Short-term effects of a virtual, community-based, task-oriented group exercise programme incorporating a healthcare–community partnership compared to a waitlist control on increasing everyday function among adults with mobility limitations: protocol for the TIME™ at Home randomised controlled trial

Abstract

Introduction

While group, task-oriented, community-based exercise programs (CBEPs) delivered in-person can increase exercise and social participation in people with mobility limitations, challenges with transportation, cost and human resources, threaten sustainability. A virtual delivery model may help overcome challenges with accessing and delivering in-person CBEPs. The study objective is to estimate the short-term effect of an 8-week, virtual, group, task-oriented CBEP called TIME™ (Together in Movement and Exercise) at Home compared with a waitlist control on improving everyday function in community-dwelling adults with mobility limitations.

Methods and analysis

A randomised controlled trial incorporating a type 1 effectiveness-implementation hybrid design is being conducted in four Canadian metropolitan centres. We aim to stratify 200 adults with self-reported mobility limitations by site, participation alone or with a partner, and functional mobility level, and randomise them using REDCap software to either TIME™ at Home or a waitlist control group. During TIME™ at Home classes (2 classes/week, 1.5 hours/class), two trained facilitators stream a 1-hour exercise video and facilitate social interaction prevideo and postvideo using Zoom. A registered healthcare professional at each site completes three e-visits to monitor and support implementation. Masked evaluators with physical therapy training evaluate participants and their caregivers at 0, 2 and 5 months using Zoom. The primary outcome is the change in everyday function from 0 to 2 months, measured using the physical scale of the Subjective Index of Physical and Social Outcome. The study is powered to detect an effect size of 0.4, given α=0.05, power=80% and a 15% attrition rate. Secondary outcomes are mobility, well-being, reliance on walking aids, caregiver assistance, caregiver mood, caregiver confidence in care-recipient balance and cost-effectiveness. A multimethod process evaluation is proposed to increase understanding of implementation fidelity, mechanisms of effect and contextual factors influencing the complex intervention. Qualitative data collection immediately postintervention involves interviewing approximately 16 participants and 4 caregivers from the experimental group, and 8 participants and 4 caregivers from the waitlist control group, and all healthcare professionals, and conducting focus groups with all facilitators to explore experiences during the intervention period. A directed content analysis will be undertaken to help explain the quantitative results.

Ethics and dissemination

TIME™ at Home has received ethics approval at all sites. Participants provide verbal informed consent. A data safety monitoring board is monitoring adverse events. We will disseminate findings through lay summaries, conference presentations, reports and journal articles.

Trial registration number

NCT06245135.

Strengths and limitations of this study.

• Virtual delivery of a community-based, task-oriented group exercise programme incorporating a healthcare–community partnership for people with mobility limitations addresses barriers to attending in-person programmes.

• A multimethod process evaluation will increase understanding of implementation fidelity, mechanisms of effect and contextual factors influencing the complex intervention.

• Equitable access to research participation is promoted through use of recruitment strategies to target underrepresented groups, and provision of mobile devices, data plans and accommodations for evaluations, if required.

• Conducting the study across four metropolitan sites, using virtual methods of data collection and programme delivery, and strategies to promote equitable access, is expected to optimise participant diversity and external validity of study findings; however, compared with urban settings, participation from rural areas may be lower.

• Participants cannot be masked to the intervention they receive and may participate in alternate exercise programmes which may introduce bias.

Introduction

Background and rationale

Mobility limitations are highly prevalent,¹,³ limit everyday functioning³ and increase the need for caregiver assistance⁴ in people with chronic health conditions. Mobility is defined as changing and maintaining body position; carrying, moving and handling objects; walking and moving; and moving around using transportation.⁵ Evidence from non-randomised⁶,¹⁰ and randomised studies¹¹,¹³ suggests that group, community-based exercise programmes (CBEPs), incorporating task-oriented training of mobility, have the potential to improve balance, balance self-efficacy, leg strength, walking capacity and everyday functioning, as well as cognition, in people with mobility limitations.¹⁴ CBEPs are also associated with caregiver benefits: observing the care recipient participate in a meaningful activity may improve caregiver mental health, and improvement in everyday mobility and activities of daily living (ADLs) of care recipients through exercise may reduce the level of caregiver assistance needed.⁹ CBEPs that include a healthcare–community partnership (CBEP-HCP)¹⁰ foster programme safety and quality and can leverage the infrastructure and expertise of healthcare and community organisations with complementary mandates to support exercise participation.^{10 15} Healthcare partners are healthcare professionals who support CBEPs in various roles, such as by providing training, facilitating referral, prescribing exercise and monitoring programme safety and fidelity.^{16 17} Supported by this evidence base, CBEP-HCPs have emerged in Canada¹⁸,²⁰ and in other countries.^{7 8 11 13}

There are challenges, however, to accessing and delivering in-person CBEP-HCPs. Exercise participants face issues with availability and cost of transportation,^{15 21} inclement weather,²² inadequate building access given their mobility limitations,²³ programme cost,¹⁵ risk of infection²⁴ and pandemic-related programme closures.²⁴ The COVID-19 pandemic particularly affected older adults with mobility limitations who experienced a reduction in meaningful social contact and mental health, increased sedentary behaviour, and decline in mobility and participation in life roles.^{24 25} Caregivers commonly provide transportation to CBEPs, which can take time away from paid employment and other daily activities.^{9 21} Community centre managers identify challenges with operating the programme at a financial loss due to the need for instructors with specialised skills and a high level of exercise supervision.^{15 26} Healthcare managers have to consider the time, cost, staff availability and impact on clinical care of providing a healthcare partner to train fitness instructors and travel to complete programme visits at community centres.^{26 27}

