Table 6.

Proposed revisions to the study protocol based on pilot results

Challenges during pilot study	Proposed changes to study protocol
Recruitment • Unable to achieve recruitment target of 20 per site.	• Engage a member of the stroke team preferably in the out-patient department, known to patients, to refer people to the study. • For prospective recruitment, ensure access to patients near the time of discharge. • Highlight the type and benefits of exercises in the program in recruitment materials. • Consider targeting other clinical populations to boost recruitment given the exercise program is not specific to any health condition.
• Some participants could not fully engage with the exercise program due to a low level of physical function, comorbidity, and cognitive decline.	• Revise eligibility criteria to require individuals to have the capacity to perform sit-to-stand independently, walk 10 m independently with or without a walking aid but without assistance or supervision of another individual, and pass a cognitive screen.
• Only 68% of caregivers were recruited.	• Develop caregiver-specific recruitment materials that highlight the role of caregivers in the exercise program and potential benefits for the caregiver.
Length of wait time for control group
• 12-month waitlist period was too long and led to drop-outs and potentially co-interventions.	• Reduce the wait time in the control group to 6 months.
Evaluations
• Inclement weather and inadequate access to transportation were perceived as barriers to attending the exercise program and evaluations. • Evaluations were considered lengthy. • Monthly follow-up calls for falls monitoring were challenging to complete for ~ 25% of participants.	• Schedule evaluations and intervention periods during good-weather months if possible. • Provide participants with information about transportation services available in their region at the time of recruitment. • Budget for reimbursement for parking, adapted transportation, and driving services for remote areas. • Provide participants with gift cards as an incentive to attend evaluations, and the option to receive an evaluation summary. • Streamline the number of study measures to reduce evaluation length. • Provide flexible data collection options for those unable to attend in person, e.g., administer self-report measures by telephone. • Remove monthly falls monitoring given the exercise program was deemed safe.
Fitness instructor training
• Issues with fitness instructor availability necessitated identification and training of new instructors.	• Train 3–5 instructors annually per site to improve instructor availability and mitigate potential turnover.
Program delivery
• Participants found it distracting when other classes were being run in the same room and when rooms and class times changed between sessions.	• Ensure no other classes are being run in the same room. • Recommend using the same room and time for both classes each week.
Potential effect
• Improvement on measures of walking capacity over the 3-month exercise program was not observed. • In new sites, fitness instructors and volunteers deliver the TIMETM program for the first time during the experimental phase and may lack the expertise to progress participants.	• Incorporate additional practice of exercises for fitness instructors in the training workshop. • Have fitness instructors deliver the exercise program to an initial group of participants prior to randomization.