variables adjusted for age, sex and education. In adjusted models, finger-tapping and key-tapping variability of non-dominant hand were associated (corr.: 0.254, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Key-tapping frequency of each hand was correlated with grip strength of the same hand (non-dominant corr.: 0.228, \u003cem\u003eP\u003c/em\u003e: 0.002; dominant corr.: 0.245, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Key-tapping variability (dominant hand) was associated with grip strength of the same hand (corr.: -0.252, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Key-tapping frequency and variability (non-dominant hand only) were associated with NfL (frequency corr.: -0.253, \u003cem\u003eP\u003c/em\u003e: 0.008, variability corr.: 0.426, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Grip strength of either hand was associated with NfL (non-dominant corr.: -0.248, \u003cem\u003eP\u003c/em\u003e: 0.003 and dominant corr.: -0.240, \u003cem\u003eP\u003c/em\u003e: 0.004).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003e3.2 Classification accuracy of motor variables and NfL\u003c/h2\u003e\u003cp\u003eSupplementary Table\u0026nbsp;4 presents the accuracy of individual upper limb motor variables and NfL for classifying cognitive diagnoses. Finger-tapping frequency of either hand improved classification of the SCI group (but not dementia or MCI) from HC (AUC: \u0026ge; 0.79, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026le;\u0026thinsp;0.006). Compared to the Null Model, key-tapping frequency and variability of either hand improved classification of dementia (all AUCs\u0026thinsp;\u0026ge;\u0026thinsp;0.94; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026le;\u0026thinsp;0.035), and MCI (all AUCs\u0026thinsp;\u0026ge;\u0026thinsp;0.83; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026le;\u0026thinsp;0.008) from HC, but not SCI. Grip strength did not aid with classification of any diagnostic group from HC. NfL improved discrimination of only the dementia group from HC.\u003c/p\u003e\u003cp\u003eCompared to the Null Model, no measure improved classification of dementia, or MCI from SCI, or aided with discriminating dementia from MCI.\u003c/p\u003e\u003cp\u003e3.3 The best combination of hand motor tests and NfL to optimize classification accuracy of cognitive diagnosis groups\u003c/p\u003e\u003cp\u003eTo differentiate dementia from HC, the best combination of key-tapping variables was key-tapping frequency of non-dominant hand (AUC: 0.95). Combining NfL with the non-dominant hand key-tapping frequency slightly improved the classification (AUC: 0.97, \u003cem\u003eP\u003c/em\u003e: 0.021) of dementia from HC. The best combination of variables to discriminate MCI from HC was key-tapping frequency of both dominant and non-dominant hands and variability of the non-dominant hand (AUC: 0.96). The addition of NfL slightly improved accuracy (AUC: 0.97). The best combination of variables to discriminate SCI from HC was all four key-tapping variables, e.g., frequency of both dominant and non-dominant hands and variability of both dominant and non-dominant hands, (AUC: 0.76). Adding NfL to the model improved classification accuracy (AUC: .81).\u003c/p\u003e\u003cp\u003eFigure 2 and Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e present classification accuracy of combined hand motor and NfL measures for cognitive diagnosis groups compared to the Null Model and the model comprising covariates and hand motor measures in the subset of participants with both hand motor and NfL measures.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eClassification accuracy of combined upper limb motor and NfL measures for cognitive diagnosis groups in an adjusted model.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAUC\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003e95% CI\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003eP\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eHC v Dementia\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003en: 59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNull\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.85\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.75; 0.95\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNull\u0026thinsp;+\u0026thinsp;Key-tapping frequency (N)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.95\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.88; 1.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.073\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNull\u0026thinsp;+\u0026thinsp;Key-tapping frequency (N)\u0026thinsp;+\u0026thinsp;NfL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.93; 1.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.021\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eHC v MCI\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003en: 60\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNull\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.73\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.60; 0.87\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNull\u0026thinsp;+\u0026thinsp;Key-tapping frequency (N\u0026thinsp;+\u0026thinsp;D)\u0026thinsp;+\u0026thinsp;Key-tapping variability (N)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.92; 1.