Virtual delivery of CBEP-HCPs may help to overcome participant and provider challenges. With virtual CBEPs, people can exercise in their own home, and the programme can be delivered with fewer human, space and equipment resources than in-person programmes. In an ongoing scoping review²⁸ characterising 32 studies²⁹,⁶⁰ of virtual exercise programmes involving populations commonly experiencing mobility limitations, most studies employed designs, including a quasi-experimental or pilot randomised controlled trial (RCT) design (ie, single group pre–post study,³⁰,3436 39 41 44 46 47 51 53 55 56 59 60 non-RCT,^{37 45} dose-escalation feasibility study,⁴² pilot RCT^{3843 48},⁵⁰) that lack adequate control for confounding or statistical power.⁶¹ Only seven studies employed an RCT design.^{29 35 40 52 54 57 58} The most common study populations were people with cancer³⁰³²,34 38 43 44 55 and neurological conditions^{3135 36 39},42 46 48 50 52; however, none of the studies targeted adults with mobility limitations due to any health condition. Of the 21 studies reporting, the majority of exercise programmes were delivered in a group format^{29 31 32 34 37 41 47 50 52 54 55 57 58} which can help foster social support.⁹ Exercise interventions most commonly involved strength training,³⁰³²,^{35 38 41} aerobic training,³⁰³²,35 38 43 45 47 48 51 53 flexibility training³⁰³²,34 38 40 41 43 45 47 53 55 59 or balance training.^{34 35 38 41 43 47 49 53 55 58 59} Only one study³⁶ evaluated task-oriented training of functional everyday mobility tasks shown feasible to implement in community settings.¹⁶ Most often, exercise programmes were offered two2627 30 36 37 41 45,47 50 54 56 59 or three2933 34 40 42,44 48 49 52 53 58 times per week for 1 hour per class^{2629 30 34},^{37 39} for 8 weeks.2930 33 34 37 42 47,^{49 52 58} Two-thirds of programmes were synchronous (ie, exercises delivered in real-time)^{2930 34 36},^{50 52} and all programmes but one were delivered by healthcare professionals, specialised fitness/yoga instructors, or research staff, which can challenge community implementation due to the cost and challenge associated with hiring, scheduling and retaining exercise instructors with specialised skills.²⁷ Among studies using Zoom (an online videoconferencing platform),2931 34 36 38 40 41 43,^{47 50 52} many^{31 34 36 38 41 43} reported that Zoom software was feasible and acceptable for participants. Study participants were predominantly female/woman, white/Caucasian, with high education levels; studies generally lacked detailed reporting of characteristics (eg, gender, race/ethnicity, language) important to identify potential inequities and generalisability of study results.⁶² These findings suggest that efforts to increase access to underrepresented groups in evaluations of virtual CBEPs are needed. Study outcomes commonly included health-related quality of life,²⁹,3537 39 40 46 51 balance,^{29 32 34 40 41 47 50 55 58 60} cardiorespiratory fitness,^{30 31 33 34 40 42 47 55 56} physical function,^{29 37 43 46 49 55 57 59} muscle strength^{32 40 41 45 47 55} and anxiety.^{30 32 33 41 46} Everyday functioning and mood require attention given the extent to which older adults experienced social isolation during the COVID-19 pandemic.²⁴ Reduced capacity for ADL can contribute to caregiver burden and limit participation in meaningful activities.⁶³ Thus, it is important to understand the potential for virtual exercise programmes to reduce the level of caregiver assistance needed for everyday activities and improve caregiver mood.

We recently developed a virtual, group-based, task-oriented CBEP-HCP for adults with mobility limitations during which people exercise at home. The programme uses a set of exercise videos called TIME™ (Together in Movement and Exercise) at Home. The CBEP-HCP model has innovative features that address challenges associated with accessing and implementing in-person CBEPs. The 1-hour TIME™ at Home videos show therapists demonstrating everyday mobility tasks using materials (eg, step, chair) commonly found at home. Two live facilitators, who do not require exercise expertise, stream the videos using Zoom and facilitate prevideo and postvideo social interaction. This approach provides a feasible and low-cost alternative to hiring and scheduling exercise instructors to deliver the programme. A healthcare partner completes three e-visits to foster programme safety and quality and serves as a resource for facilitators and a referral source for the programme.

We conducted a series of feasibility studies to develop and evaluate select programme components, including a preprogramme safety video, the TIME™ at Home videos, and use of trained facilitators to stream the exercise videos via video-conferencing, and facilitate a postvideo social time in a drop-in style class offered twice a week.⁶⁴ We then evaluated these components in a group-based TIME™ at Home program model in a pre–post feasibility study⁶⁵ (ClinicalTrials.gov, NCT05369741, registered on 11 May 2022). The intervention included the preprogramme safety video, the TIME™ at Home videos, and use of trained facilitators to stream the exercise videos via video-conferencing and facilitate a postvideo social time twice a week for 8 weeks for a group of 10 participants. Twenty older adults with mobility limitations and 11 caregivers participated. Findings supported programme safety (no serious adverse events), feasibility (consistent intervention fidelity, mean attendance >75% benchmark) and potential benefit of the programme based on significant improvement observed on measures of everyday function, leg strength, mobility, balance self-efficacy, mood and perceived health status, consistent with qualitative findings. While caregiver mood improved, the extent of change was not significantly different. Because 35% of participants reported participating in a cointervention, such as physical therapy or an exercise programme, during the intervention period, it is unclear to what extent participation in the TIME™ at Home program contributed to the benefits observed.

A rigorous evaluation is needed to provide evidence of the efficacy of virtual CBEP-HCPs on improving everyday function and caregiver mood and assistance in adults with mobility limitations. Additionally, with complex interventions⁶⁶ like the TIME™ at Home program, which have multiple components and outcomes, there is a need to conduct a process evaluation⁶⁷ to better understand implementation, mechanisms of impact and contextual factors. Compared with evaluations of programme effectiveness alone, innovative hybrid designs that pair evaluations of programme effectiveness and implementation provide more comprehensive evidence to inform scale up, policy and funding to support and sustain these programmes.⁶⁸

Objectives

The objectives of this study are to (1) evaluate the short-term effects of an 8-week, virtual, group-based, task-oriented CBEP-HCP (TIME™ at Home) compared with a waitlist control on improving everyday function (primary outcome), mobility, reliance on walking aids, well-being, caregiver assistance, caregiver mood and caregiver confidence in care-recipient balance (secondary outcomes) in community-dwelling adults with mobility limitations; (2) determine whether site, mobility level, participation with a partner, sex or gender modifies the effect of the TIME™ at Home; (3) assess the cost-effectiveness of TIME™ at Home and (4) as part of a planned process evaluation, describe programme implementation and explore exercise participants’, caregivers’ and programme providers’ experiences during the intervention phase within and between study sites to better understand the intended and unintended consequences of programme components, inclusivity during social time, and potential influences of mobility level, sex and gender.

