Horizontal lines indicate the 95% CIs. The size of the square data marker refers to the proportional weight of each study. The diamond represents the pooled effect size.
Horizontal lines indicate the 95% CIs. The size of the square data marker refers to the proportional weight of each study. The diamond represents the pooled effect size. MMSE indicates Mini-Mental State Examination; TMT-B, Trail Making Test, Part B; and WMS-LM, Wechsler Memory Scale–Logical Memory.
eTable. Database Search Strategy
eFigure 1. Funnel Plot for Attention, Executive Function, Global Cognition, Memory, Processing Speed, and Verbal Fluency
eFigure 2. Summary of RoB Criteria Using Cochrane’s RoB Tool
Customize your JAMA Network experience by selecting one or more topics from the list below.
Salzman T, Sarquis-Adamson Y, Son S, Montero-Odasso M, Fraser S. Associations of Multidomain Interventions With Improvements in Cognition in Mild Cognitive Impairment: A Systematic Review and Meta-analysis. JAMA Netw Open. 2022;5(5):e226744. doi:10.1001/jamanetworkopen.2022.6744
Are multidomain interventions associated with better cognitive outcomes than single interventions in older adults with mild cognitive impairment (MCI)?
This meta-analysis of 28 studies with 2711 participants examined global cognition, attention, executive function, memory, processing speed, and verbal fluency effects. After intervention, significant improvements favoring multidomain interventions were observed in global cognition, executive function, memory, and verbal fluency compared with the single-intervention, active control.
In this study, multidomain interventions were more strongly associated with improving global cognition, memory, executive function, and verbal fluency in older adults with MCI than single interventions, but evidence is needed to determine the optimal length of multidomain interventions.
Older adults with mild cognitive impairment (MCI) have the highest risk of progressing to dementia. Evidence suggests that nonpharmacological, single-domain interventions can prevent or delay progressive declines, but it is unclear whether greater cognitive benefits arise from multidomain interventions.
To determine whether multidomain interventions, composed of 2 or more interventions, are associated with greater improvements in cognition among older adults with MCI than a single intervention on its own.
MEDLINE, Embase, PsycInfo, AgeLine, CINAHL, and Cochrane Central Register of Controlled Trials were systematically searched from database inception to December 20, 2021.
Included studies contained (1) an MCI diagnosis; (2) nonpharmacological, multidomain interventions that were compared with a single active control; (3) older adults aged 65 years and older; and (4) randomized clinical trials.
Data Extraction and Synthesis
Data were screened and extracted by 3 independent reviewers. Following Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines, random-effects meta-analyses were used to calculate effect sizes from the standardized mean difference (SMD) and 95% CIs.
Main Outcomes and Measures
Postintervention cognitive test scores in 7 cognitive domains were compared between single-domain and multidomain groups. Exposure to the intervention was analyzed.
A total of 28 studies published between 2011 and 2021, including 2711 older adults with MCI, reported greater effect sizes in the multidomain group for global cognition (SMD, 0.41; 95% CI, 0.23-0.59; P < .001), executive function (SMD, 0.20; 95% CI, 0.04-0.36; P = .01), memory (SMD, 0.29; 95% CI, 0.14-0.45; P < .001), and verbal fluency (SMD, 0.30; 95% CI, 0.12-0.49; P = .001). The Mini-Mental State Examination (SMD, 0.40; 95% CI, 0.17-0.64; P < .001), category verbal fluency test (SMD, 0.34; 95% CI, 0.13-0.56; P = .002), Trail Making Test–B (SMD, 0.46; 95% CI, 0.13-0.80; P = .007), and Wechsler Memory Scale–Logical Memory I (SMD, 0.47; 95% CI, 0.15-0.80; P < .001) and II (SMD, 0.26; 95% CI, 0.07-0.45; P < .001) favored the multidomain group. Exposure to the intervention varied between studies: the mean (SD) duration was 71.3 (36.0) minutes for 19.8 (14.6) weeks with sessions taking place 2.5 (1.1) times per week, and all interventions lasted less than 1 year.
Conclusions and Relevance
In this study, short-term multidomain interventions (<1 year) were associated with improvements in global cognition, executive function, memory, and verbal fluency compared with single interventions in older adults with MCI.