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNull\u0026thinsp;+\u0026thinsp;Key-tapping frequency (N\u0026thinsp;+\u0026thinsp;D)\u0026thinsp;+\u0026thinsp;Key-tapping variability (N)\u0026thinsp;+\u0026thinsp;NfL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.94; 1.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eHC v SCI\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003en: 46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNull\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.54\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.36; 0.72\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNull\u0026thinsp;+\u0026thinsp;Key-tapping frequency (N\u0026thinsp;+\u0026thinsp;D)\u0026thinsp;+\u0026thinsp;Key-tapping variability (N\u0026thinsp;+\u0026thinsp;D)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.76\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.61; 0.90\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.067\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNull\u0026thinsp;+\u0026thinsp;Key-tapping frequency (N\u0026thinsp;+\u0026thinsp;D)\u0026thinsp;+\u0026thinsp;Key-tapping variability (N\u0026thinsp;+\u0026thinsp;D)\u0026thinsp;+\u0026thinsp;NfL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.68; 0.93\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.018\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e \u003cb\u003eLegend\u003c/b\u003e: Model Two was developed using the variables with the strongest AUC and p-value from Model One. \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05 indicates added variable improves prediction of diagnosis over and above the Null Model. Abbreviations: HC, cognitively healthy controls; MCI, Mild Cognitive Impairment; SCI, Subjective Cognitive Impairment; AUC, area under the Receiver Operating Characteristic curve; CI, Confidence Interval; \u003cem\u003eP\u003c/em\u003e, p-value. N, nondominant hand; D, dominant hand; NfL, neurofilament light.\u003c/p\u003e\u003c/div\u003e"},{"header":"4 DISCUSSION","content":"\u003cp\u003eWe examined three digitised upper limb motor function tests and a blood-based biomarker of neurodegeneration in large groups of clinically diagnosed participants with subjective and objective cognitive impairment, as well as healthy controls. In models adjusted for age, sex and education we found associations between slower, less rhythmic key-tapping, weaker grip strength and higher NfL levels. We also found that individually, classification accuracy of upper limb motor function and NfL measures for diagnostic groups were different. Our main finding was that combining key-tapping with NfL measures significantly improved classification of all cognitive diagnosis groups from HC with adjusted accuracies of 97%, 97% and 81% for dementia, MCI and SCI respectively. Considering the best results only required data from a short test of motor function on a standard computer keyboard, our findings are significant in developing an accessible clinical tool.\u003c/p\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003e4.1 Associations between motor variables and NfL\u003c/h2\u003e\u003cp\u003eWe report a novel finding: that impaired hand function, indicated by lower key-tapping frequency, greater key-tapping variability and lower grip strength, was associated with increased NfL levels. These correlations have not been examined in groups with a cognitive diagnosis and require further investigation; Two previous studies investigated associations between grip strength and NfL in community-dwelling adults without a cognitive diagnosis but had contradictory findings; the first, a cross-sectional study of 1925 adults aged 20\u0026ndash;75 years, reported those with higher NfL levels had lower grip strength.\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e However, another study of 507 adults aged 71\u0026ndash;81 years examining grip strength and NfL cross-sectionally and at 2-year follow-up found no associations at either analysis.\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eAssociations found between each of the hand motor tasks were weaker than we expected considering available evidence from neuroimaging studies. Looking at finger-tapping and key-tapping tasks first, sparse brain MRI studies in healthy individuals show activation of similar brain areas including the primary and supplementary motor cortices, parietal lobe, and basal ganglia during both tasks.\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e This would imply similarities in these tasks\u0026rsquo; neural pathways, but two recent studies examined the associations between tapping tests and brain structure in people with cognitive impairment present a slightly different picture. One investigated the age/sex-adjusted associations between computer key-tapping measures and hippocampal volume on MRI scans in 26 persons with AD, 27 aMCI and 47 HC.\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e They found lower speed and more variability on key-tapping associated with smaller hippocampal volumes. On the other hand, another study investigated the age/sex-adjusted associations between finger-tapping measures and whole brain MRI scan in 71 individuals with AD and 65 MCI and reported lower tapping frequency associated with reduced grey matter volume in the primary motor cortex.\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eWhile both tapping tasks activate the motor cortex, the degree and distribution of activation may vary.