Cognitive Groove community-based rehabilitation for older adults: study protocol for a multisite parallel randomized controlled trial

Abstract

Background

Promoting and maintaining intrinsic capacity, the composite of an individual’s physical and mental abilities, is central to the World Health Organization’s Integrated Care for Older People (ICOPE) framework. Intrinsic capacity reflects a person’s overall functional ability and resilience. Strengthening intrinsic capacity through community-based, person-centered interventions is key to enabling older adults to live independently and maintain well-being. Dance-based rehabilitation represents a promising, holistic approach that supports multiple domains of intrinsic capacity, including locomotor, cognitive, and psychological function, within a single, engaging program.

Methods

This two-arm, parallel-group, single-blind trial compares Cognitive Groove to wait-listed control. Participants will attend 1-h group classes twice weekly for 3 months at YMCA sites. Eligibility includes a FRAIL Scale score ≥ 2, ability to walk 25 m (with or without assistive devices), and follow two-step instructions. The target sample is 250 older adults (125 per arm). The primary outcome is physical performance, measured by the Short Physical Performance Battery (SPPB), representing locomotor capacity within the ICOPE framework. Secondary outcomes align with other ICOPE domains and include measures of cognitive, psychological, sensory, and vitality capacities. Outcomes will be analyzed using generalized linear model (GLM) analysis to assess between-group differences over time. This protocol follows the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) guidelines.

Discussion

By evaluating Cognitive Groove in real-world settings, the study will generate critical evidence to support the implementation of scalable, non-pharmacological strategies for frailty management in community-dwelling older adults.

Trial registration

This trial is registered on ClinicalTrials.gov (NCT06870149).

Introduction

Background and rationale {9a}

Frailty is a condition characterized by reduced physical strength and function [1]. In Canada, over 1.6 million older adults are currently living with frailty, a number projected to exceed 2 million within the next decade [2] and affecting up to 50% of adults over 85 years old [3]. A major contributor to frailty progression is the cycle of deconditioning, which can result from prolonged hospitalization or immobility at home [1]. Early intervention, particularly among those in pre-frail or early frail stages, offers a critical window to prevent further functional decline and preserve independence [4]. Despite the potentially treatable and reversible nature of frailty, access to specialized geriatric and rehabilitation services remains limited in many communities. There is a pressing need for scalable, evidence-based, non-pharmacological interventions that can be implemented in primary care and community settings.

Physical exercise has consistently been shown to be one of the most effective interventions for addressing frailty [5]. Our meta-analyses of randomized controlled trials (RCTs) have demonstrated that exercise improves mobility (standardized mean difference (SMD) 0.60, 95% CI 0.37 to 0.83) and frailty status (SMD − 1.29, 95% CI − 2.22 to − 0.36) with moderate certainty of evidence [6]. International guidelines strongly recommend multifactorial exercise programs that challenge balance and incorporate functional movements with up to 30% reduction in falls [7, 8]. Exercise programs for fall prevention for community-dwelling older adults should be individualized, conducted at least three times per week, and progressed in intensity for > 12 weeks (Grade 1A: Strong Recommendation) [9]. However, there remains a gap in the availability of evidence-based exercise interventions specifically tailored for those living with frailty. Dance-based interventions are uniquely promising because they are multimodal, simultaneously targeting strength, balance, coordination, and cognitive function—key domains impacted by frailty, which is a significant factor for cognitive impairment and dementia [10]. A systematic review of 11 RCTs (N = 1412) confirmed that dance interventions improve cognition, with significant gains in global cognition (Mini-Mental State Examination [MMSE], MD = 1.58, 95% CI: 0.21–2.95) and executive function (Trail Making Test Part B [TMT-B], MD = − 4.12, 95% CI: − 21.28 to 13.03) [11].

GERAS DANCE is an evidence-based rehabilitation program co-designed by geriatric medicine and rehabilitation clinicians, fitness instructors, and older adults [12–14]. Dance as a rehabilitation approach inherently presents a complex sensorimotor activity, combining both physical and cognitive training. Dance steps are taught in a graded approach designed to increase endurance while targeting large muscle groups involved in functional movement patterns [12]. In a quasi-experimental study involving 106 older adults, 3 months of GERAS DANCE led to meaningful improvements in mobility, balance, and strength, with the greatest gains observed among participants who were frail or had sarcopenia [13]. Notably, improvements in walking speed met or exceeded the minimal clinically important difference (MCID) of 0.1 m/s [13]. These physical benefits were accompanied by high adherence rates (90%) and positive participant feedback, particularly in relation to program enjoyment and opportunities for social connection [13]. Based on participant feedback, the program was rebranded as Cognitive Groove to reflect its emphasis on cognitive engagement, rhythm, and social interaction. While the name highlights these dimensions, Cognitive Groove aims to improve both cognitive and physical function which are core components of intrinsic capacity. The program aligns with the World Health Organization’s Integrated Care for Older People (ICOPE) framework (Fig. 1) [15] targeting multiple domains of intrinsic capacity including locomotor, cognitive, psychosocial, sensory, and vitality domains. It bridges the gap between general healthy aging initiatives and the need for accessible, culturally diverse, and intensive rehabilitation for older adults living with frailty.