Methods and analysis

Trial design

The TIME™ at Home study (registered on 7 February 2024 with ClinicalTrials.gov NCT06245135) is a two-parallel-group, stratified type I hybrid⁶⁸ randomised controlled trial (allocation ratio: 1:1) with a multimethod process evaluation in which all aspects of the trial (recruitment, data collection, intervention) are conducted virtually (protocol version 1, 7 Fefruary 2024). One-to-one group allocation is concealed, and outcome evaluators are masked. We are running the trial at four sites in Canada including Edmonton (pop. 1 568 000),⁶⁹ Ottawa (pop. 1 021 765),⁶⁹ Toronto (pop. 2 928 879⁶⁹) and Winnipeg (pop. 777 346⁶⁹) to accrue a sufficient sample size and build capacity across provinces to continue programme implementation after the research is completed. We are using REDCap,⁷⁰ a secure data management software, for online data management.

The Standard Protocol Items: Recommendations for Interventional Trials statement⁷¹ was used to guide reporting of the protocol. The Consolidated Standards of Reporting Trials statement⁷² is being used to guide study design and reporting. The Template for the Intervention Description and Replication extension to telehealth interventions was used to guide the description of the intervention.⁷³ We are using the Trial Forge Guidance 3^{74 75} to optimise recruitment and retention of people from ethnic minority groups to optimise participant diversity and generalisability of results. Table 1 provides details of strategies we are using to promote equitable access to research participation and to generate evidence of equity and inclusion in this study. Figure 1 provides an overview of the study design.

Table 1. Study strategies to optimise equity, diversity, inclusion and accessibility.

Study component	Strategy
Overall study design	Study is guided by Trial Forge Guidance 3 recommendations for inclusion and retention of individuals from minority ethnic groups. Trial targets a population, specifically community-dwelling adults with a mobility limitation, regardless of the underlying health condition, that experiences difficulty accessing exercise opportunities designed for the general public. Study is conducted at 4 sites across Canada and partnerships with not-for-profit community organisations that support people with disabilities at national and regional/local levels. Virtual delivery of a prerecorded exercise video streamed by facilitators and exercises requiring a minimum of equipment is a low-cost model expected to increase access to exercise participation.
Recruitment/eligibility criteria	Recruitment materials for the exercise participant and the caregiver include simple terminology and photos of individuals from different ethnic backgrounds. Recruitment strategies to increase access to under-represented groups include: Recruitment from hospital-based outpatient programmes that provide service to diverse clinical populations with mobility limitations Advertisement through organisations that cater to underserviced communities (eg, 2SLBTQI+, Black, South Asian, Chinese, Korean). Participants can participate with a partner (social capital). No exclusion criteria are based on having sufficient space in the home environment to accommodate performance-based tests (eg, 3 m walk test) or the virtual exercise programme. Site coordinator screens participants for technology needs. As needed, consenting participants are provided with: A mobile device (ie, tablet and stand) for the duration of the study A data plan for Internet access An email account A written guide and educational session to review how to use Zoom.
Data collection	PROGRESS Plus framework98 was used to guide quantitative data collection to ensure ability to describe diversity of study participants. Ontario Anti-Racism Data Standards116 were used to guide the wording of questionnaire items to collect data on PROGRESS Plus variables. 2 video-recorded TIME™ at Home classes (weeks 3 and 7) will be reviewed to evaluate the extent to which interactions align with programme provider training on fostering an inclusive environment. Qualitative interviews and focus groups explore exercise participants’, caregivers’ and programme providers’ perceptions of inclusivity.
Virtual evaluations of participants using Zoom	A ‘study partner’ must be present during Zoom evaluations to assist (eg, with technology and equipment). When some participants reported challenges with attaining a study partner, we amended the study protocol (Amendment approved on 14 Novemver 2024) to allow for the site coordinator to serve as the study partner. Evaluators also judge whether it is safe to conduct follow-up evaluations without a study partner. Evaluators modify test protocols (eg, use a shorter distance for the 3 m walk test) to suit the home environment.
Stratification and randomisation	In case the Timed ‘up and go’ test cannot be completed due to inadequate space, we amended the study protocol (Amendment approved on 14 November 2024) to stratify participants based on 30 s sit-to-stand test performance as this test only requires a chair.
Intervention	Programme provider training: Facilitators and healthcare partners complete an online learning module to develop awareness of how to facilitate an inclusive environment Facilitators and healthcare partner training includes suggestions of topics for promoting inclusive conversation during the social component Facilitators are provided with a list of activities for social time based on participant feedback from previous feasibility study Exercise videos encourage self-pacing and provide exercises at two different levels of challenge to support individuals with varying abilities. Exercise videos show two therapists demonstrating the exercises who differ by gender and race/ethnicity. No space or equipment requirements to participate in the programme. Provision of two class times (morning and afternoon) at Toronto site to accommodate different schedules. Provision of evening exercise classes at Toronto site for those who work is being explored.

2SLBTQI, 2-spirit, lesbian, gay, bisexual, transgender, queer/questioning, intersex; TIMETM, Together in Movement and Exercise.

Figure 1. Overview of study design.

Patient and public involvement

People with mobility limitations participated in foundational feasibility studies and provided feedback on the safety, feasibility, acceptability and benefit of the study intervention and the safety, feasibility and acceptability of data collection methods.^{64 65} Findings from the quantitative pre–post evaluations and qualitative interviews informed the selection of primary and secondary outcomes for this study protocol. People with mobility limitations and representatives of non-profit organisations that offer programmes to people with disability provided input on recruitment strategies to increase access to the study among under-represented groups in a concurrent study.²⁸ Community organisations (eg, Heart and Stroke Foundation, March of Dimes Canada) that have partnered with the research team will be involved with decisions on how to disseminate study results to participants and relevant communities.

Eligibility criteria: exercise participants

The target population for the trial is community-dwelling adults with mobility limitations.