Mild cognitive impairment (MCI) refers to an intermediate stage between normal aging and dementia in which individuals report cognitive concerns and demonstrate objective cognitive deficits that do not interfere with daily functioning.1 Recent evidence has demonstrated that the prevalence of MCI increases from 6.7% in older adults aged 60 to 64 years to 25.2% in those aged 80 to 84 years.2 Although MCI may be considered a prodromal stage of dementia,2 35% of older adults with MCI revert back to a normal cognitive state.3,4
Nonpharmacological interventions, including cognitive training, can help improve one’s mood and preserve cognitive functions such as memory.5,6 Exercise can also lead to elevated cerebral metabolism and brain-derived neurotrophic factor (BDNF), which support brain plasticity and angiogenesis in the hippocampus.7-9 Multidomain interventions, composed of 2 or more interventions, may have even greater benefits than cognitive training or exercise alone.10-12 The effects of multidomain interventions have been examined in healthy older adults.13-15 However, there is inconclusive evidence to support significantly improved cognitive outcomes following multidomain vs single-domain interventions in MCI.16,17 Additionally, reviews on this topic have primarily focused on memory or are limited to investigations of cognitive and physical training, while interventions such as mindfulness and nutrition are overlooked.17-19
The purpose of this systematic review and meta-analysis is to compare cognitive outcomes immediately following nonpharmacological single-domain and multidomain interventions in older adults with MCI. This article (1) examines whether multidomain interventions are associated with greater improvements in cognition than single-domain interventions; (2) evaluates which cognitive domains are associated with improvements following multidomain interventions; and (3) assesses length of exposure to the multidomain intervention that is associated with positive outcomes in MCI. Understanding the benefits of multidomain interventions will help design more robust intervention strategies that preserve cognition or delay the onset of decline in older adults with MCI.
This review has been registered in the Prospective Register of Systematic Reviews (CRD42019126899). Results followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting guideline.20
Inclusion and exclusion criteria were outlined using the PICO (population, intervention, comparison, and outcome) model.21 Eligible studies were randomized clinical trials that examined older adults aged 65 years or older who have been diagnosed with MCI using Petersen criteria and confirmation from clinicians who assessed subjective cognitive concerns, objective deficits compared with age-related norms, no functional deficits related to cognition, and the absence of dementia.1 Interventions were considered multidomain if they included 2 or more nonpharmacological components (eg, cognitive and aerobic exercise) that were completed simultaneously or sequentially.7 Additionally, studies had to compare the multidomain intervention with an active control (ie, a single intervention). Systematic reviews, studies without full texts, interventions with a pharmacological component, and studies comparing multidomain interventions with inactive controls (ie, waiting list controls) were excluded. Additionally, studies whose primary aim was to examine the effects of multidomain interventions on MCI and a cooccurring condition that can affect cognition (eg, MCI and a neurological disorder) were excluded.
The literature search was conducted in December 2021 in MEDLINE, Embase, AgeLine, PsycInfo, CINAHL, and Cochrane Central Register of Controlled Trials (CENTRAL) with no restrictions on publication date or language. The search strategy combined medical subject heading terms and keywords encompassing 4 key concepts: cognitive status, age, intervention, and study design (eTable in the Supplement).
Studies were independently screened by 3 of us (S.F., S.S., and T.S.) in 2 steps: (1) title and abstract screening and (2) full-text screening. Relevant abstracts were retained for the full-text review based on the inclusion and exclusion criteria. Study and participant characteristics and outcome measures were extracted from each study by 3 of us (S.F., S.S., and T.S.). During the screening and extraction stages, disagreements were resolved by consensus between authors.
Risk of bias (RoB) was assessed by 3 of us (S.F., S.S., and T.S.) using the Cochrane RoB tool for randomized controlled studies that evaluated selection, performance, detection, attrition, reporting, and other sources of bias.22 Each criterion was classified according to a low, high, or unclear risk of bias. Disagreements within these categories were resolved by consensus between authors. The Egger test of funnel plot asymmetry was used to visualize publication bias.23
Cognitive outcomes were analyzed by cognitive domain (eg, executive function), and meta-analyses were conducted when there were at least 3 studies for a given outcome. A random-effects model was used to account for methodological differences and between-study variance. Review Manager version 5.4 was used to calculate Hedge g and pooled effect size estimates.24 Hedge g was derived directly from the published data, and means and SDs were requested from the corresponding authors when they were not reported.25 Differences between single-domain and multidomain interventions were quantified as the change between scores at baseline and immediately after the intervention. The standardized mean difference (SMD) was calculated by pooling SDs for pre-post change scores and pooling posttest SDs for posttest scores.26 The primary outcome was reported when studies used multiple measures for the same cognitive domain, and the comparison of the marginal analyses was used to assess factorial trials.27 Mean scores were multiplied by −1 to ensure higher scores favored improvements in the multidomain group.28 Effect sizes were then interpreted using small (d = 0.20), medium (d = 0.50), and large (d = 0.80) effect size categories.29 For all analyses, a 2-tailed P < .05 was considered statistically significant. Exposure to the intervention was assessed by the number of sessions, duration, and frequency.
