Active Participation of Care Partners in a Physical Activity Intervention Alongside People with Stroke: A Feasibility Study

Abstract

Purpose: This study investigated the feasibility of a physical activity intervention for people with stroke and their care partners and the role social support plays in physical activity adherence. Method: The study used a single-group, pretest–posttest design with follow-up. Participants were adults with chronic stroke and their care partners. The intervention consisted of 8 weeks of structured, group-based physical activity classes, followed by 19 weeks of self-directed physical activity. Recruitment, adherence, safety, and retention were assessed. Familial social support was assessed before and after the 8-week structured portion and again 19 weeks later. Results: A total of 21 participants (15 people with stroke, 6 care partners), mean age 67.6 (SD 11.6) years, were recruited; 19 (90.5%) completed the 19-week assessment. No adverse events were experienced during the programme. Attendance during the 8-week portion was better than during the 19-week portion (mean difference 0.95; p < 0.001; 95% CI: 0.71, 1.19 visits/wk). No relationship was found between social support and physical activity adherence (p > 0.05). Conclusions: Involvement of care partners in a physical activity intervention is feasible and safe. Both people with stroke and their care partners may require ongoing support to participate in long-term physical activity. The relationship between social support and physical activity adherence requires further study.

Currently, an estimated 6.8 million people in the United States live with stroke;¹ by 2030, stroke prevalence is expected to exceed 10 million.² Stroke is a leading cause of long-term disability,³ and it often results in muscle weakness,⁴ balance difficulties,⁵ depression,⁶ and fatigue,⁷ as well as an increased risk of subsequent stroke.⁸ Because of stroke-related deficits, the majority of people with stroke require assistance from a care partner (often the spouse) to perform daily activities.⁹ Unfortunately, studies have shown that providing care is detrimental to one’s own health because it increases the risk of developing poor health outcomes, such as cardiovascular disease.^10–12

Physical activity may reduce the negative effects of stroke,¹³ the risk of subsequent stroke,¹⁴ and, in people without stroke, the risk of cardiovascular disease.¹⁵ Physical activity may also have a positive impact on psychological health by reducing depression and anxiety.¹⁶ Despite the benefits of physical activity, however, most people with stroke are largely inactive, taking less than half the number of steps per day than the general population and spending up to 80% of their waking hours being sedentary.¹⁷

Although rehabilitation after stroke can restore physical function,¹⁸ it typically lasts a short time. The median length of stay for people with stroke in the United States includes 4 days in an acute inpatient facility and 15 days in an inpatient rehabilitation facility.¹⁹ After undergoing inpatient care, people with stroke may discharge home with home health, to an outpatient facility, or home without services. All in all, structured rehabilitation may end after 6 months or less.¹⁹ Thus, a patient-to-participant gap is created, leaving many people with stroke and their care partners independently responsible for restoring or improving physical function.²⁰

Physical inactivity is also prevalent among care partners. As a group, care partners of those with stroke have been classified as “low active,” based on the number of daily steps they take.²¹ This inactivity may be due to the increased demands associated with providing care for people with stroke. These data suggest that physical activity interventions targeting care partners are also warranted.

Lack of physical activity after stroke may be related to post-stroke deficits,²² and people with stroke report barriers including the cost and availability of physical activity programmes,²³ limited knowledge of how to exercise safely,²³ and low levels of self-efficacy.²⁴ In contrast, motivation, planned activities, and higher levels of social support have been identified as facilitators of physical activity.^22,24 Specifically, social support may have a positive effect on community participation after stroke, which could include physical activity and may also be important for care partners to initiate and continue participation in interventions involving both parties (i.e., people with stroke and their care partners).^25–27 Still, the relationship between social support and physical activity has not been fully investigated among people with stroke and especially not among their care partners.

Although involving care partners in interventions for people with stroke is strongly recommended,²⁸ interventions tend to focus on the overall health of people with stroke or the psychological, rather than physical, health of their care partners. When interventions involve both parties (often referred to as dyad interventions), the role of the care partner is often passive and aimed at improving outcomes for the person with stroke. Despite care partners having similar cardiovascular health as people with stroke,¹² to our knowledge, no studies to date have examined the feasibility of having care partners actively participate in a physical activity intervention alongside people with stroke.

The primary aim of this study was to investigate the feasibility of actively involving care partners alongside people with stroke in the Neurological Exercise Training (NExT) programme, a group-based physical activity programme designed to teach and encourage people with stroke and their care partners to safely exercise in the community. The NExT programme took place at a local YMCA and consisted of an 8-week supervised physical activity programme followed by 19 weeks of self-directed physical activity.