Inclusion criteria

Individuals meeting the following inclusion criteria, confirmed through self-report, are considered eligible: (1) adult defined as age 18 years or older; (2) mobility limitations, defined as need to use a walking aid such as a cane or walker; or difficulty stepping onto curbs, ramps or stairs or difficulty walking on uneven surfaces such as grassy fields or loose gravel; or difficulty walking up and down hills; or challenges to walk while running errands in the community; (3) living independently in the community (in own home or assisted living settings); (4) able to walk a minimum of 10 m independently (with walking aids if used) without assistance and/or supervision from another person; (5) able to stand up from and sit down onto a chair independently, without supervision; (6) able to maintain balance while exercising in standing (eg, marching on the spot) holding onto the back of a chair or a sturdy countertop; (7) has a study partner (eg, family member or friend) willing to be present, at minimum, during the first evaluation in the home and (8) able to speak and read English to understand informed consent and follow instructions for study procedures and exercises.

Exclusion criteria

Individuals meeting any of the following criteria, confirmed through self-report, are excluded: (1) involvement in another formal exercise or rehabilitation programme in the next 2 months; (2) previous participation in the TIME™ at Home exercise programme; (3) health conditions or symptoms (eg, unstable cardiovascular disease, significant back or joint pain) preventing participation in exercise; (4) cognitive impairment, defined as a score of <11/15 on the 5 min Montreal Cognitive Assessment⁷⁶ given that sufficient cognition is needed to understand exercise instructions, choose to follow the exercise based on difficulty level and use sound judgement to self-pace during the exercise without visual monitoring from facilitators; (5) severe visual impairment, defined as self-reported difficulty while wearing usual eyewear viewing an exercise video on an electronic device, like a computer screen or tablet, that would hinder participation and safety in the virtual exercise programme and (6) severe hearing impairment, defined as self-reported difficulty, while wearing usual hearing aids, hearing an exercise video on an electronic device, like a computer screen or tablet, that would hinder participation and safety in the virtual exercise programme.

Eligibility criteria: caregivers

Caregivers of consenting participants are invited to participate in the study to complete caregiver-related study measures and participate in qualitative interviews about study experiences.

Inclusion criteria

Individuals meeting the following inclusion criteria, confirmed through self-report, are considered as eligible caregivers: (1) caregiver defined as an individual who helps the exercise participant to live at home by providing support and assistance with at least one basic (eg, self-care) and/or instrumental ADL (eg, doing groceries, cleaning, managing finances, making meals, doing laundry, etc) at least once a week⁷⁷ and (2) able to speak and read English.

Exclusion criteria

Individuals who are paid personal support workers are excluded.

Recruitment and informed consent

Community-based recruitment strategies include placing advertisements in local newspapers, and newsletters or emails from health-promoting charities/organisations and city councillors; recruiting from hospital programmes (eg, geriatric and neurorehabilitation) involving clinical populations with mobility limitations; posters in community locations (eg, grocery store, places of worship) and social media (eg, LinkedIn).

The site coordinator reviews the study with interested individuals by phone, screens for eligibility, reviews the consent form (see online supplemental file 1) and obtains verbal informed consent. In addition to providing consent to participate in the study, participants are asked to provide additional consents for: (1) having two exercise classes videorecorded if assigned to the experimental group; (2) being contacted to participate in a qualitative interview and (3) having their deidentified study data made available in a database for other researchers to use in future. Once the exercise participant has consented to participate, the site coordinator recruits the caregiver, if there is one, using a similar approach. Exercise and caregiver participants are then scheduled for the baseline evaluation. To encourage recruitment and adherence to study procedures, exercise participants and caregivers are provided with a gift card ($C15 value) for each evaluation completed. Exercise participants, caregivers and programme providers receive a gift card ($C25 value) after participating in a qualitative interview or focus group. The date of enrolment of the first participant in the study was 19 July 2024. We expect to complete data collection by 31 March 2026.

Intervention

Participants are randomly assigned to receive the TIME™ at Home exercise programme immediately (experimental group) or 5 months later (waitlist control group). A waitlist control group was considered appropriate as it mimicked usual care during the intervention period and would ultimately provide all participants with the intervention. Figure 2 presents the conceptual framework for the TIME™ at Home intervention developed based on self-efficacy theory,⁷⁸ research supporting the benefits of CBEPs⁷,¹² incorporating task-oriented training^{79 80} and evaluations of the TIME™ at Home model.^{64 65} The TIME™ at Home exercise programme is licensed by the University Health Network (UHN). A community organisation signs the licence to receive a toolkit and training to deliver TIME™ at Home. The TIME™ at Home program model in this study is based on our preliminary research⁶⁵ and involves delivering a 1.5-hour group session, twice a week, for 8 weeks using Zoom. Each session comprises: (1) a 15 min prevideo safety check and social time; (2) streaming a 1-hour prerecorded exercise video and (3) an optional 15 min postvideo social time. Two trained facilitators run each class, with a maximum group size of 14 people. The programme incorporates safety features that optimise safe, home-based exercise participation.

Figure 2. Conceptual framework for the TIME™ at Home intervention. TIME™, Together in Movement and Exercise.

The TIME™ at Home program model evaluated in this study has the following components:

1. Participant waiver and release form: Community organisations who provide the programme ask participants to sign a waiver and release form that advises examination by a physician prior to beginning the exercise programme and participant agreement to view a safety video.

2. Safety video: A 5 min YouTube™ video reviews strategies to optimise safety, including optimal home setup to exercise, importance of hand-holds, self-pacing and easy access to water and a phone in case of an emergency.

3. Pre-exercise video safety check and informal social time: During the 15 min period prior to streaming the exercise video, programme facilitators welcome everyone, troubleshoot technology issues and perform safety scans of participants and their environment. Participants are asked to keep webcams on throughout the 1.5-hour class.

4. Prerecorded exercise videos: The 1-hour exercise videos include a seated warm-up and cool down, and 40 min of exercise performed in standing. A physiotherapist and TIME™ programme lead with over 15 years of experience in neurorehabilitation selected exercises for the video from the in-person TIME™ programme^{10 18} that were considered safe for adults with mobility limitations to perform at home without supervision. These exercises include standing up and sitting down onto a chair, stepups, lunges and walking. The exercise programme is self-paced. The exercise video shows two healthcare professionals, one demonstrating a lower and one demonstrating a higher difficulty level of each exercise. There are two videos, a level 1 and level 2 video. The level 2 video continues to present exercises at the lower difficulty level but provides advanced options for the higher difficulty level such as reducing base of support (narrow stance walking); increasing limits of stability (smaller to larger weight shifts); changing the speed and direction of movement (rapid high-tempo and multidirectional stepping); adding cognitive challenge (dual-task walking) and increasing the repetitions and duration of functional tasks to challenge fitness. The programme begins with the level 1 video and switches to the level 2 video mid-programme.