Fig. 1.

Intrinsic capacity domains adapted from the World Health Organization’s Integrated Care for Older People (ICOPE) framework

Objectives {10}

We will undertake a randomized controlled trial (RCT) to evaluate the effectiveness of Cognitive Groove, a community-based, multimodal exercise intervention versus a wait-list control in improving intrinsic capacity in community-dwelling older adults living with frailty. Specifically, the proposed RCT will answer (Table 1):

Table 1.

Research questions

Primary	In community-dwelling older adults (≥ 65 years) living with frailty, is Cognitive Groove more effective than wait-list control in improving physical performance as measured by the Short Physical Performance Battery (SPPB) from baseline to 3 months?
Secondary	In older adults (≥ 65 years) living with frailty, is Cognitive Groove more effective than wait-list control in improving intrinsic capacity including locomotor, cognitive, psychosocial, sensory, and vitality domains from baseline to 3 months?
Tertiary	As a community-based rehabilitation intervention, is Cognitive Groove a cost-effective intervention embedded within the community for older adults living with frailty after 9 months?

Methods: patient and public involvement and trial design

Patient and public involvement {11}

Participants were actively involved in the development of the Cognitive Groove intervention based on feedback from the pilot work. Additionally, older adults with lived experience are members of the trial steering committee as part of the dissemination plan.

Trial design {12}

This study is a two-arm, parallel-group, single-blind RCT with 250 older adults (aged 65+) living with frailty, randomly distributed between intervention and control groups. We used the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) guidelines to guide the reporting of our trial protocol [16]. We will utilize the Consolidated Standards of Reporting Trials (CONSORT) checklist to ensure adherence to standardized reporting guidelines and enhance transparency and research quality [17]. Participants will receive 3 months of either Cognitive Groove delivered in 1-h classes 2 times weekly or wait-list control (i.e., continue with their normal activities, with the option to participate in Cognitive Groove after the intervention period). Main group comparisons will be made at 3 months, with blinded assessors conducting clinical assessments at baseline and 3 months. A cost-effectiveness evaluation will be conducted at 9 months. The study design is outlined in Fig. 2.

Fig. 2.

Study flow diagram. Note: Participants will be recruited from primary care, specialists, and the community. Following eligibility screening and baseline assessment, individuals will be randomized (stratified by sex and Short Physical Performance Battery (SPPB score)) to a 3-month Cognitive Groove intervention or wait-listed control group. Post-intervention (3 months) assessments will include SPPB and secondary outcomes. All participants will be followed for 6 months to monitor emergency room visits, hospitalization, and institutionalization. Wait-listed controls will receive the intervention after follow-up

Methods: participants, interventions, and outcomes

Trial setting {13}

Over the course of approximately 18 months, rolling recruitment will form eight cohorts of 32 participants receiving Cognitive Groove (16 per arm) (Table 2).

Table 2.

Characteristics of the people who are needed for the trial

Characteristic	The people we would expect to see included
Age	≥ 65 years
Sex	Male, female
Gender	Man, woman, non-binary/gender-fluid, two-spirit, prefer to self-describe
Race, ethnicity and ancestry	Racial/ethnic identity, place of birth, citizenship status, language(s) spoken
Socioeconomic status	Housing, education, employment status, income
Geographic location	Canada
Other characteristics relevant to the trial	N/A

Demographic and health-related information: We will collect information about each participant’s age, sex, gender, housing, education, smoking habits, health conditions, past surgeries, any history of falls or broken bones, use of mobility aids, and involvement in other activity or social programs. We will also measure height, weight, and blood pressure, and record any medications or supplements being taken. Participants will be asked about any challenges they face with exercise or getting to the YMCA, as well as what helps or motivates them to stay active and healthy.