The I2 statistic was used to evaluate potential sources of heterogeneity and was classified into small (≤25%), medium (26-74%), and large (≥75%) groups.30 Subgroup analyses were conducted when there was medium or large heterogeneity across studies. When sufficient studies were available, subgroups were created based on 5 categories: (1) recruitment source (ie, community, clinic-based); (2) multidomain intervention type (eg, cognitive-physical); (3) single-intervention type (ie, a single component of the multidomain intervention or an alternate intervention); (4) multidomain intervention style (ie, group, individual); and (5) order of the multidomain intervention components (ie, sequential, simultaneous). Additionally, specific cognitive tests (eg, Mini-Mental State Examination [MMSE]) were analyzed within each cognitive domain.
Database searches in MEDLINE, Embase, PsycInfo, CINAHL, AgeLine, and CENTRAL yielded 5206 studies that were managed in Covidence. Titles and abstracts were screened following the removal of 1477 duplicates, and 83 studies remained for the full-text review. Fifty-five studies were then excluded for having an abstract only, wrong patient population, no multidomain intervention, wrong comparator, wrong outcome, or wrong study design (Figure 1). Finally, 28 studies, with 2711 participants, were retained for the meta-analyses.
The publication period ranged from 2011 to 2021, as summarized in Table 1.31-58 Seventeen studies recruited participants from the community31,32,36,37,39,40,48-58 whereas 11 recruited from memory clinics.33-35,38,41-47 Total sample sizes ranged from 22 to 555, with a mean (SD) age of 71.6 (3.4) years. MCI was diagnosed using a combination of Petersen criteria, MMSE, Montreal Cognitive Assessment (MoCA), clinical dementia rating (CDR) scores, and evaluations by physicians.
Multidomain interventions included cognitive and physical components35-37,39,41-45,48,52; multiple exercise33,40,49,50,55,56 or cognitive components34; nutritional supplements33,47; mind-body38,46,54; education57,58; cognitive, physical, and social components31; combined cognitive training with transcranial direct current stimulation (tDCS)51; and exercise with music (Table 1).53 Eighteen studies conducted the intervention components sequentially,31,33,34,36-38,40-44,46-49,54,57,58 whereas 10 studies conducted them simultaneously.32,35,39,45,50-53,55,56 Interventions were also completed in a group setting in 19 studies31,33-36,38-40,45,47-50,52-56,58 compared with the remaining 9, which were conducted individually.32,37,41-44,46,51,57 Lastly, the active control in 21 studies contained 1 component of the multidomain intervention,32,34-48,51,53,54,57,58 whereas 7 studies used alternative interventions.31,33,49,50,52,55,56 Cognitive outcomes converged under 7 domains: attention, executive function, global cognition, memory, processing speed, verbal fluency, and other (eg, reaction time, visuospatial) (Table 2).
Twenty studies evaluated global cognition, of which 12 reported improvements34,39,40,42-44,49,52,54-56,58 and 7 found no differences in the multidomain intervention compared with the control (Table 2).31,33,45,46,48,53,57 One study reported increased global cognition scores in the multidomain group, but the changes were smaller than the control.37 The pooled effect size favored the multidomain intervention (SMD, 0.41; 95% CI, 0.23-0.59; P < .001) such that participants in the multidomain group had improved global cognition immediately after the intervention compared with those in the single-domain control (Figure 2A). A medium level of heterogeneity (I2 = 62%) was present, leading to stratification of the sample by recruitment source, order of multidomain intervention components, multidomain and control group intervention styles, and multidomain intervention type, but there were no differences between the subgroups.