\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e Key-tapping requires more precision and speed control when alternately tapping small keys; hence, as a more complex task may activate larger areas of the brain compared to finger-tapping. With no need for precision or following visual input, finger-tapping can be considered a simpler task that uses established/learned motor patterns in the motor cortex.\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e The complexity of these tasks, therefore, might explain the generally weak associations found between finger-tapping and key-tapping measures.\u003c/p\u003e\u003cp\u003eCorrelations between tapping tasks and grip strength were also weak although brain MRI analysis during grip strength test shows activation of brain regions similar to key-tapping and finger-tapping tests.\u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e This finding might also be related to differences in tasks\u0026rsquo; complexity and cognitive requirements; compared to tapping tasks, grip strength measure requires attention and controlled activation of muscles.\u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e This test might also rely more on the palm/digits\u0026rsquo; tactile feedback than on visual input, hence using slightly different neural pathways than used in tapping tasks.\u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e Further research is necessary to understand the associations between upper limb motor function and brain structure.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003e4.2 Classification accuracy of hand motor variables and NfL\u003c/h2\u003e\u003cp\u003eConsidering the associations found between finger-tapping frequency and cognitive groups, we expected this variable to classify cognitive diagnosis groups, especially dementia and MCI. However, we found finger-tapping frequency only accurately classified SCI from HC. Finger-tapping frequency of our SCI group was higher than all other groups, including HC. The exact reason for this finding is unclear, but it may be related to prioritization of an aspect of the task by different groups. Given the same instruction was given to all participants, the HC group appears to have prioritized amplitude and tap \u0026ldquo;big\u0026rdquo;, hence having a lower frequency while cognitive diagnosis groups put speed first and tapped faster. To our knowledge, no other study has tested finger-tapping performance in SCI, hence further research is required.\u003c/p\u003e\u003cp\u003eSCI is a heterogeneous and less explored stage of the continuum; contradictory findings are reported by the few studies investigating upper limb motor-cognitive associations in SCI. Our previous study showed greater key-tapping variability was associated with SCI as well as MCI and dementia\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e, but another study reported no differences in key-tapping frequency or variability across 11 individuals with AD, 19 aMCI, 23 SCI and 12 HC.\u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e However, they had small sample sizes and used a different key-tapping test, involving 10s tapping of a lever with the index finger that might be less demanding on neural pathways than an alternate key-tapping test. Different methods of key-tapping tests make comparison of findings complicated, underlining careful consideration of the methods used to test motor function before drawing conclusions.\u003c/p\u003e\u003cp\u003eIn our study, grip strength did not improve classification of any diagnostic group, although it has been reported in longitudinal studies as a predictor of future conversion to dementia.\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e,\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e Our finding, however, aligns with recent evidence suggesting grip strength is often preserved in the early stages of dementia and that testing upper limb functions such as key-tapping provides more information on motor-cognitive associations.\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eWe found that NfL only improved classification accuracy of dementia from HC. NfL-cognition associations need further exploration as current evidence is contradictory. For example, a four-year longitudinal study reported no association between NfL and cognition\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e, and another study with 15\u0026ndash;30 months follow-up found significant associations.\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e In our study, NfL levels were associated with cognitive diagnosis groups and yet, NfL did not improve classification of MCI, or SCI, from HC. This highlights the importance of future research with larger sample sizes.\u003c/p\u003e\u003cp\u003e4.3 The best combination of hand motor tests and NfL to optimize classification accuracy of cognitive diagnosis groups\u003c/p\u003e\u003cp\u003eCombining key-tapping measures with NfL significantly improved classification accuracy of all cognitive diagnoses including SCI. Our findings are important especially considering cognitive tests cannot discriminate SCI from HC. Using a combination of measures from different sources is common practice in clinical investigation of health disorders including dementia. \u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e Advances in identification of dementia-related pathology in the brain and blood has led to development of tests such as PET scans and blood-based biomarkers. These tests have made the investigation of novel multimodal models possible.