5. Postvideo social time: Participants are encouraged to participate in an optional 15 min, postvideo social time after each exercise session. Facilitators are provided with a list of potential topics and activities, such as ‘Virtual Show and Tell’ to facilitate conversation in an online environment.

6. Facilitators: Two live facilitators meeting the following criteria host each programme: (1) volunteers or paid employees of the community provider; (2) experience hosting virtual group meetings and (3) excellent communication skills. Experience with group facilitation is an asset. One facilitator streams the exercise video via Zoom each class, and the other provides technological support and monitors the Zoom gallery and chat box. Facilitators ensure the smooth running of the programme and are available if an emergency occurs. Facilitators do not suggest exercise modifications during the classes; they may consult the healthcare partner if they have questions.

7. Healthcare partner e-visits: A registered healthcare professional with expertise teaching functional mobility (eg, physical therapist, kinesiologist), called the healthcare partner, completes three e-visits, timed as follows: (1) the first class (to ensure there are no safety concerns and to establish rapport with participants and facilitators); (2) either first or second class after the switch to level 2 video to ensure there are no safety concerns and (3) towards the end of the programme (eg, the 15th class). Healthcare partners answer questions from participants during the postvideo session and debrief with facilitators afterward to make suggestions or problem-solve as appropriate. During each e-visit, healthcare partners complete an intervention fidelity checklist.

Standardisation of the intervention

The use of prerecorded exercise videos with healthcare professionals demonstrating each exercise provides a standardised approach to administering the exercise intervention across all sites. Facilitators and healthcare partners complete a 1.5-hour training session led by TIME™ programme leads at UHN and receive written resources to learn about the TIME™ at Home program and their role and responsibilities. Facilitators and healthcare partners complete a 0.5-hour online learning module to develop awareness of how to facilitate an inclusive environment.

Data collection

Participants complete three evaluations conducted using Zoom, at 0 (baseline), 2 and 5 months with an evaluator masked to intervention allocation. Evaluators are trained as physical therapists with >1 year of clinical experience evaluating balance and mobility and using standardised assessment tools. Evaluations are video recorded to enable study personnel to verify participant performance on study measures. A study partner must be available in the home for the baseline evaluation via Zoom to assist with technology (eg, using Zoom, computer or tablet, adjusting the webcam), test setup and emergency situations. Exercise participants are asked to review a home setup guide and video that describes how to prepare the home for an evaluation session and safety and emergency procedures. We mail participants a 3 m measuring tape to optimise accuracy of home setup for walk tests.

At each evaluation, the evaluator administers self-report and performance-based measures with participants, and self-report measures with caregivers, based on protocols outlined in an evaluator manual. Procedures for administering performance-based measures via Zoom were previously deemed safe and feasible.⁶⁵ We use the REDCap⁷⁰ survey tool to enable direct data entry during evaluations. Evaluators share the screen to show a REDCap data collection form to administer self-reported items, precoded to optimise data quality. To administer performance-based measures, evaluators document scores on paper forms and enter the data into REDCap immediately postevaluation.

Evaluator training includes: (1) viewing a 30 min video providing an overview of evaluation procedures; (2) reviewing resources (ie, evaluator manual, REDCap templates, troubleshooting guide, participant home setup guide); (3) completing a calibration exercise by scoring videos of performance-based measures and (4) completing a practice evaluation using REDCap and discussing calibration results with an experienced physical therapy evaluator. We remind participants not to reveal their assigned study group during evaluations, and evaluators document unmasking.

Quantitative data collection

Primary outcome of everyday function

The primary study outcome of everyday function is evaluated using the physical scale of the self-report Subjective Index of Physical and Social Outcome (SIPSO).⁸¹ The SIPSO physical scale (SIPSO-P) is a 5-item scale. Each item describes an activity (eg, dressing, moving around the home, shopping) that participants score from 0 to 4 points where a higher score indicates a better level of integration with physical function. Item level scores are summed to produce a subscale score that can range from 0 to 20 points. The SIPSO-P has demonstrated excellent test-reliability in community-dwelling people with stroke (intraclass correlation coefficient (ICC)=0.91).⁸¹ Correlations of SIPSO scores with Barthel Index, Frenchay Activities Index and Wakefield Depression Inventory scores ranging from 0.73 to 0.80 support the construct validity of SIPSO.⁸¹

Secondary outcomes of mobility, well-being, caregiver assistance and mood

Mobility is a secondary outcome that is evaluated using six indicators. These six indicators are balance, leg strength, functional mobility, walking speed, endurance and balance self-efficacy, measured using the seven-item Berg balance scale,^{82 83} 30 s sit-to-stand⁸⁴,⁸⁶ (30STS) test, Timed ‘up and go’ (TUG) test,^{87 88} 3 m walk test (comfortable and fast pace),^{88 89} 2 min step test⁸⁵ and the activities-specific balance confidence (ABC) scale,^{90 91} respectively. We are also collecting information specific to the participant’s reliance on walking aids indoors and outdoors.

Well-being is a secondary outcome that is evaluated using four indicators. These four indicators are mood, social isolation, social integration and health status that are measured using the Centre for Epidemiological Studies Depression⁹² (CES-D) scale, the PROMIS V.2.0 Social Isolation short form (PROMIS),⁹³ the SIPSO social integration scale⁸¹ and the EuroQoL-5 Dimension-5 Level^{94 95} (EQ-5D-5L) Visual Analogue Scale, respectively. Caregiver assistance and mood are evaluated using the Caregiver Assistance Scale^{96 97} and the CES-D,⁹² respectively. Caregivers’ level of confidence in the balance ability of the exercise participant is evaluated using an adapted version of the ABC scale.

Cost data

A generic measure of health-related quality of life, the EQ-5D-5L,^{94 95} will be used in cost-effectiveness analyses. Scores on the EQ-5D-5L convert to a single index value used to calculate quality-adjusted life-years. The site coordinator tracks the hours spent by the community partner to deliver the programme. The site coordinator administers a questionnaire at the 2-month follow-up evaluation to the experimental group only, to track costs that participants and caregivers incur from programme participation, including expenses and time and willingness to pay for the programme.