Five studies measured cognition using the Alzheimer Disease Assessment Scale–Cognitive Subscale (ADAS-Cog), but cognitive outcomes were not significantly different between multidomain and control groups. Thirteen studies using the MMSE demonstrated that the overall pooled effect size favored the multidomain intervention (SMD, 0.40; 95% CI, 0.17-0.64; P < .001) such that there was a greater increase in MMSE scores after the intervention in the multidomain group compared with the control group (Figure 3A).31,33,34,43-45,48,49,52,54-57 Medium heterogeneity (I2 = 72%) was present, but stratifying the studies by control group intervention type, recruitment source, and order of the multidomain intervention components found that subgroups were not significantly different.
Twenty studies evaluated executive function, but 3 studies38,48,49 were omitted from analyses due to missing data.31-37,39-41,44,45,47,52,56-58 Five studies found positive associations44,45,49,57,58 compared with 1 that reported a smaller effect size in the multidomain group compared with the control group.37 The remaining studies found no differences between intervention and control groups.31-36,38-41,47,48,52,56 Overall, the pooled effect size favored the multidomain intervention (SMD, 0.20; 95% CI, 0.04-0.36; P = .01) (Figure 2B). Medium heterogeneity (I2 = 30%) was present, leading to stratification of the sample by recruitment source, order of multidomain intervention components, multidomain and control group intervention styles, and multidomain intervention type, but there were no differences between the subgroups.
By examining specific tests of executive function, 7 studies measured the Trail Making Test Part A31,34,39,41,44,45,57 and 4 studies measured the Stroop test,35,36,44,56 but there were no differences between groups. There was, however, a medium effect size across 6 studies that demonstrated significant improvements on the Trail Making Test Part B in the multidomain group (SMD, 0.46; 95% CI, 0.13-0.80; P = .007) with medium heterogeneity (I2 = 43%). The small number of studies prevented further subgroup analyses (Figure 3B).31,33,34,39,45,57
Seventeen studies evaluated memory, whereby 15 were included31,33,34,37,39,40,42,44,47,48,51,52,55,56,58 in the meta-analyses because of incomplete data in 2 studies.38,49 Seven studies reported no differences,34,37,38,40,47,48,51 1 reported greater improvements in the control group,33 and 9 reported greater changes in the multidomain group.31,39,42,44,49,52,55,56,58 This is supported by the pooled effect size; greater improvements in memory were observed immediately after the intervention in the multidomain group compared with the control group (SMD, 0.29; 95% CI, 0.14-0.45; P < .001) (Figure 2C). Given medium heterogeneity (I2 = 32%), studies were stratified according to recruitment source, multidomain and single intervention type, multidomain intervention style, and intervention order, but there were no differences between subgroups.
The Wechsler Memory Scale Logical Memory I (WMS-LM I; SMD, 0.47; 95% CI, 0.15-0.80; P < .001) and II (WMS-LM II; SMD, 0.26; 95% CI, 0.07-0.45; P < .001) favored greater improvements in memory in the multidomain intervention group compared with the control group (Figure 3C and D). An insufficient number of studies prevented subgroup analyses.
Nine studies evaluated verbal fluency, whereby 1 study was excluded from analyses due to missing data.49 Five studies found a positive association in the multidomain intervention group33,39,42,49,52 compared with 3 that revealed no differences40,48,56 and 1 study that found greater changes in the control group.37 The overall effect size favored improvements in the multidomain intervention group (SMD, 0.30; 95% CI, 0.12-0.49; P = .001) but with medium heterogeneity (I2 = 30%) (Figure 2D). The studies were further subdivided based on the order of the multidomain invention components and the multidomain and single intervention type, but there were no differences between subgroups. In addition, 6 studies used category verbal fluency tests, which favored the multidomain group (SMD, 0.34; 95% CI, 0.13-0.56; P = .002) (Figure 3E). A medium level of heterogeneity was identified (I2 = 38%), but there were no differences based on the order of interventions and single intervention type.
Six studies measured attention, but there were no differences between the multidomain intervention and control group in 5 studies38,40,47,49,51 compared with 1 that found improvements in the multidomain group.44 The overall effect size was, therefore, not significantly different between groups (SMD, 0.13; 95% CI, −0.15 to 0.41; P = .36).
Ten studies evaluated processing speed, whereby 2 found improvements in the multidomain group44,45 and the remaining studies did not find differences between groups.31,35-38,51,56,57 Therefore, there were no discernable improvements after the intervention (SMD, 0.46; 95% CI, −0.04 to 0.96, P = .07).