\u003c/p\u003e\u003cp\u003eIn recent years, various combinations of biomarkers, including cognitive, motor, genetic, brain imaging and blood-based, have been investigated for classification of dementia and MCI and there are also studies currently under way.\u003csup\u003e\u003cspan additionalcitationids=\"CR57 CR58\" citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e However, to our knowledge, no study has examined a combination of hand motor function and NfL. Current evidence shows that combining different biomarkers generally improves classification accuracy of dementia and MCI.\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e,\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e For example, one study reported combining MRI and CSF biomarkers classified 93% of AD and 76% of MCI compared to 87% and 72% individually.\u003csup\u003e60\u003c/sup\u003e Still, models using brain imaging and CSF are limited by costs and low accessibility. Emerging evidence on digital biomarkers of dementia has provided new opportunities in developing combined models. Our study used digitised tests, accessible using everyday devices, and a plasma NfL measure that are considerably less costly and invasive.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003e4.4 Strengths and limitations\u003c/h2\u003e\u003cp\u003eWe examined three upper limb motor tests in large samples of consecutively recruited participants from a cognitive clinic and included NfL. Diagnosis of our cognitive groups was based on a robust interdisciplinary consensus diagnosis after comprehensive gold-standard assessment, rather than a global cognitive screening test. We used a well-established protocol to measure grip strength and a simple, short and validated key-tapping test. We also tested finger-tapping; commonly used in clinics and research.\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e Additionally, we adjusted all models for age, sex and education.\u003c/p\u003e\u003cp\u003eLimitations are also acknowledged; although participants had rests in-between tests, the finger-tapping results could have been affected by fatigue as they were completed after grip strength and key-tapping tests. Future investigators could consider alternating tests\u0026rsquo; order to alleviate fatigue. We acknowledge that the durations of finger-tapping and key-tapping tasks were not the same in this study and that it may have affected our findings. Additionally, we did not investigate other measures of neurodegeneration such as hippocampal volume in this cross-sectional study. Future research could consider investigating these associations longitudinally, consider using the same duration for tapping tasks and include other neurodegeneration measures. Finally, future studies should consider adjusting NfL levels for renal function and other comorbidities that may impact NfL levels.\u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eIn conclusion, a combination of digitised upper limb motor test variables and NfL classified all symptomatic stages of the dementia continuum, indicating potential benefits of further exploring this multimodal approach in detecting those in early stages of dementia.\u003c/p\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eCONFLICT OF INTEREST STATEMENT\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors have no competing interests to declare.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSOURCES OF FUNDING\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRoyal Hobart Hospital Research Foundation RHHRF (20-003) grant supported plasma NfL analysis in this study. Grant title: Blood-based biomarkers for neurodegeneration dementia, Major Project (2022), University of Tasmania.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eKR is supported by an Australian Government Research Training Program living allowance and University of Tasmania tuition fee scholarship. The ISLAND Clinic and ISLAND Project were supported by grants from the Medical Research Future Fund, the National Health and Medical Research Council, Tasmanian Masonic Medical Research Foundation, St Lukes Health and the Hopkins Foundation. The funding bodies have no direct role in the study\u0026rsquo;s design, data collection, analysis, interpretation, or manuscript preparation. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCONSENT STATEMENT\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll study participants provided written informed consent.\u0026nbsp;This study was approved by the Human Research Ethics Committee of the University of Tasmania.