Eligibility criteria for participants {14a}

Inclusion criteria

• Community-dwelling aged ≥ 65 years

• At high risk for mobility disability/functional limitations as assessed by the FRAIL Scale ≥ 2 [18]

• Able to ambulate independently (25 m) with or without an assistive device

• Able to follow two-step instructions

• Medical clearance from the referring clinician, or for self-referrals, medical clearance from the family physician to safely participate in exercise

• Can arrange transportation to the YMCA up to 2 times per week

Exclusion criteria

• Unable to speak or understand English

• Current regular participation (i.e., 2 or more times per week for the past month) in a structured exercise program

• Receiving active physical therapy services

• Severe cardiac or pulmonary disease

• Unstable angina or heart failure

• Severe osteoarthritis (e.g., awaiting joint replacement)

• Parkinson’s disease or other progressive neurological disorder

• Other medical, psychiatric, behavioral factors that in the judgment of the physician or principal investigator, may interfere with study participation

• Life expectancy is expected to be less than 12 months (i.e., palliative care)

• Travel/commitments requiring missing more than 2 weeks of the intervention

• Previously participated in Cognitive Groove programming

Eligibility criteria for sites and those delivering interventions {14b}

The intervention will be delivered at participating YMCA sites by Certified Cognitive Groove instructors who have been trained by the research staff in study-specific procedures.

Who will take informed consent? {32a}

Interested older adults may self-refer by contacting the Geras research team (phone number provided on advertising materials) or may be referred by their clinician who will take preliminary consent for the participant to be contacted by the research team. Research staff will follow up by phone to confirm eligibility, share study details, and obtain verbal consent. A study information sheet will be emailed or mailed to participants for review before the scheduled consent call.

Additional consent provisions for collection and use of participant data and biological specimens {32b}

Not applicable. Biological specimens will not be collected.

Intervention and comparator

Intervention and comparator description {15a}

Cognitive Groove

Classes follow a consistent sequence of activities (orientation and socialization (10 min); warm-up (5 min); structured dance (30 min); cool-down (5 min); and socialization and review of weekly hand-outs for balance exercises (10 min)). Classes will consist of approximately 16 participants. Instructors will guide participants through seated and standing exercises. After each class, participants will use the Borg Rating of Perceived Exertion (RPE) to self-report their perceived exertion [19]. During the first class, they will receive an intervention-specific study manual that describes the homework exercises for each week. They will also indicate whether they were able to complete any of the standing exercises during each session in their weekly homework completion logs. The curriculum schedule includes five foundations and seven routines. Cognitive Groove foundations emphasize learning the ABCs (Agility, Balance, and Coordination) of movement and increasing participants’ confidence to move their bodies forward and backward and side-to-side with improved speed and rhythmicity. Cognitive Groove routines combine the foundational skills into full choreographed dances to music from the 50 s and 60 s (e.g., rock and roll, jazz, salsa, cha cha cha, rhythm soul, swing, Bollywood and disco). The end of the program curriculum is used to repeat the participants’ favorite class dances. Instructor and end-user feedback is utilized to adapt Cognitive Groove programming to match the functional abilities of the cohort. This proposed program follows the Template for Intervention Description and Replication (TIDieR) for transparent evaluation and replication [20]. Certified Cognitive Groove instructors can actively help monitor progress, set goals, and make informed decisions on frailty management. Rationale/evidence: Cognitive Groove was developed with an activity analysis [12], a systematic evaluation in the field of occupational therapy, to tailor the curriculum with appropriately matched grading (duration, sequence, instructions) and motor complexity increases in agility, balance, and coordination demands. Certified Cognitive Groove instructors undergo specialized 20-h (12 h online portal modules, 8 h in-person) training to learn the curriculum, which uses a balance systems approach to target multiple biomechanical [21] and cognitive functions [22] simultaneously. Intervention fidelity is assessed through in-class tracking logs completed by instructors, which document the Cognitive Groove routine and/or foundation completed during each class. These logs are then reviewed by research staff to evaluate fidelity.

Wait-list control

Participants randomized to the control arm will continue with their wait-list control and daily activities for the duration of the study. They will not receive the Cognitive Groove intervention during the trial period but will be placed on a wait-list and offered full access to the program after completing all study-related assessments. This approach ensures ethical and equitable access to the intervention while preserving the integrity of the controlled evaluation. Brief, monthly check-ins will support participant safety and also help mitigate the risk of differential dropout between groups. Additionally, the structured contact with control participants is intended to minimize the influence of attention bias and the Hawthorne effect—where individuals may modify their behavior simply due to awareness of being observed—thereby helping to isolate the true effects of the intervention [23].

Explanation for the choice of comparator {9b}

Participants in the control group will not take part in Cognitive Groove during the study but will continue with their wait-list control and normal activities. They will have the opportunity to participate in Cognitive Groove after completing their involvement in the study.