To determine feasibility, the study aimed for a recruitment rate of 50% or more of all screened participants, an attendance rate of 80% or more of all classes during the 8-week portion of the programme, no adverse events occurring while participating in the programme, an 80% or higher retention rate,²⁹ and the majority (i.e., > 50%) of participants reporting positive changes in exercise ability.

Secondary and exploratory aims included examining the relationship among social support, reported physical activity level, and YMCA attendance, in addition to comparing changes in exercise self-efficacy between people with stroke who have care partners and those who do not. We aimed to remove commonly reported barriers to physical activity and foster social support, thus enabling people with stroke and their care partners to transition from supervised to self-directed physical activity. Although further evaluation of the physical health of care partners should be conducted, this study takes the first step in addressing the physical health of both people with stroke and their care partners.

Methods

Study design and participants

This study used a single-group, pretest–posttest design with follow-up. A convenience sample of people with stroke and their care partners was recruited from a database of past research participants (or those who had been excluded from previous research but had indicated interest in future studies) at the University of South Carolina. Participants were also recruited through word of mouth and from a local stroke support group. To be included in the study, participants had to be (1) community-dwelling adults (aged ≥ 18 y) at least 6 months post-stroke or no longer receiving rehabilitation for a stroke; (2) able to follow instructions in English and able to report (verbally or nonverbally) fatigue, pain, shortness of breath, and exertion; (3) able to provide consent; (4) able to ambulate at least 10 metres with or without the use of an assistive device; and (5) able to travel to and from the YMCA in which the intervention took place.

Participants were excluded if they required one-on-one supervision for safety during physical activity (e.g., if they required someone to walk alongside them to prevent a fall). Also, following YMCA policy, people with stroke were excluded if they did not obtain their physician’s clearance to perform moderate-intensity physical activity.

Care partners of individuals with stroke were included if they were aged 18 years or older, met Inclusion Criteria 2–5, and had a partner (i.e., person with stroke) who was eligible for the study. Care partners without an eligible person with stroke were excluded. Care partners were defined as adults who provided assistance to people with stroke as a result of impairments experienced after a stroke. For example, a spouse who provided transport for an individual because they were unable to drive after a stroke was considered a care partner. Care partners were verbally identified by either the person with stroke or the care partner themselves.

Before we initiated the study, we obtained approval from the University of South Carolina’s Institutional Review Board (IRB). The IRB considered our study to be a programme evaluation (i.e., not human subjects research), and thus the study was exempt.

Neurological Exercise Training programme

After pre-programme data were collected, the participants were provided with a free, 6-month membership to the YMCA, which allowed them to use all its features, including all equipment and group classes. For the first 8 weeks of the YMCA membership (January–March 2018), a structured, supervised, and group-based physical activity programme was provided. The programme was designed and implemented by NExT instructors (two licensed physical therapists with assistance from two doctoral physical therapy students); people with stroke and their care partners completed the intervention as a group. The 8-week supervised portion was designed to minimize commonly reported barriers to physical activity after stroke, build participant knowledge and self-efficacy, and facilitate social support through group classes (Table 1).

Table 1 .

Addressing Barriers and Facilitators to Physical Activity after Stroke

Barriers and facilitators	How the barriers were minimized or facilitators were used
Barrier
Cost	Participants received a free 6-mo YMCA membership.
Programme availability	There were minimal exclusion criteria.
	We developed an 8-wk, physical therapist–led supervised physical activity programme.
	Participants used a centrally located downtown YMCA with accessible parking, washrooms, showers, and exercise equipment.
Limited knowledge	We distributed an information sheet on tips for successful exercise, how to exercise safely (including checking blood pressure and ensuring no contraindications to exercise), and local options for physical activity.
	During the 8-wk portion, instructors educated on specific modes of exercise (e.g., Tai Chi and aerobic exercise) and their importance for health.
	During the 19-wk portion, instructors and physical therapy students were available to meet with participants to facilitate independent exercise at the YMCA.
Low self-efficacy	NExT instructors provided encouragement (verbal persuasion)30 throughout the 8-wk portion of the programme.
	Participants were given training on blood pressure assessment and contraindications to exercise.
	All exercises were demonstrated (vicarious experiences)31 and modified to allow each individual to safely perform them rather than exclude participants from certain activities (e.g., participants were encouraged to stand when possible, but for some with balance impairments, the activity was performed while sitting).
	As a reward for regular attendance, T-shirts with the programme’s name on them were given to participants during the last 2 wk of the 8-wk portion.
Facilitator
Social support	The 8-wk portion was performed as a group. The group format fosters social support by having people who share similar experiences (e.g., experiencing a stroke) and goals (exercising for health) perform activities together.
	Participants were encouraged to exchange contact information.
	Care partners were actively involved.
	During the 19-wk portion, 2 recreational outings were held for all participants (1 at a local park in which recreational sports were played and 1 at a restaurant).
Motivation	To enhance enjoyment during the 8-wk portion, participants were exposed to various modes of exercise, with several repeated classes for consistency and repetition.
	Instructors fostered a fun environment by using group games (e.g., recreational sports during the last class and counting the number of laps the entire group could complete in a set amount of time).
Planned activities	The 8-wk portion was scheduled to occur at the same time each week.