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eNichols E, Vos T. The estimation of the global prevalence of dementia from 1990-2019 and forecasted prevalence through 2050: An analysis for the Global Burden of Disease (GBD) study 2019. \u003cem\u003eAlzheimer\u0026apos;s \u0026amp; Dementia\u003c/em\u003e. 2021;17(S10):e051496. doi:https://doi.org/10.1002/alz.051496\u003c/li\u003e\n\u003cli\u003eJack Jr CR, Knopman DS, Jagust WJ, et al. Tracking pathophysiological processes in Alzheimer\u0026apos;s disease: an updated hypothetical model of dynamic biomarkers. \u003cem\u003eThe Lancet Neurology\u003c/em\u003e. 2013;12(2):207-216. doi:https://doi.org/10.1016/s1474-4422(12)70291-0\u003c/li\u003e\n\u003cli\u003eGarcia-Ptacek S, Eriksdotter M, Jelic V, Porta-Etessam J, K\u0026aring;reholt I, Manzano Palomo S. Subjective cognitive impairment: Towards early identification of Alzheimer disease. \u003cem\u003eNeurologia (English Edition)\u003c/em\u003e. 2016/10/01/ 2016;31(8):562-571. doi:https://doi.org/10.1016/j.nrleng.2013.02.011\u003c/li\u003e\n\u003cli\u003eDubois B, Villain N, Frisoni GB, et al. Clinical diagnosis of Alzheimer\u0026apos;s disease: recommendations of the International Working Group. \u003cem\u003eThe Lancet Neurology\u003c/em\u003e. 2021;20(6):484-496. doi:https://doi.org/10.1002/alz.051167\u003c/li\u003e\n\u003cli\u003eMitchell AJ, Beaumont H, Ferguson D, Yadegarfar M, Stubbs B. Risk of dementia and mild cognitive impairment in older people with subjective memory complaints: meta-analysis. \u003cem\u003eActa Psychiatrica Scandinavica\u003c/em\u003e. 2014;130(6):439-451. doi:https://doi.org/10.1111/acps.12336\u003c/li\u003e\n\u003cli\u003eJack Jr CR, Bennett DA, Blennow K, et al. NIA‐AA research framework: toward a biological definition of Alzheimer\u0026apos;s disease. \u003cem\u003eAlzheimer\u0026apos;s \u0026amp; dementia\u003c/em\u003e. 2018;14(4):535-562. doi:https://doi.org/10.1016/j.jalz.2018.02.018\u003c/li\u003e\n\u003cli\u003eMattsson N, Andreasson U, Zetterberg H, Blennow K, Initiative AsDN. Association of plasma neurofilament light with neurodegeneration in patients with Alzheimer disease. \u003cem\u003eJAMA neurology\u003c/em\u003e. 2017;74(5):557-566. doi:https://doi.org/10.1001/jamaneurol.2016.6117\u003c/li\u003e\n\u003cli\u003eGaetani L, Blennow K, Calabresi P, Di Filippo M, Parnetti L, Zetterberg H. Neurofilament light chain as a biomarker in neurological disorders. \u003cem\u003eJournal of Neurology, Neurosurgery \u0026amp; Psychiatry\u003c/em\u003e. 2019;90(8):870-881. doi:https://doi.org/10.1136/jnnp-2018-320106\u003c/li\u003e\n\u003cli\u003eWang X, St George RJ, Bindoff AD, et al. Estimating presymptomatic episodic memory impairment using simple hand movement tests: A cross-sectional study of a large sample of older adults. \u003cem\u003eAlzheimer\u0026apos;s \u0026amp; Dementia\u003c/em\u003e. 2024;20(1):173-182. doi:https://doi.org/10.1002/alz.13401\u003c/li\u003e\n\u003cli\u003eMollica MA, Tort-Merino A, Navarra J, et al. Early detection of subtle motor dysfunction in cognitively normal subjects with amyloid-\u0026beta; positivity. \u003cem\u003eCortex\u003c/em\u003e. 2019;121:117-124. doi:https://doi.org/10.1016/j.cortex.2019.07.021\u003c/li\u003e\n\u003cli\u003eRudd KD, Lawler K, Callisaya ML, et al. Hand Motor Dysfunction Is Associated with Both Subjective and Objective Cognitive Impairment across the Dementia Continuum. \u003cem\u003eDementia and Geriatric Cognitive Disorders\u003c/em\u003e. Jul 29 2024:1-11. doi:https://doi.org/10.1159/000540412\u003c/li\u003e\n\u003cli\u003eCui M, Zhang S, Liu Y, Gang X, Wang G. Grip strength and the risk of cognitive decline and dementia: a systematic review and meta-analysis of longitudinal cohort studies. \u003cem\u003eFrontiers in aging neuroscience\u003c/em\u003e. 2021;13:625551. doi:https://doi.org/10.3389/fnagi.2021.625551\u003c/li\u003e\n\u003cli\u003eBuchman AS, Wilson RS, Boyle PA, Bienias JL, Bennett DA. Grip strength and the risk of incident Alzheimer\u0026rsquo;s disease. \u003cem\u003eNeuroepidemiology\u003c/em\u003e. 2007;29(1-2):66-73. doi:https://doi.org/10.1159/000109498\u003c/li\u003e\n\u003cli\u003eLi R. \u003cem\u003eUsing artificial intelligence (AI)-based computer vision methods to investigate associations between hand movement and Alzheimer\u0026rsquo;s Disease (AD) risk\u003c/em\u003e. University of Tasmania; 2024. \u003c/li\u003e\n\u003cli\u003eKoppelmans V, Ruitenberg MFL, Schaefer SY, et al. Classification of Mild Cognitive Impairment and Alzheimer\u0026rsquo;s Disease Using Manual Motor Measures. \u003cem\u003eNeurodegenerative Diseases\u003c/em\u003e. 2024:54-70. doi:https://doi.org/10.1159/000539800\u003c/li\u003e\n\u003cli\u003eLi R, Wang X, Lawler K, et al. Brief webcam test of hand movements predicts episodic memory, executive function, and working memory in a community sample of cognitively asymptomatic older adults. \u003cem\u003eAlzheimer\u0026apos;s \u0026amp; Dementia: Diagnosis, Assessment \u0026amp; Disease Monitoring\u003c/em\u003e. 2024/01/01 2024;16(1):e12520. doi:https://doi.org/10.1002/dad2.12520\u003c/li\u003e\n\u003cli\u003eHooyman A, Malek‐Ahmadi M, Fauth EB, Schaefer SY. Challenging the relationship of grip strength with cognitive status in older adults. \u003cem\u003eInternational Journal of Geriatric Psychiatry\u003c/em\u003e. 2021;36(3):433-442. doi:https://doi.org/10.1002/gps.5441\u003c/li\u003e\n\u003cli\u003eRudd KD, Callisaya ML, Lawler K, Noyce AJ, Vickers JC, Alty J. Stepping and tapping: combining motor tasks improves cognitive classification. \u003cem\u003eGeroScience\u003c/em\u003e. 2025/05/08 2025;doi:https://doi.org/10.1007/s11357-025-01678-7\u003c/li\u003e\n\u003cli\u003eHuang G, Li R, Bai Q, Alty J. Multimodal learning of clinically accessible tests to aid diagnosis of neurodegenerative disorders: a scoping review. \u003cem\u003eHealth Information Science and Systems\u003c/em\u003e. 2023/07/22 2023;11(1):32. doi:https://doi.org/10.1007/s13755-023-00231-0\u003c/li\u003e\n\u003cli\u003eAlty J, Lawler K, Salmon K, et al. A new one-stop interdisciplinary cognitive clinic