Criteria for discontinuing or modifying allocated intervention/comparator {15b}

For participants randomized to either study group, the assigned study intervention may need to be modified or discontinued by trial investigators if the participant experiences harm, which could include injury, hospitalization, decline in health or mobility status which impacts the safety of exercise, or withdrawal of participant consent.

Strategies to improve adherence to intervention/comparator {15c}

The Cognitive Groove RCT design emphasizes the study’s contextual factors and real-world applicability. We designed our intervention in collaboration with older adults and community stakeholders. Almost 90% of participants reported practicing at home “most days/every day” in the pilot work [12]. Adherence to the Cognitive Groove intervention will be recorded by both the dance instructor (i.e., class attendance and physical exertion level) and participants (logbook of home practice). Class attendance will be tracked via YMCA attendance monitoring. Study personnel will conduct monthly check-ins by phone to assist with challenges pertaining to their intervention.

Concomitant care permitted or prohibited during the trial {15d}

Not applicable. No concomitant care is prohibited during the trial beyond the exclusion criteria described in Sect. 14a.

Ancillary and post-trial care {34}

In the case of research-related side effects or injury as a direct result of taking part in this study, participants will be referred for appropriate medical care. There are no other plans for additional health care or compensation.

Outcomes {16}

The primary outcome is physical performance within the locomotor capacity domain, assessed with the SPPB [24]. SPPB total scores range from 0 to 12, and ≤ 8 is indicative of physical frailty. SPPB sub-component scores, which include measures of functional mobility, comprise walking speed, chair rise and standing balance tests (each scored from 0 to 4). The SPPB has demonstrated validity in community-dwelling older adults as a measure of physical frailty and sarcopenia [25–27]. To comprehensively evaluate the effects of the intervention on physical and cognitive function, outcomes were mapped to the WHO ICOPE framework of intrinsic capacity, encompassing locomotor, cognitive, psychosocial, sensory, and vitality domains [15]. The following outcomes will be collected for all participants (the specific measurement variables and collection schedules are listed in Fig. 3):

1. Additional measures of locomotor capacity. Balance will be assessed using the Single Leg Stance Test, recorded as the maximum time (up to 30 s) a participant can maintain single-leg stance; longer times indicate better balance [28]. Endurance will be measured using the 400-m Walk Test, with total completion time recorded; faster times indicate better endurance [29]. Functional mobility will be evaluated using the Timed Up and Go (TUG) test, measured as the total time to stand, walk 3 m, return, and sit; shorter times indicate better mobility [30].

2. Cognitive capacity. Global cognition will be assessed using the Montreal Cognitive Assessment (MoCA). Higher scores (range 0–30) reflect better overall cognitive function [31]. Executive function will be measured with the TMT Part A (TMT-A) and Part B (TMT-B) [32]. This test assesses mental flexibility, task-switching, and processing speed. Longer completion times (in seconds) indicate poorer executive performance. Processing speed and attention will be evaluated using the Digit Symbol Substitution Test (DSST) [33]. Higher scores (range 0–100) represent better cognitive performance in these domains. Dual-task performance will be measured via the TUG-Cognitive Test, where longer times suggest lower performance [34].

3. Psychosocial capacity. Depressive symptoms will be measured using the Patient Health Questionnaire-9 (PHQ-9) [35]. This validated screening tool evaluates the presence and severity of depressive symptoms, with higher total scores (range 0–27) indicating greater levels of depression. Social isolation and loneliness will be assessed using the UCLA Loneliness Scale [36]. This measure evaluates subjective feelings of loneliness and social disconnectedness, with higher scores (range 20–80) reflecting more severe loneliness. Balance confidence will be evaluated using the Activities-specific Balance Confidence (ABC) Scale (higher scores = more confidence) [37], and the Falls Efficacy Scale International (FES-I) (higher scores = greater fear of falling) [38].

4. Sensory capacity. Self-reported hearing and vision will be captured using the ICOPE screening questions (“Do you have difficulty seeing, even if wearing glasses?”/“Do you have difficulty hearing, even if using a hearing aid?”). Responses will be used to describe sensory status at baseline and explore its relationship with other intrinsic capacity domains.

5. Vitality. Muscle strength will be assessed using a handgrip dynamometer, with higher values indicating better strength [39].

6. Multiple domains of intrinsic capacity. Frailty and overall intrinsic capacity will be measured via the Fit-Frailty Index (higher scores = greater frailty) [40, 41]. Community mobility and social engagement will be captured using the Life Space Assessment [42]. This tool assesses the frequency, independence, and extent of movement through different life spaces (e.g., home, neighborhood, community), with higher scores (range 0–120) indicating greater social participation and functional autonomy.