NExT = Neurological Exercise Training.

The NExT programme intended to transition participants from supervised physical activity (8-wk portion) to self-directed physical activity (19-wk portion). The participants were exposed to various modes of exercise and physical activity (e.g., resistance, aerobic, and balance training; Table 2) to help them find activities they enjoyed and would potentially continue long term. Several modalities (e.g., resistance and aerobic training) were repeated to maintain a degree of consistency and repetition.

Table 2 .

Content of 8-Week Supervised Exercise Classes

Week	Class focus
	Tuesday	Thursday
1	Orientation and getting moving	Resistance training (machines and free weights)
2	Circuit training (combined resistance and aerobic training)	Mobility and flexibility
3	Resistance training (free weights and resistance bands)	Tai Chi and balance
4	Circuit training (combined resistance and aerobic training)	Yoga
5	Resistance training (machine weights)	No class;* participants encouraged to choose their own workout based on previous ones
6	Circuit training (combined resistance and aerobic training)	Tai Chi and balance
7	Mobility and flexibility	Circuit training (combined resistance and aerobic training)
8	Group choice	Recreational sports (whiffle ball, bag toss, soccer)

Notes: Classes were offered 2×/wk for 8 wk. Activity was performed for 30–40 min, followed by an additional 20–30 min of participants’ chosen aerobic activity (e.g., treadmill walking, recumbent cycling, recumbent stepping, and indoor rowing).

^*Instructors were attending a professional conference.

At the beginning of the 8-week portion, the NExT instructors provided written education on how to safely exercise and ways to increase physical activity, including local options for where to be active. Verbal education on the benefits of exercise and the purpose of specific exercise modalities was also provided throughout the 8 weeks. Classes were held twice per week (Tuesdays and Thursdays) for 8 consecutive weeks (15 total classes) and lasted 90 minutes. Each class started and ended with assessments of blood pressure, heart rate, fatigue, and pain (on a scale of 0–10, with 0 indicating no pain and 10 indicating the most pain)³² to ensure that there were no contraindications to exercise.

For the first few weeks of the programme, blood pressure was assessed by the NExT instructors (e.g., physical therapists or physical therapy students), but as the programme continued, the participants began taking their own blood pressure using automated cuffs (with supervision and instruction provided by the instructors). They were not allowed to participate and were encouraged to follow up with their physician if they exhibited any of the following: a systolic blood pressure reading higher than 180 mmHg or less than 90 mmHg, diastolic blood pressure higher than 100 mmHg, heart rate less than 60 or more than 100 beats per minute,³³ pain or fatigue levels that interfered with their ability to safely mobilize, or pain increasing with exercise.

Each class then began with a structured warm-up consisting of movements that could be performed in either sitting or standing. The participants were encouraged to stand if they could do so safely. Warm-up movements were performed for 60 seconds each and included shoulder rolls, cervical and thoracic rotation, bending over to touch the floor with the hands and then raising the hands overhead, lower extremity marching in place, lateral lunges, diagonal movement patterns of the upper extremities, and heel raises. The warm-up concluded with all participants walking in a circular pattern around the gym for 5 minutes as the instructors set up the day’s activities.