model tackles rural health inequality and halves the time to diagnosis: Benchmarked against a national dementia registry. \u003cem\u003eInternational Journal of Geriatric Psychiatry\u003c/em\u003e. 2023;38(8):e5988. doi:https://doi.org/10.1002/gps.5988\u003c/li\u003e\n\u003cli\u003eSachdev PS, Blacker D, Blazer DG, et al. Classifying neurocognitive disorders: the DSM-5 approach. \u003cem\u003eNature Reviews Neurology\u003c/em\u003e. 2014/11/01 2014;10(11):634-642. doi:https://doi.org/10.1038/nrneurol.2014.181\u003c/li\u003e\n\u003cli\u003eBartlett L, Doherty K, Farrow M, et al. Island study linking aging and neurodegenerative disease (ISLAND) targeting dementia risk reduction: protocol for a prospective web-based cohort study. \u003cem\u003eJMIR Research Protocols\u003c/em\u003e. Mar 1 2022;11(3):e34688. doi:https://doi.org/10.2196/34688\u003c/li\u003e\n\u003cli\u003eBruno D, Schurmann Vignaga S. Addenbrooke\u0026rsquo;s cognitive examination III in the diagnosis of dementia: a critical review. \u003cem\u003eNeuropsychiatric disease and treatment\u003c/em\u003e. 2019:441-447. doi:https://doi.org/10.2147/NDT.S151253\u003c/li\u003e\n\u003cli\u003ePendlebury ST, Mariz J, Bull L, Mehta Z, Rothwell PM. MoCA, ACE-R, and MMSE versus the National Institute of Neurological Disorders and Stroke\u0026ndash;Canadian Stroke Network vascular cognitive impairment harmonization standards neuropsychological battery after tia and stroke. \u003cem\u003eStroke\u003c/em\u003e. 2012;43(2):464-469. doi:https://doi.org/10.1161/STROKEAHA.111.633586\u003c/li\u003e\n\u003cli\u003eBowie CR, Harvey PD. Administration and interpretation of the Trail Making Test. \u003cem\u003eNature protocols\u003c/em\u003e. 2006;1(5):2277-2281. doi:https://doi.org/10.1038/nprot.2006.390\u003c/li\u003e\n\u003cli\u003eLi R, Lao C, St. George R, et al. Rapid-Motion-Track: Markerless Tracking of Fast Human Motion with Deeper Learning. \u003cem\u003eIntelligence-Based Medicine\u003c/em\u003e. 2024;10(100162)doi:https://doi.org/10.1016/j.ibmed.2024.100162\u003c/li\u003e\n\u003cli\u003eLi R, Tran S, Garg S, Lawler K, Alty J, Bai Q. Parallel multi-scale networks with deep supervision for hand keypoint detection. \u003cem\u003earXiv preprint arXiv:211210275\u003c/em\u003e. 2021;doi:https://doi.org/10.48550/arXiv.2112.10275\u003c/li\u003e\n\u003cli\u003eBologna M, Guerra A, Colella D, et al. Bradykinesia in Alzheimer\u0026apos;s disease and its neurophysiological substrates. \u003cem\u003eClin Neurophysiol\u003c/em\u003e. Apr 2020;131(4):850-858. doi:https://doi.org/10.1016/j.clinph.2019.12.413\u003c/li\u003e\n\u003cli\u003eNoyce AJ, Nagy A, Acharya S, et al. Bradykinesia-akinesia incoordination test: validating an online keyboard test of upper limb function. \u003cem\u003ePLoS One\u003c/em\u003e. 2014;9(4):e96260. doi:https://doi.org/10.1371/journal.pone.0096260\u003c/li\u003e\n\u003cli\u003eNewby R, Muhamed S, Alty J, et al. Geste antagoniste effects on motor performance in dystonia\u0026mdash;a kinematic study. \u003cem\u003eMovement Disorders Clinical Practice\u003c/em\u003e. 2022;9(6):759-764. doi:https://doi.org/10.1002/mdc3.13505\u003c/li\u003e\n\u003cli\u003eSuzumura S, Kanada Y, Osawa A, et al. Assessment of finger motor function that reflects the severity of cognitive function. \u003cem\u003eFujita Med J\u003c/em\u003e. 2021;7(4):122-129. doi:https://doi.org/10.20407/fmj.2020-013\u003c/li\u003e\n\u003cli\u003eTomita Y, Tanaka S, Takahashi S, Takeuchi N. Detecting cognitive decline in community-dwelling older adults using simple cognitive and motor performance tests. \u003cem\u003eGeriatrics \u0026amp; Gerontology International\u003c/em\u003e. 2020;20(3):212-217. doi:https://doi.org/10.1111/ggi.13863\u003c/li\u003e\n\u003cli\u003eWilliams S, Zhao Z, Hafeez A, et al. The discerning eye of computer vision: Can it measure Parkinson\u0026apos;s finger tap bradykinesia? \u003cem\u003eJ Neurol Sci\u003c/em\u003e. Sep 15 2020;416:117003. doi:https://doi.org/10.1016/j.jns.2020.117003\u003c/li\u003e\n\u003cli\u003eAmo-Salas J, Olivares-Gil A, Garc\u0026iacute;a-Bustillo \u0026Aacute;, Garc\u0026iacute;a-Garc\u0026iacute;a D, Arnaiz-Gonz\u0026aacute;lez \u0026Aacute;, Cubo E. Computer Vision for Parkinson\u0026rsquo;s Disease Evaluation: A Survey on Finger Tapping. \u003cem\u003eHealthcare\u003c/em\u003e. 2024;12(4):439. doi:https://doi.org/10.3390/healthcare12040439\u003c/li\u003e\n\u003cli\u003eBohannon R. Test-Retest Reliability of Measurements of Hand-Grip Strength Obtained by Dynamometry from Older Adults: A Systematic Review of Research in the PubMed Database. \u003cem\u003eThe Journal of frailty \u0026amp; aging\u003c/em\u003e. 2017;6(2):83-87. doi:https://doi.org/10.14283/jfa.2017.8\u003c/li\u003e\n\u003cli\u003eRoberts HC, Denison HJ, Martin HJ, et al. A review of the measurement of grip strength in clinical and epidemiological studies: towards a standardised approach. \u003cem\u003eAge and ageing\u003c/em\u003e. 2011;40(4):423-429. doi:https://doi.org/10.1093/ageing/afr051\u003c/li\u003e\n\u003cli\u003eBeaudart C, Rolland Y, Cruz-Jentoft AJ, et al. Assessment of muscle function and physical performance in daily clinical practice: a position paper endorsed by the European Society for Clinical and Economic Aspects of Osteoporosis, Osteoarthritis and Musculoskeletal Diseases (ESCEO). \u003cem\u003eCalcified tissue international\u003c/em\u003e. 