7. Health economics. Healthcare utilization (e.g., hospitalization, emergency department visits, physician visits, and visits to other healthcare professionals) will be collected at baseline, 3 months, 6 months and 9 months using a short economic questionnaire developed for the study. Higher healthcare resource usage will reflect greater healthcare needs. Health-related quality of life will be measured at baseline, 3, 6, and 9 months with the Health Utilities Index (HUI), where higher values indicate better overall health status [43].

8. Fidelity, safety, and program evaluation. Program engagement and adherence will be tracked through class attendance and homework completion records. Higher levels of participation indicate stronger adherence to the intervention and greater engagement with program activities. Falls, fall-related injuries, and other adverse events (AEs) will be tracked through self-reported data and electronic health records (EHRs), providing insight into fall risk and safety outcomes. Participant perspectives will be explored using a post-program experience survey, which gathers qualitative feedback on their motivations, perceived benefits, social experiences, and overall impressions of the program [44].

Fig. 3.

Schedule of enrollment, interventions, and assessments. Note: ABC = Activities-specific Balance Confidence; DSST = Digit Symbol Substitution Test; FES-I = Falls Efficacy Scale International; HUI = Health Utilities Index; MoCA = Montreal Cognitive Assessment; PHQ-9 = Patient Health Questionnaire-9; SPPB = Short Physical Performance Battery; TMT-A/B = Trail Making Test A and B; TUG = Timed Up and Go; TUG-Cognitive = Timed Up and Go – Cognitive; UCLA = University of California, Los Angeles (Loneliness Scale)

Harms {17}

Study personnel will conduct monthly check-ins by phone to monitor for AEs. The participants will also be instructed to contact the study coordinator if they experience any unfavorable signs or symptoms. AEs or harm from any source will be reported to the research team and recorded in a structured form. Any serious adverse events (SAEs) will be reported to the Research Ethics Board within 24 h. The independent Data Safety Monitoring Board (DSMB) will review safety data from the trial and advise the investigators and the Steering Committee on the future management of the trial. AEs will also be collected through Epic EHR software (Epic Systems Corporation, US) at Hamilton Health Sciences (HHS) and St. Joseph’s Healthcare Hamilton, where most participants will be registered as patients.

Participant timeline {18}

The schedule of enrollment, interventions, and assessments is detailed in Fig. 3.

Sample size {19}

Assuming at least a 0.8-point difference at 3-month follow-up (SPPB = 9.3 in the intervention group versus 8.5 in the control group), with a standard deviation of 2.0, an alpha of 0.05 and power of 80%, and factoring a drop-out rate of 20%, we will need a total sample size of 250 individuals (n = 125 per arm).

Recruitment {20}

Based on our protocol timeline, enrolling 250 participants over 18 months necessitates an enrollment rate of approximately 14 older adults with frailty per month. During our pilot studies, we enrolled approximately 20 older adults with frailty from the community each month. We will recruit participants from diverse locations (inner-city vs. suburban) to enhance generalizability and enable a comprehensive evaluation of the intervention’s effectiveness across various settings and populations [45]. This approach mitigates selection bias and ensures a representative sample. Healthcare Providers and Sites: Clinicians in acute care, primary care, and rehabilitation settings will help identify potential participants. With appropriate approvals, these providers may share study information directly or display recruitment materials (e.g., posters, brochures) in their settings. Within the ethical guidelines of our hospital system, recruitment may also involve searching EHRs or contacting individuals who have consented with the hospital system to be approached for research. Community and Self-Referral: Community-based outreach will include partnerships with local organizations (e.g., Alzheimer Society, community centers), distribution of study materials at public locations and events, and outreach via local media. Interested individuals may self-refer by contacting the study team by phone, email, or the Geras Centre website. A brief telephone pre-screen will be conducted to assess interest and eligibility, and participants can choose to receive study information by mail or email. Advertising Platforms: The study will also be promoted through newspapers and our social media platforms.

Assignment of interventions: randomization

Sequence generation: who will generate the sequence {21a} and type of randomization {21b}

An independent statistician will create computer-generated stratified block randomization allocation tables for each cohort with SAS software and upload them into Research Electronic Data Capture (REDCap) tools hosted at McMaster University. REDCap is a secure, web-based software platform designed to support data capture for research studies [46, 47]. Participants will be stratified based on sex and baseline SPPB scores ≤ 8 points, i.e., a proxy for the presence of physical frailty.