Participant characteristics

At baseline, we are collecting data on sociodemographic and health characteristics of participants that include PROGRESS Plus variables^{62 98} (ie, place of residence, racial background, language, occupation, gender/sex, religion, education, income (socioeconomic status), marital status (social capital), cohabitation (social capital), availability of a caregiver (social capital), and homecare services received (social capital), sexual and gender diversity, as well as age, health conditions (Charlson Comorbidity Index^{99 100}), hearing, use of hearing aids, vision, use of glasses, employment, smoking status. For caregiver-participants, we are collecting data on age, sex, gender, racial background, language, relationship to exercise participant, education, employment and occupation, at baseline.

Harms and cointerventions

Adverse events

Programme facilitators and evaluators are documenting the occurrence and nature of adverse events observed or reported during participation in TIME™ at Home or study evaluations, respectively, using a standardised form.

Cointerventions

At 2-month and 5-month evaluations, we are collecting data on participation in cointerventions, including type of programme (eg, physical therapy, home programme, exercise class), programme duration and frequency of programme attendance.

Quantitative process evaluation

In alignment with recommended approaches for evaluating complex interventions,⁶⁶ we are undertaking a quantitative process evaluation to evaluate implementation fidelity and participant engagement. Online supplemental file 2 describes the site-level and participant-level process indicators we are using. The evaluation of implementation fidelity involves completion of fidelity checklists by healthcare partners during e-visits, on provision of TIME™ at Home intervention components. The evaluation of participant engagement involves documenting exercise participants’ attendance at each class, including the prevideo social time, exercise video and postvideo social time. We use REDCap to administer an online questionnaire to programme providers (ie, facilitators and healthcare partners) to collect information on age, sex/gender, education, race, sexual and gender diversity, religion, profession, credentials, and experience; and obtain permission to be video recorded during two exercise classes and to be contacted in future to participate in a qualitative interview or focus group.

Qualitative process evaluation

In alignment with recommended approaches for evaluating complex interventions,⁶⁶ qualitative methods,¹⁰¹ specifically a qualitative descriptive approach,¹⁰² are being used to obtain an in-depth understanding of participants’ experiences during the intervention period to help explain quantitative results for primary and secondary endpoints. We are specifically interested in understanding the intended and unintended consequences of programme components, inclusivity during social time and contextual factors (eg, mobility level, space in the home, participation in other exercise programmes) that may influence exercise participation or study outcomes. We are interviewing individuals in the waitlist control group to understand contextual factors, such as participation in cointerventions, which may influence quantitative study outcomes.

Approximately 16 participants and 4 caregivers from the experimental group, and 8 participants and 4 caregivers from the waitlist control group are invited to participate in separate Zoom interviews with a trained interviewer from the Toronto site who has experience with qualitative data collection. We are purposively sampling by site, mobility level (TUG <17 s vs ≥17 s), and gender, as well as race and participation in postvideo social time in the experimental group. Following the intervention period, site coordinators identify and invite participants to participate in semistructured interviews lasting approximately 60 min. A semistructured interview guide was designed to address the aims of the process evaluation with probes to obtain rich description. Interviews are digitally recorded. A qualitative analyst will compare transcriptions to the audio recordings to verify accuracy and replace identifiers with a code. The interviewer documents reflective notes after each interview related to new or recurring comments, emotional responses or themes. We are also inviting all programme facilitators and healthcare partners by email and obtaining verbal informed consent to participate in focus groups or interviews using Zoom to explore their experiences with programme delivery and any differences across sites. Finally, a qualitative analyst will review the video recorded exercise classes to observe the interactions between exercise participants and facilitators to discern the extent to which these interactions align with our training on fostering an inclusive environment. The insights derived from these observations will serve as a supplementary source of data to ensure a more complex, in-depth, understanding¹⁰³ of the qualitative data obtained through interviews and focus groups conducted on programme completion.

Table 2 provides the timeline for quantitative and qualitative data collection and intervention.

Table 2. Schedule for quantitative and qualitative data collection and interventions.

Study activity	Study time point
	0 months	2 months	5 months	7 months
Quantitative data collection: study outcome: measure
Everyday functioning: Subjective Index of Physical and Social Outcome (SIPSO)–Physical scale	X	X	X
Mobility (six indicators):
Balance: 7-item Berg balance scale	X	X	X
Leg strength: 30 s sit-to-stand test	X	X	X
Functional mobility: Timed ‘up and go’ test	X	X	X
Walking speed: 3 m walk test (fast and comfortable pace)	X	X	X
Endurance: 2 min step test	X	X	X
Balance self-efficacy: Activities-specific balance confidence (ABC) scale	X	X	X
Reliance on walking aids: Mobility aid used when walking indoors vs outdoors, 2 self-report items	X	X	X
Well-being (four indicators):
Mood: Centre for Epidemiological Studies Depression Scale (CES-D)	X	X	X
Social isolation: PROMIS V.2.0 Social Isolation short form	X	X	X
Social integration: SIPSO–Social integration scale	X	X	X
Health status: EuroQoL-5 Dimension-5 Level (EQ-5D-5L) Visual Analogue Scale	X	X	X
Caregiver assistance: Caregiver assistance scale	X	X	X
Caregiver mood: CES-D	X	X	X
Caregiver confidence in care-recipient balance: ABC scale (adapted for caregiver)	X	X	X
Cost-effectiveness
Quality-adjusted life-years: EQ-5D-5L	X	X	X
Programme delivery costs		X
Exercise participant costs		X
Caregiver personal costs and time		X
Exercise participant willingness to pay		X
Exercise participant and caregiver sociodemographic and health characteristics	X
Programme provider sociodemographic characteristics	X
Process evaluation (implementation fidelity, participant engagement)	X	X
Harms and co-interventions
Adverse events (standardised form)	X	X	X	X
Co-intervention (standardised questions)		X	X
Intervention
TIME™ at Home program (experimental group)	X	X
TIME™ at Home program (waitlist control group)			X	X
Qualitative data collection
Videorecording of 2 TIME™ at Home classes (experimental group)	X	X
Interviews: exercise participants and caregivers		X
Focus groups: programme providers		X

TIMETM, Together in Movement and Exercise.