2019;105:1-14. doi:https://doi.org/10.1007/s00223-019-00545-w\u003c/li\u003e\n\u003cli\u003eMcGrath R, Vincent BM, Jurivich DA, et al. Handgrip strength asymmetry and weakness together are associated with functional disability in aging Americans. \u003cem\u003eThe Journals of Gerontology: Series A\u003c/em\u003e. 2021;76(2):291-296. doi:https://doi.org/10.1093/gerona/glaa100\u003c/li\u003e\n\u003cli\u003eCollins JM, Bindoff AD, Roccati E, Alty JE, Vickers JC, King AE. Does serum neurofilament light help predict accelerated cognitive ageing in unimpaired older adults? \u003cem\u003eFrontiers in Neuroscience\u003c/em\u003e. 2023;17:1237284. doi:https://doi.org/10.3389/fnins.2023.1237284\u003c/li\u003e\n\u003cli\u003eMielke MM, Syrjanen JA, Blennow K, et al. Plasma and CSF neurofilament light: relation to longitudinal neuroimaging and cognitive measures. \u003cem\u003eNeurology\u003c/em\u003e. 2019;93(3):e252-e260. doi:https://doi.org/10.1212/WNL.0000000000007767\u003c/li\u003e\n\u003cli\u003eYamada Y, Kobayashi M, Shinkawa K, et al. Characteristics of Drawing Process Differentiate Alzheimer\u0026rsquo;s Disease and Dementia with Lewy Bodies. \u003cem\u003eJournal of Alzheimer\u0026rsquo;s Disease\u003c/em\u003e. 2022;90(2):693-704. doi:https://doi.org/10.3233/JAD-220546\u003c/li\u003e\n\u003cli\u003eKoppelmans V, Duff K, Schaefer SY. Using a Functional Upper Limb Motor Task to Predict Cognitive and Functional Decline in Mild Cognitive Impairment and Alzheimer\u0026rsquo;s Disease. \u003cem\u003eNeurorehabilitation and Neural Repair\u003c/em\u003e. 2024;0(0):15459683241309591. doi:https://doi.org/10.1177/15459683241309591\u003c/li\u003e\n\u003cli\u003eOrchard SG, Polekhina G, Ryan J, et al. Combination of gait speed and grip strength to predict cognitive decline and dementia. \u003cem\u003eAlzheimer\u0026apos;s \u0026amp; dementia (Amsterdam, Netherlands)\u003c/em\u003e. 2022;14(1):e12353. doi:https://doi.org/10.1002/dad2.12353\u003c/li\u003e\n\u003cli\u003eDilixiati D, Kadier K, Qiao B, et al. Association between serum neurofilament light chain levels and grip strength among US adults: a cross-sectional study using National Health and Nutrition Examination Survey data from 2013 to 2014. \u003cem\u003eBMJ open\u003c/em\u003e. 2025;15(4):e084766. doi:https://doi.org/10.1136/bmjopen-2024-084766\u003c/li\u003e\n\u003cli\u003eHe L, de Souto Barreto P, Giudici KV, et al. Cross-Sectional and Longitudinal Associations Between Plasma Neurodegenerative Biomarkers and Physical Performance Among Community-Dwelling Older Adults. \u003cem\u003eThe Journals of Gerontology: Series A\u003c/em\u003e. 2020;76(10):1874-1881. doi:https://doi.org/10.1093/gerona/glaa284\u003c/li\u003e\n\u003cli\u003eGiancardo L, S\u0026aacute;nchez-Ferro A, Butterworth I, Mendoza CS, Hooker JM. Psychomotor Impairment Detection via Finger Interactions with a Computer Keyboard During Natural Typing. \u003cem\u003eScientific Reports\u003c/em\u003e. 2015/04/16 2015;5(1):9678. doi:https://doi.org/10.1038/srep09678\u003c/li\u003e\n\u003cli\u003eBuijink AWG, Broersma M, van der Stouwe AMM, et al. Rhythmic finger tapping reveals cerebellar dysfunction in essential tremor. \u003cem\u003eParkinsonism \u0026amp; Related Disorders\u003c/em\u003e. 2015/04/01/ 2015;21(4):383-388. doi:https://doi.org/10.1016/j.parkreldis.2015.02.003\u003c/li\u003e\n\u003cli\u003eKoppelmans V, Ruitenberg MFL, Schaefer SY, et al. Delayed and More Variable Unimanual and Bimanual Finger Tapping in Alzheimer\u0026rsquo;s Disease: Associations with Biomarkers and Applications for Classification. \u003cem\u003eJournal of Alzheimer\u0026rsquo;s Disease\u003c/em\u003e. 2023;95(3):1233-1252. doi:https://doi.org/10.3233/JAD-221297\u003c/li\u003e\n\u003cli\u003eSuzumura S, Sugioka J, Sakurai K, et al. Finger motor skills and related brain regions in patients with cognitive disorder. \u003cem\u003eJournal of Alzheimer\u0026rsquo;s Disease\u003c/em\u003e. 2024:13872877251344223. doi:https://doi.org/10.1177/13872877251344223\u003c/li\u003e\n\u003cli\u003eWitt ST, Laird AR, Meyerand ME. Functional neuroimaging correlates of finger-tapping task variations: an ALE meta-analysis. \u003cem\u003eNeuroimage\u003c/em\u003e. 2008;42(1):343-356. doi:https://doi.org/10.1016/j.neuroimage.2008.04.025\u003c/li\u003e\n\u003cli\u003eWard N, Frackowiak R. Age‐related changes in the neural correlates of motor performance. \u003cem\u003eBrain\u003c/em\u003e. 2003;126(4):873-888. doi:https://doi.org/10.1093/brain/awg071\u003c/li\u003e\n\u003cli\u003eJiang R, Westwater ML, Noble S, et al. Associations between grip strength, brain structure, and mental health in \u0026gt; 40,000 participants from the UK Biobank. \u003cem\u003eBMC Medicine\u003c/em\u003e. 2022/09/09 2022;20(1):286. doi:https://doi.org/10.1186/s12916-022-02490-2\u003c/li\u003e\n\u003cli\u003ePrigatano GP, Braga LW, McElvogue M, et al. Motor correlates of finger tapping variability in subjective memory complaints, mild cognitive impairment and probable Alzheimer\u0026apos;s disease. \u003cem\u003eJournal of Alzheimer\u0026rsquo;s Disease\u003c/em\u003e. 2025:13872877241312984. doi:https://doi.org/10.1177/13872877241312984\u003c/li\u003e\n\u003cli\u003eMontero-Odasso M, Pieruccini-Faria F, Ismail Z, et al. CCCDTD5 recommendations on early non cognitive markers of dementia: A Canadian consensus. \u003cem\u003eAlzheimer\u0026apos;s \u0026amp; Dementia: Translational Research \u0026amp; Clinical Interventions\u003c/em\u003e. 