Allocation concealment mechanism {22}

To ensure concealment of the sequence, only a researcher not involved in the study will have access to a computer-generated allocation list. This list will be uploaded directly into the REDCap randomization module. Group assignments will be permanent within the participant record and will be trackable, but not modifiable, in the audit log.

Implementation {23}

After obtaining written informed consent and participant screening, the research coordinator will confirm the study participant’s eligibility. To protect against ascertainment bias, baseline measures will be assessed before randomization. Research staff will randomize the participants using a REDCap randomization system, a secure web application for building and managing online surveys and databases.

Assignment of interventions: blinding

Who will be blinded {24a}

Those blinded to the intervention will include the blinded assessor, who will conduct assessments, the investigative team, and members of the steering committee.

How will be blinding be achieved {24b}

Blinded assessors will work off-site and will not meet to discuss participants with study staff. Every effort will be made to reduce bias; unfortunately, complete blinding may not be possible as participants may disclose group allocation to assessors. The study coordinator, local research assistants, and participants will not be blinded.

Procedure for unblinding if needed {24c}

Not applicable. Participants are not blinded to the allocated intervention.

Data collection and management

Plans for assessment and collection of outcomes {25a}

Outcomes will be measured at baseline (within 2 weeks of starting the program) and after 3 months (within 2 weeks of the last intervention/control class). Physical performance and cognitive outcomes will be assessed in-person at the YMCA. All other questionnaires will be administered over the phone by a blinded assessor. Healthcare utilization data will be collected for the 9-month period after baseline (3-month, 6-month, and 9-month timepoints).

Plans to promote participant retention and complete follow-up {25b}

In our clinical pilot study, 5 of 25 individuals discontinued Cognitive Groove; two of these were due to unrelated injuries/death; one was uninterested, and two were due to logistical reasons (e.g., transportation). Our previous studies show a successful retention rate of 75–80%, which is higher than the typical retention rate of 65–75% for exercise programs [39]. We anticipate similar levels of discontinuation for intervention/wait-list control. Further, given that outcome assessments will be conducted on-site at the YMCA, we anticipate a low loss to follow-up assessments (< 5%). Utilizing a participant-centered ecological model of retention, our retention working group will develop an evidence-informed retention protocol to maximize participant reach, engagement, and retention [48–50].

Data management {26} and confidentiality {33}

All data will be collected and managed using the secure, web-based software platform REDCap database hosted at the Geras Centre for Aging Research. The data management plan includes detailed data collection, storage, sharing, and preservation procedures, ensuring compliance with ethical and regulatory requirements. Electronic data will be stored in a secured online research drive under a password-protected network with Hamilton Health Sciences (HHS). All de-identified electronic data will be entered in a REDCap database.

Statistical methods

Statistical methods for primary and secondary outcomes {27a}

The study statistician will analyze the study data and report it using the CONSORT checklist [17]. Our primary analyses will be conducted using the intention-to-treat principle (ITT). A per-protocol analysis will also be performed, including only participants who complete 75% of classes. To examine the influence of covariates on our dependent variables, including adherence, the generalized linear model (GLM) technique, assuming an autoregressive correlation structure, will be used to compare differences between intervention groups. To determine the impact of sex and gender differences, sex (female; male) and gender identity (defined by CIHR as woman, man, both, neither, or somewhere along the gender spectrum) variables will be included in the primary analysis as covariates [51]. We will also examine the influence of locales (YMCA sites) and other covariates on our outcomes. Descriptive data will be presented, and group differences in baseline values will be compared using independent t-tests (continuous variables), Fisher’s exact tests (binary variables), and Poisson regression (count variables). The significance level will be set at 5% (p < 0.05). The current proposal’s secondary outcomes and sensitivity analyses will be considered exploratory rather than confirmatory. For analyses of secondary outcomes and subgroup analyses, the Sequential Holm-Šidák approach will be used to adjust for multiple significance testing [52]. This controls for the overall type I error rate while improving the statistical power of the tests compared to a non-sequential approach.

Who will be included in each analysis {27b}

All randomized participants will be included in the ITT analyses. The per-protocol analysis will include participants who completed 75% of classes.

How missing data will be handled in the analysis {27c}

Our primary analyses will be conducted using the ITT principle. To handle missing data from the ITT perspective, we will conduct multiple imputations using a Markov chain Monte Carlo method that includes 10 imputed datasets [53]. Missing data sensitivity analyses on the imputed datasets with best-case worst-case scenarios will be conducted [53].