Data monitoring

A data safety monitoring board (DSMB) is reviewing adverse events. The DSMB is composed of three researchers with expertise in the rehabilitation of older adults who are free of conflict of interest. The DSMB meets three times a year to review the occurrence of adverse events. A serious adverse event in this study is defined as one that leads to death, a life-threatening adverse event, inpatient hospitalisation or a persistent or significant disability or incapacity that limits ADL and lasts more than 48 hours.^{104 105} Falling is a primary concern given that participants are individuals with mobility limitations who are performing balance and mobility exercises in their homes, without in-person supervision of an instructor. A fall is defined as ‘a sudden or unintended change in position resulting in the individual landing at a lower level such as the floor, the ground, another object or surface including falling back on a chair.¹⁰⁶ Thus, our stopping rule is based on the estimated rate of 7% of injurious falls in community dwelling older adults in Canada.¹⁰⁷

Stopping rule

If the rate of serious adverse events resulting from the TIME™ at Home exercise programme (ie, occurring during an exercise class or as a result of the exercise) exceeds 7% of those assigned to the experimental group, then the investigators will consider halting the trial. The DSMB will discuss whether there is sufficient evidence to stop the trial across all sites. The decision to stop the trial will be made by the principal investigators, in consultation with the DSMB and coinvestigators.

Sample size

Sample size was calculated¹⁰⁸ for our primary analysis based on a two-sample t-test as we aim to compare the mean change in SIPSO-P scores from baseline to 2 months between intervention groups. Randomisation will, on average, yield study groups that are similar in known and unknown confounders.⁶¹ Therefore, the sample size for an unadjusted analysis was estimated. A sample size of 86 per group (α=0.05, power=80%) is needed to detect a between-group mean difference of 1.5 points on the SIPSO-P given the SD of SIPSO-P change scores of 3.5 observed in our pre–post feasibility study.⁶⁵ A difference of 1.5 points on the SIPSO-P corresponds to an effect size of 0.4. To account for a 15% withdrawal rate observed preintervention to postintervention in our feasibility study of TIME™ at Home,⁶⁵ a total sample size of 198 is required. We aim to recruit a total sample size of 200, 100 per group. The Toronto site aims to recruit 80 exercise participants, and the Edmonton, Ottawa and Winnipeg sites each aim to recruit 40 exercise participants. Based on experiences in our pre–post feasibility study,⁶⁵ we expect to recruit 110 caregivers in the proposed trial.

Stratification and randomisation

Following the baseline evaluation, the site coordinator is using the randomisation module in REDCap⁷⁰ to stratify participants by site, participant type (participating as an individual or with a partner¹⁰⁹) and functional mobility level based on TUG score (high mobility: <17 s, low mobility: ≥17 s), and randomly assign them in blocks of 2 or 4 to receive the TIME™ at Home program either immediately (experimental group) or 5 months later (waitlist control group). TUG scores have been previously found to partially predict SIPSO-P scores in people with stroke.¹¹⁰ Analysis of baseline data (n=17) from the TIME™ at Home pre–post feasibility study⁶⁵ showed that 2 levels of mobility based on a TUG score of <17 and ≥17 discriminated SIPSO-P scores. In case TUG data are missing due to insufficient space in the home, we amended the study protocol (Amendment approved on 14 November 2024) to permit stratification by functional mobility level based on 30 STS score (high mobility: ≥7 stands, low mobility: <7 stands). A biostatistician external to the research team prepared the randomisation module.

Quantitative data analysis

Primary endpoint

In the primary analysis, we will use a mixed-effects multivariable regression model¹¹¹ to test for differences between the TIME™ at Home program and the waitlist control group for the primary endpoint (everyday function) across measurement occasions. Specifically, the model will include a random intercept, and fixed effects of group, time (categorical variable), the two-way interaction of group and time, stratification variables (site, participant type, mobility level) and other grouping variables (ie, cohort). Linear contrasts will be used to test for differences between the groups in the difference between the baseline and 2-month, and baseline and 5-month measurement occasions. The primary analysis will be intention to treat and include all participants randomised in the intervention groups to which they were assigned, with a secondary per protocol analysis.⁷² We will examine the potential missing data mechanism¹¹² to determine an optimal approach to address missing data on the primary outcome. We will test for effect modification by mobility level (TUG <17 s vs TUG ≥17 s), sex and gender. A dose-response analysis will be undertaken to examine whether the extent of change in SIPSO-P score would vary based on the total number of TIME™ at Home sessions participants attend.

Primary hypothesis

The mean change in score on the SIPSO-P from baseline to 2 months will be significantly higher in individuals assigned to the immediate TIME™ at Home group than in individuals assigned to the waitlist group, after accounting for site, participant type, mobility level and cohort.

Secondary endpoints

We will take a two-step approach to analysing secondary endpoints. First, we will use a mixed-effects multivariable regression model¹¹¹ to test for differences between the TIME™ at Home program and the waitlist control group for two composite secondary endpoints of mobility and well-being across measurement occasions. The composite secondary endpoints are used to summarise intervention effects. Scores on each measure underlying each secondary outcome will be transformed into a z-score at each timepoint to standardise scores based on the same value of the baseline mean and baseline SD pooled across intervention groups. For example, a z-score for a measure at 2 months will be computed as:

z-score_2mon=individual score_2mon–mean of scores_0mon / SD of scores_0mon pooled across intervention groups.¹⁰⁹

For each composite secondary outcome, we will build a regression model using an approach similar to that used for the primary endpoint. For each model, we will enter each relevant z-scored difference as the dependent variable. Independent variables include a clustered ‘measure’ variable; intervention group (TIME™ at Home vs waitlist); an interaction term between intervention group and the measure variable; and site and mobility level. For each secondary outcome, we will report the regression parameter for the association of intervention group and the z-scored difference outcome (0–2 months and 0–5 months change) along with the associated 95% CI. Summary models will be run with listwise deletion and interpretation will be verified with results from models employing last value carried forward or last value carried backward to impute for missing data.

Second, we will use a mixed-effects multivariable regression model to test for differences between the TIME™ at Home program and the waitlist control group for each individual measure of secondary outcomes (outlined in table 2), including mobility, reliance on walking aids, well-being, caregiver assistance, caregiver mood, and caregiver confidence in care-recipient balance, across measurement occasions, and change in everyday function from 0 to 5 months. This step will enable an in-depth understanding of intervention effects. This analysis approach will also be used to analyse intervention effects on the primary endpoint.