2020;6(1):e12068. doi:https://doi.org/10.1002/trc2.12068\u003c/li\u003e\n\u003cli\u003eKumar S, Rani S, Sharma S, Min H. Multimodality Fusion Aspects of Medical Diagnosis: A Comprehensive Review. \u003cem\u003eBioengineering\u003c/em\u003e. 2024;11(12):1233. doi:https://doi.org/10.3390/bioengineering11121233\u003c/li\u003e\n\u003cli\u003eJonell P, Mo\u0026euml;ll B, H\u0026aring;kansson K, et al. Multimodal capture of patient behaviour for improved detection of early dementia: clinical feasibility and preliminary results. \u003cem\u003eFrontiers in Computer Science\u003c/em\u003e. 2021;3:642633. doi:https://doi.org/10.3389/fcomp.2021.642633\u003c/li\u003e\n\u003cli\u003eAlty J, Bai Q, Li R, et al. The TAS Test project: a prospective longitudinal validation of new online motor-cognitive tests to detect preclinical Alzheimer\u0026apos;s disease and estimate 5-year risks of cognitive decline and dementia. \u003cem\u003eBMC Neurol\u003c/em\u003e. Jul 18 2022;22(1):266. doi:https://doi.org/10.1186/s12883-022-02772-5\u003c/li\u003e\n\u003cli\u003ePark JE, Gunasekaran TI, Cho YH, et al. Diagnostic blood biomarkers in Alzheimer\u0026rsquo;s disease. \u003cem\u003eBiomedicines\u003c/em\u003e. 2022;10(1):169. doi:https://doi.org/10.3390/biomedicines10010169\u003c/li\u003e\n\u003cli\u003eBi X-a, Cai R, Wang Y, Liu Y. Effective diagnosis of Alzheimer\u0026rsquo;s disease via multimodal fusion analysis framework. \u003cem\u003eFrontiers in genetics\u003c/em\u003e. 2019;10:976. doi:https://doi.org/10.3389/fgene.2019.00976\u003c/li\u003e\n\u003cli\u003eZhang D, Wang Y, Zhou L, Yuan H, Shen D, Initiative AsDN. Multimodal classification of Alzheimer\u0026apos;s disease and mild cognitive impairment. \u003cem\u003eNeuroimage\u003c/em\u003e. 2011;55(3):856-867. doi:https://doi.org/10.1016/j.neuroimage.2011.01.008\u003c/li\u003e\n\u003cli\u003eRudd KD, Lawler K, Callisaya ML, Alty J. Investigating the associations between upper limb motor function and cognitive impairment: a scoping review. \u003cem\u003eGeroScience\u003c/em\u003e. 2023;45:3449\u0026ndash;3473. doi:https://doi.org/10.1007/s11357-023-00844-z\u003c/li\u003e\n\u003cli\u003eAkamine S, Marutani N, Kanayama D, et al. Renal function is associated with blood neurofilament light chain level in older adults. \u003cem\u003eScientific reports\u003c/em\u003e. 2020;10(1):20350. doi:https://doi.org/10.1038/s41598-020-76990-7\u003c/li\u003e\n\u003cli\u003eWu J, Xiao Z, Wang M, et al. The impact of kidney function on plasma neurofilament light and phospho-tau 181 in a community-based cohort: the Shanghai Aging Study. \u003cem\u003eAlzheimer\u0026apos;s Research \u0026amp; Therapy\u003c/em\u003e. 2024;16(1):32. doi:https://doi.org/10.1186/s13195-024-01401-2\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"University of Tasmania","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Upper Limb Motor Function, MCI, Subjective Cognitive Impairment, Biomarkers, ISLAND, Early Diagnosis, Cognitive Impairment, Finger tapping","lastPublishedDoi":"10.21203/rs.3.rs-7209767/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7209767/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cb\u003eBACKGROUND\u003c/b\u003e\u003c/p\u003e\u003cp\u003eEmerging research suggests hand motor biomarkers help identify dementia, but it is unclear which test is best, or whether combining with a blood-based biomarker may further improve classification. We evaluated combinations of hand motor measures and plasma neurofilament light (NfL) to distinguish dementia, Mild Cognitive Impairment (MCI) and Subjective Cognitive Impairment (SCI) from cognitively-healthy controls (HC).\u003c/p\u003e\u003cp\u003e\u003cb\u003eMETHODS\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThree-hundred and seventeen participants (71 dementia, 105 MCI, 59 SCI, 82 HC) completed key-tapping, finger-to-thumb-tapping, grip strength, and NfL analysis in a cognitive clinic. Age-, sex-, and education-adjusted Receiver-Operating-Characteristic curves measured classification accuracy.\u003c/p\u003e\u003cp\u003e\u003cb\u003eRESULTS\u003c/b\u003e\u003c/p\u003e\u003cp\u003eLower-frequency and higher-variability of key-tapping associated with weaker grip strength and higher NfL levels. Models combining key-tapping and NfL best classified dementia (97%), MCI (97%) and SCI (81%) from HC, outperforming individual measures.\u003c/p\u003e\u003cp\u003e\u003cb\u003eCONCLUSION\u003c/b\u003e\u003c/p\u003e\u003cp\u003eIntegration of a brief hand motor test and NfL could aid with distinguishing cognitive diagnosis groups. Findings support multimodal approaches for early dementia detection.\u003c/p\u003e","manuscriptTitle":"Digitised hand movement and plasma NfL are complementary biomarkers of the dementia continuum","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-28 04:27:14","doi":"10.21203/rs.3.rs-7209767/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e6bf1b32-4ea7-42ed-9507-12e907687b43","owner":[],"postedDate":"July 28th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":52092106,"name":"Cognitive Neuroscience"},{"id":52092107,"name":"Geriatrics \u0026 Gerontology"}],"tags":[],"updatedAt":"2025-07-28T04:27:14+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-28 04:27:14","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7209767","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7209767","identity":"rs-7209767","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}