Methods for additional analyses (e.g., subgroup analyses) {27d}

Sensitivity analysis will be conducted to evaluate the per-protocol cohort (participants who completed the trial) and the effect of covariates on our dependent variables, including adherence. We will apply a GLM analysis to assess how stratified variables (age and sex) influence the outcome measures. Based on the Instrument to assess the Credibility of Effect Modification Analyses (ICEMAN) criteria for an RCT, we hypothesize that primary outcomes will show greater changes in women compared to men and in older adults compared to younger individuals [54]. Additionally, subgroup analysis will be performed using the Borg RPE, and the proportion of seated and standing participants will be used to measure physical intensity.

Economic analysis: A 9-month trial-based economic evaluation will be conducted to compare the costs and quality-adjusted life years (QALYs) between treatment arms from a public payer perspective. The study conduct and reporting will follow Canadian [55] and international [56–58] guidelines for economic evaluations of healthcare programs. The cost of developing and administering Cognitive Groove will be captured as part of the trial. Healthcare resource utilization (e.g., hospitalization, emergency department visits, physician visits, and visits to other healthcare professionals) collected at baseline, 3 months, 6 months and 9 months will be costed using publicly available costs. Health utility scores derived from the HUI [59], which will also be collected at baseline, 3 months, 6 months, and 9 months, will be weighted by time spent in health states using an area-under-the-curve approach to calculate QALYs. To take into account censored data, stratification variables and potential differences in baseline characteristics between treatment arms (e.g., baseline HUI score), regression-based models (e.g., net benefit regression [60], seemingly unrelated regression [61]) will be used as appropriate. The uncertainty will be summarized with cost-effectiveness acceptability curves (CEACs). The CEAC will show the probability that Cognitive Groove is cost-effective compared to wait-list control at different willingness to pay thresholds (e.g., $50,000 or $100,000 per QALY gained).

Interim analyses {28b}

Not applicable. There will be no interim analyses.

Protocol and statistical analysis plan {5}

The trial protocol and statistical analysis plan (refer to the “Statistical methods” section) are described in this manuscript.

Oversight and monitoring

Composition of the coordinating center and trial steering committee {3d}

The study coordinating center is the Geras Centre for Aging Research. Overseen by the PI and clinical scientists, the central site research team will be responsible for trial management. The study research assistants will be responsible for submitting and maintaining research ethics board documents, liaising with clinical partners for recruitment, enrolling participants, receiving and storing consent forms, monitoring adherence and adverse events (AEs), overseeing data collection and entry, and all other trial day-to-day operations. To provide overall supervision of the trial, the study steering committee will meet every 6 months or more frequently, as needed.

Composition of the data monitoring committee, its role and reporting structure {28a}

The DSMB is an independent multidisciplinary group consisting of a biostatistician and clinicians who, collectively, have experience in the management of older adults who are pre-frail or frail and in the conduct and monitoring of randomized clinical trials. For each DSMB meeting, Open and Closed Reports will be provided. Open Reports, available to all individuals who attend the DSMB meeting, will include data on recruitment and baseline characteristics and pooled data on eligibility violations, completeness of follow-up, and compliance. Closed Reports, available only to those attending the Closed Sessions of the DSMB meeting, will include analyses of AEs and symptom severity and Open Report analyses that are displayed by the intervention group.

Frequency and plans for auditing trial conduct {29}

Not applicable. There are no plans for formal auditing to be conducted.

Protocol amendments {31}

All protocol updates (i.e., changes to eligibility criteria, outcomes, procedures, etc.) will be communicated to relevant parties (i.e., investigators, research teams, participants, etc.) and submitted to the ethics board for review.

Dissemination policy {8}

Our commitment to knowledge mobilization includes (1) engaging stakeholders with lived experience at all stages of the research process and dissemination (older adults with lived experience are part of our steering committee); (2) fostering collaborative partnerships with policymakers, healthcare providers, and community organizations to co-create relevant, meaningful, and applicable knowledge for real-world contexts—our trial is embedded in the YMCA ecosystem, setting the stage for future sustainability; (3) developing dissemination strategies using multiple channels, such as peer-reviewed publications, conference presentations, social media, local geriatric community and regional specialized geriatric service programs to ensure wide-reaching impact and accessibility; (4) capacity-building initiatives with training and skill development on how to support effective knowledge mobilization practices; and (5) evaluation of knowledge mobilization for continuous improvement and impact.