Qualitative data analysis

A directed content analysis¹¹³ of qualitative data will be undertaken. In the first phase of analysis, the qualitative research analyst (RA) will review the transcripts by site and develop an initial codebook informed by the study objectives and the conceptual framework for the TIME™ at Home program model (figure 2). Investigators with qualitative expertise and the principal investigator will familiarise themselves with the data by reviewing selected transcripts and meet with the RA to discuss and provide feedback on the codebook. Following the discussion, the RA will revise the codebook as needed. Following this, the RA and a second researcher will independently code a sample of the transcripts, meet to discuss and reconcile discrepancies and finalise the codebook. The RA will then apply the coding scheme to all transcripts using NVivo software. Given the study has multiple sites, the RA will review coded data within sites and then compare and contrast findings across sites.¹¹⁴ There will be iterative meetings between the RA and the researchers to review the clustering of similar codes to identify themes. The RA will maintain an audit trail, including documentation of decision-making and reflections, and quotes will be used to support themes.

Study organisation and data management

Each site includes an academic institution overseeing the research and a community partner organisation, including Westend Seniors Activity Centre (Edmonton), YWCA-YMCA Winnipeg (Winnipeg), March of Dimes Canada (Toronto), City of Ottawa (Ottawa), that implements the TIME™ at Home program. Each site is staffed by a site coordinator and evaluators. Each community partner has trained facilitators, who host the TIME™ at Home program, and a healthcare partner.

NMS leads the central coordinating centre in Toronto and is responsible for all aspects of trial implementation. MTB from the Toronto Rehabilitation Institute-UHN is the co-principal knowledge user for this study. CAJ, RB, NMS and HS/LS are lead investigators for the Edmonton, Winnipeg, Toronto and Ottawa sites, respectively, and, with a site coordinator, oversee study implementation at that site. NMS will coordinate data analysis. RM will guide the quantitative data analysis, and JIC and SM will guide the qualitative data analysis, respectively, in collaboration with other investigators.

A unique ID number is generated by REDCap for each participant. Data exported from REDCap for analysis will not include personal identifiers. Site coordinators maintain a master list connecting REDCap ID numbers and names and contact information of participants, securely stored in password-protected servers overseen by site investigators. Data entered by each site into the University of Toronto REDCap projects are stored on the secure University of Toronto server. Sites have access to site-specific data only. NMS and the Toronto-based coordinator monitor data entry in REDCap and report to sites on data quality. Each site stores and shares confidential data using Microsoft OneDrive.

Zoom recordings are generated from a site-specific study account with restricted access. Zoom recordings of the two exercise classes and all evaluations are uploaded to a secure password-protected OneDrive folder at the University of Toronto, and the recordings on the Zoom Cloud are subsequently deleted. Zoom recordings of interviews and focus groups are generated by the Toronto site, encrypted and uploaded to Microsoft OneDrive and subsequently deleted from the Zoom Cloud. A qualitative RA will access the Zoom transcripts in OneDrive, compare the transcript to the audio recording, replace any names with codes, save the revised transcripts, and subsequently delete the interview recordings in OneDrive. Hard copies of study documents containing deidentified data are stored in a secure space within a locked cabinet in a locked office.

Ethics and dissemination

Research ethics approval and protocol amendments

Research ethics approval has been received by the Research Ethics Board at the University of Toronto (study ID: 45343, approved on 19 January 2024), Sunnybrook Health Sciences Centre (study ID: 6195, approved on 24 May 2024), University of Alberta (study ID: Pro00141069, approved on 29 May 2024), University of Manitoba (study ID: HS26453 (H2024:145), approved on 5 July 2024) and Bruyère Health Research Institute (study ID: M16-24-012, approved on 23 May 2024). After describing the study, site coordinators confirm eligibility, review the consent form and obtain verbal informed consent from interested individuals. Substantive protocol modifications (eg, changes to eligibility criteria, outcomes, analyses) will be reviewed and agreed on by all site investigators, and formal amendments submitted to and approved by the relevant Research Ethics Boards prior to implementing the change. On 14 November 2024, an amendment was approved to: (1) increase the group size during TIME™ at Home classes from 10 to 14 participants; (2) permit the site coordinator to serve as the study partner and (3) to stratify based on 30STS performance if the TUG could not be completed.

Access to data

Deidentified individual participant data reported on after the trial will be made available beginning 6 months and ending 36 months following article publication.¹¹⁵ The final trial dataset will be made available through Microsoft Dataverse, a secure cloud-based data platform. The study protocol, statistical analysis plan, informed consent form and analytical code will be made available. Researchers who have institutional ethics approval to conduct a secondary analysis to: (1) verify study results; (2) identify the characteristics of people who benefit from the TIME™ at Home program or (3) examine relationships among study variables, will be permitted data access. Data will not be shared with or used for private industry. Proposals may be submitted up to 36 months following publication of results on primary and secondary outcomes. Information regarding submitting proposals and accessing data will be provided in the article reporting on trial results.

Ancillary and post-trial care

Given Canada has a publicly funded healthcare system, participants may follow-up with their healthcare provider in the case that health issues (eg, new or exacerbation of existing health condition) arise. Each site will be responsible for reimbursing the participant for any out-of-pocket expenses required for medical care in the event that they experience an adverse health event (eg, injury) as a result of participation in the study.

Dissemination policy

All individuals who meet the International Committee of Medical Journal Editors criteria for authorship will be listed as coauthors on study reports. Tailored lay summaries will be used to share study findings with participants and healthcare and community organisations involved in this study. Study partners, including the Heart and Stroke Foundation of Canada, March of Dimes Canada, Alberta Health Services Provincial Seniors Health and Continuing Care, and the Centre intégré universitaire de santé et de services sociaux, will assist with sharing study findings within their networks, including the TIME™ network, charities, non-governmental organisations and stroke networks. TIME™ program leads (MC, NH) will integrate findings into training and marketing materials to promote program scale-up. We will post study results to the trial registration website at ClinicalTrials.gov (Identifier: NCT06245135) that is available to the public. We will use conference presentations and journal publications to reach scientists, educators and developers of physical activity recommendations, guidelines and CBEP-HCPs. We will develop a public health brief to support community organisations and partners with implementing the programme and obtaining funding. The study will build capacity to deliver the TIME™ at Home program among community partner organisations involved in the trial.

The Canadian Institutes of Health Research has no role in the design, conduct or reporting of this study.