Eight studies targeted other cognitive domains, such as visuospatial, reaction time, and sensorimotor tests.32,33,36,40,42,47,49,50 Differences favored the multidomain group on the Wechsler Adult Intelligence Scale arithmetic, block design, and picture arrangement subtests, and MoCA abstraction and orientation subtests (Table 2).
The multidomain interventions varied in terms of session duration (30-135 minutes), frequency (1-7 times per week), and length (8-52 weeks) (Table 2). The mean (SD) duration was 71.3 (36.0) minutes for 19.8 (14.6) weeks with sessions taking place 2.5 (1.1) times per week, and all interventions lasted less than a year. There were no adverse events related to the multidomain intervention, and adherence remained greater than 80% across all studies, with no differences between single and multidomain groups.
The Egger test revealed that attention, executive function, global cognition, memory, processing speed, and verbal fluency funnel plots were symmetrical, indicating minimal risk of publication bias (eFigure 1 in the Supplement). A summary of all RoB criteria using Cochrane’s RoB tool is presented in eFigure 2 in the Supplement. In particular, the blinding of participants and personnel proved to be the most variable, whereby 9 studies presented a low risk of bias,31-33,37,39,48,51,52,55 9 studies demonstrated a high risk of bias,34,41-44,46,49,54,58 and 10 studies were unclear.35,36,38,40,45,47,50,53,56,57 However, this factor is unlikely to have affected the outcomes of this analysis.
The purpose of this systematic review and meta-analysis was to evaluate whether multidomain interventions were associated with greater improvements in cognition than single interventions in older adults with MCI. Findings from this review predominantly featured multidomain cognitive-physical interventions, which reflects previously reported interventions in the literature.10,17 However, nutrition, mind-body, music, and social interventions also contributed to small-medium effect sizes in global cognition, executive function, memory, and verbal fluency immediately after the intervention. Attention and processing speed did not differ between intervention groups, and an insufficient number of studies prevented pooling other cognitive outcomes, such as reaction time and visuospatial abilities.
Because of medium heterogeneity, subgroup analyses were conducted, but differences were not found across recruitment source, multidomain intervention type, single intervention type, multidomain intervention style, and order of the multidomain intervention components, which may be because of the limited number of studies in each subgroup. Nonetheless, these differences should be considered when designing new interventions, as they contributed to methodological variability. For example, diagnosis of MCI may lack uniformity among participants recruited from the community compared with a memory clinic.59 In addition, patients from a clinic may be at a greater risk of progressing to dementia.60 There is also mixed evidence on the benefits of sequential vs simultaneous cognitive-physical training interventions.10,16,61
A second analysis was conducted to determine whether specific cognitive tests acted as a potential source of heterogeneity between studies. Findings revealed improved scores on the MMSE in the global cognition domain, TMT-B in the executive function domain, category verbal fluency scores, and WMS-LM I and II in the memory domain that significantly favored the multidomain intervention group compared with the single-intervention control group. Although the effect sizes for these findings were small to medium, cognitive changes at the predementia stage could be of clinical relevance and important for public health. In all cases, effect sizes in global cognition, executive function, memory, and verbal fluency scores favored the multidomain group. Therefore, changes occurring immediately after the intervention may be used to assess the effectiveness of multidomain training on global cognition and specific cognitive domains and has the potential to inform MCI prognosis. Previous meta-analyses examining the association of cognitive or physical training alone with cognition among patient with MCI have found small to medium effect sizes favoring the intervention group.5,62 Similar findings were reported in global cognition, memory, and executive function following multidomain interventions.16,18,19 The present meta-analysis further contributes to the existing literature by demonstrating similar interactions that are not limited to cognitive and physical interventions. Findings from this study, however, must be interpreted with an understanding that effect sizes do not indicate clinical significance on tests such as the MMSE, which have limited sensitivity to detect meaningful changes in MCI. Rather, effect sizes can help with the interpretation of changes occurring in different cognitive domains.
Compared with single interventions, synergistic benefits arising from multidomain interventions align with the multifactorial nature of MCI. It is important, however, to consider the reproducibility, scalability, and adherence to such interventions. Determining the optimal exposure to a multidomain intervention is uniquely challenging in that there is no gold standard intervention combination and the association of each intervention with improvements in cognition is best understood by comparing trials with multiple arms. While most studies compared 1 intervention component to the multidomain intervention, a study comparing both arms reported decreased cognitive test scores in the multidomain group, which contradicts the outcomes of the present meta-analysis.37 This may be attributed to stress and the demands of certain multidomain intervention combinations compared with others, which may lead to decreased benefits. More evidence is needed from multigroup trials to better understand the balance between demands and benefits of multidomain interventions.