Discussion

Cognitive Groove is a multimodal community-based rehabilitation program involving progressive dance curricula with increasing motor complexity to improve physical and cognitive performance in older adults. Cognitive Groove is aligned with international consensus guidelines that recommend a multimodal approach to reduce falls and frailty in community-dwelling older adults [7]. It is grounded in the WHO ICOPE framework, which emphasizes maintaining and improving intrinsic capacity [15]. The proposed RCT is adequately powered and embedded in community health partnerships, which, if successful, could potentially change practices nationally and internationally. Our trial is important because existing community-based rehabilitation programs for high-risk populations lack high-quality research on efficacy, with most dance intervention trials focusing on healthy older adults or those with Parkinson’s disease or stroke [62, 63]. The landmark Frailty Intervention Trial, which included home-based physiotherapy and an exercise program, was embedded within specialist geriatric and rehabilitation services, and included interdisciplinary weekly case conferences [64]. However, this trial is resource-intensive and not feasible in Canadian community settings, highlighting the need for research on the effectiveness of community-based interventions. Cognitive Groove presents a novel approach to increasing or maintaining independence and quality of life by improving physical performance and mobility using reminiscent music and rhythmic movement in a fun, safe environment. Cognitive Groove helps bridge the gap in clinical care between healthy older adults and those needing more intensive rehabilitation, offering evidence-based options for prevention and rehabilitation in community settings. This is vital due to limited geriatric care resources and long wait-lists for specialized services in Canada. Cognitive Groove provides inclusive programming promoting Equity, Diversity, and Inclusion principles and anti-oppression values, with a curriculum inspired by diverse dance traditions.

Strengths and limitations

This study has several notable strengths. First, it adopts a multidimensional outcome framework that aligns with geriatric principles by evaluating physical, cognitive, psychosocial, and functional domains. This comprehensive approach reflects the complex interplay of factors that influence health in older adults. Second, the intervention is embedded within a real-world, community-based setting (YMCA), enhancing external validity and supporting future scalability and implementation. Third, the study integrates both quantitative and qualitative methodologies, allowing for a richer understanding of clinical effectiveness, participant engagement, and perceived value from the perspective of older adults.

However, there are a few important limitations to consider in this study. Given this is a group-based program, neither participants nor instructors can be blinded to the intervention, which may influence how people respond or behave. While our primary outcome and many other secondary outcomes are objective measures, some outcome measures are self-reported and subject to response bias. Although we are recruiting older adults with early frailty, it may be difficult to reach or recruit older adults who cannot easily access community programs due to health challenges, social isolation, lack of supports, economic and community mobility barriers.

Trial status

Protocol version 5.0; October 28, 2025. Recruitment started on June 10, 2025, and is expected to be completed by December 31, 2026.

Abbreviations

AE

Adverse event

ABC Scale

Activities-specific Balance Confidence Scale

CEAC

Cost-effectiveness acceptability curve

CIHR

Canadian Institutes of Health Research

CONSORT

Consolidated Standards of Reporting Trials

DSMB

Data Safety Monitoring Board

DSST

Digit Symbol Substitution Test

EHR

Electronic health record

FES-I

Falls Efficacy Scale – International

GLM

Generalized linear model

HHS

Hamilton Health Sciences

HUI

Health Utilities Index

ICEMAN

Instrument to assess the Credibility of Effect Modification Analyses

ICOPE

Integrated Care for Older People

ITT

Intention-to-treat

MCID

Minimal clinically important difference

MMSE

Mini-Mental State Examination

MoCA

Montreal Cognitive Assessment

PHQ-9

Patient Health Questionnaire – 9 item

QALY

Quality-adjusted life year

RCT

Randomized controlled trial

RPE

Rating of Perceived Exertion

SAE

Serious adverse event

SMD

Standardized mean difference

SPIRIT

Standard Protocol Items: Recommendations for Interventional Trials

SPPB

Short Physical Performance Battery

TIDieR

Template for Intervention Description and Replication Checklist

TMT

Trail Making Test

TUG

Timed Up and Go Test

TUG-Cognitive

Timed Up and Go – Cognitive

UCLA Loneliness Scale

University of California, Los Angeles Loneliness Scale

WHO

World Health Organization

Funding {7a}

This work will be carried out with the aid of grants from the Canadian Institutes of Health Research (CIHR) Project Grant. Funding No. 196006.

Data availability {6}

No datasets were generated or analyzed during the current study.

Declarations

Ethics approval and consent to participate {30}

This study was approved by the Hamilton Integrated Research Ethics Board (#18231). All participants will provide informed consent prior to enrollment.

Consent for publication

Not applicable.

Competing interests {7b}

PH, CK, GI and AP are co-developers of the Cognitive Groove intervention with McMaster University with a Canadian Copyright # 1156625 and Trademark # 0925640.