All interventions were short term and lasted less than a year, but the number of sessions, duration, and length of intervention proved to be variable between studies. The rationale for choosing a specific intervention length was often not discussed but was likely derived from the standards for single interventions. Previous reports in the literature focusing on combined cognitive-physical interventions have suggested interventions ranging between 1 to 3 hours per week for at least 16 weeks and as long as 6 months, which is consistent with the studies presented in this analysis.13,62 The effectiveness and length of exposure to more novel interventions, such as mind-body or tDCS, have not been examined to the same extent as cognitive-physical interventions.31,47,53 Future studies should continue investigating these interventions, as improved cognitive outcomes favored the multidomain group.
Studies reported high adherence to the multidomain intervention, which was similar between multidomain and active control groups. This contradicts previous reports suggesting that only one-third of older adults adhere to recommendations of single interventions, such as physical activity.63 Adherence to multidomain interventions was also found in a longer-term, 2-year trial that reported improved scores on a comprehensive neuropsychological test battery.64 Therefore, multidomain interventions remain engaging and should be contrasted with single domain interventions to examine differences in adherence during longer-term trials. Finally, there were no reports of adverse events related to the multidomain interventions, indicating their safety for older adults with MCI.
This study has limitations need to be acknowledged. First, distinct cognitive outcomes were assessed in studies from the same research groups, but bias may be introduced if participants were corecruited for these studies.41-44,55,56 To minimize reporting bias, data extraction and study outcomes were discussed among 3 independent reviewers. Additionally, the selected studies lacked sufficient data to account for different MCI subtypes (eg, amnestic MCI). Variability among different types of multidomain intervention restricted direct comparisons between novel interventions, such as music and nutrition, and more commonly reported cognitive and physical interventions. Additionally, the control groups were composed of either a single component of the multidomain intervention or an alternative intervention, such as a health education program. In both cases, however, improvements in cognition remained significant in the multidomain group compared with the single intervention control, demonstrating the advantage of combining interventions in MCI.
In this study, nonpharmacological, multidomain interventions mainly focused on physical exercise, cognitive training, mind-body, music, dietary supplements, social engagement, and education were associated with small to medium effect sizes indicating improvements in global cognition, executive function, memory, and verbal fluency. A synergistic association was found, suggesting combined interventions may be superior to single interventions to improve cognitive functioning in older adults with MCI. Furthermore, the MMSE, category fluency, TMT-B and WMS-LM scores improved in the multidomain group compared with the active control. Future studies should consider examining interventions other than combined cognitive-physical training, such as mindfulness and nutrition, as promising outcomes were found across a variety of novel interventions. More evidence is needed to determine the optimal exposure to the multidomain intervention and whether improvements are sustained longitudinally.
Accepted for Publication: January 20, 2022.
Published: May 3, 2022. doi:10.1001/jamanetworkopen.2022.6744
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2022 Salzman T et al. JAMA Network Open.
Corresponding Author: Sarah Fraser, PhD, Interdisciplinary School of Health Sciences, University of Ottawa, 25 University Private, Ottawa, ON K1N 7K4, Canada (Sarah.Fraser@uottawa.ca).
Author Contributions: Dr Fraser had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs Montero-Odasso and Fraser are both co–senior authors in this article.
Concept and design: Montero-Odasso, Fraser.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Salzman, Fraser.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Salzman.
Obtained funding: Montero-Odasso.
Administrative, technical, or material support: Sarquis-Adamson, Son, Montero-Odasso.
Supervision: Montero-Odasso, Fraser.
Conflict of Interest Disclosures: None reported.
Funding/Support: Dr Montero-Odasso’s program in Gait and Brain Health is supported by grants from the Canadian Institute of Health Research (MOP 211220; PJT 153100), the Ontario Ministry of Research and Innovation (ER11–08–101), the Ontario Neurodegenerative Diseases Research Initiative (OBI 34739), the Canadian Consortium on Neurodegeneration in Aging (FRN CNA 137794), and Department of Medicine Program of Experimental Medicine Research Award (POEM 768915), University of Western Ontario. He is also the first recipient of the Schulich Clinician-Scientist Award.
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.