The class then focused on various modes of physical activity (see Table 2) for 30–40 minutes, followed by an additional 20–30 minutes of self-directed activity, with the NExT instructors strongly encouraging aerobic activity (e.g., treadmill walking, recumbent cycling, recumbent stepping, and indoor rowing) performed at a moderate intensity (Rating of Perceived Exertion of 11–14 on a scale ranging from 6 (no exertion) to 20 (maximal exertion).³⁴ The participants were consistently given verbal education on the importance of aerobic activity, with an emphasis on the American Heart Association’s physical activity recommendations for people with stroke (a minimum of 20 min of moderate-intensity aerobic activity performed 3 d/wk).²⁸

If a participant did not want to perform aerobic activity for the recommended 20–30 minutes, the instructors had them perform an activity of their own choosing (e.g., machine weights). To ensure safety and individualization, physical activity was modified as needed on the basis of a participant’s ability and preferences. To introduce a specific activity (e.g., resistance training of the upper extremities), the instructors would explain and demonstrate the movement and then modify or adapt it so that all participants (both care partners and people with stroke) could complete it safely. Thus, the participants were never excluded from performing a certain activity but rather were given a modified version.

After the 8 weeks of supervised physical activity classes, a post-assessment was conducted, in which the NExT instructors met with the participants individually to collect data on social support, falls outside the programme, and changes in the participants’ perceived ability to exercise. In addition, the participants were given a schedule of current YMCA classes (e.g., chair yoga, SilverSneakers) and encouraged to attend or continue to use the facility independently (or with a care partner) for the remaining 19 weeks of the programme.

To facilitate the transition between the 8-week supervised portion of the programme and the 19-week self-directed portion, the NExT instructors were available on request to meet with the participants individually or in a group to help develop a physical activity regimen. Each participant received one phone call from an instructor, reminding them of the instructor’s availability; afterward, the participants were responsible for making contact with the instructors (they were given phone number and email addresses). After the 19 weeks of self-directed physical activity, a follow-up assessment was conducted to determine social support, falls, and YMCA attendance.

Outcomes and assessments

A list of all outcomes assessed, including timing of assessments, is provided in Table 3. Data were collected by licensed physical therapists between January and July 2018. To describe the sample, usual walking speed,^35,36 the Five Times Sit to Stand Test,^37–39 the 2-minute walk test,^40,41 falls history, and balance confidence (Activities-specific Balance Confidence Scale)^42,43 were assessed before the programme started. The care partners completed the same pre-programme measures as well as the Modified Caregiver Strain Index (MCSI),^44,45 a 13-item questionnaire that provides a highly reliable and valid measure of strain among informal caregivers.^45–47

Table 3 .

List of Outcomes or Assessments

Purpose	Outcome or assessment	Time point
Descriptive	Usual walking speed	Pre-programme
	Five Times Sit to Stand Test
	2-minute walk test
	Falls history (previous 6 mo)
	Activities-specific Balance Confidence Scale
	Modified Caregiver Strain Index
Feasibility	Recruitment	Throughout entire programme
	Adherence (attendance)
	Safety
	Retention
	Participant reported change in ability to exercise*	After 8-week portion
Exploratory	Sallis Social Support for Exercise Survey	Pre-programme
	Participant reported time spent exercising over previous wk	After 8-wk portion; after 19-wk portion

^*Assessed using the Global Rating of Change.

Feasibility outcomes

Programme feasibility was assessed using measures of recruitment, adherence, safety, retention, and participant-reported changes in the ability to exercise. Recruitment involved the number of prospective participants who were screened by phone and reported interest in the study, number of prospective participants contacted who did not want to participate in the study (with reasons), number of participants eligible for the study, number of eligible participants who provided consent, and number of participants who attended the initial assessment and at least one class.

Adherence was the percentage of classes attended by each participant during the 8-week supervised portion of the programme and the number of YMCA visits during the 19-week portion. For the 8-week portion, attendance was taken at each class by the NExT instructors. For the 19-week portion, all participants were required to scan their YMCA membership cards before entering the facility. Thus, attendance during the 19-week portion was collected digitally.

Safety consisted of the number of adverse events that occurred during the 6 months of the programme (8-week and 19-week portions), including any adverse events experienced while at the YMCA. Adverse events were defined as unresolving muscle soreness (i.e., lasting longer than 48 hours after activity), unmanageable self-reported fatigue, pain that interfered with mobility, and falls.^48,49 Adverse events were recorded for each class. In addition, participant-reported falls were recorded at each assessment (i.e., pre-programme, post-programme, follow-up). For example, at post-programme assessment, the participants were asked whether they had experienced a fall since the beginning of the 8-week classes.

Retention was the percentage of participants who completed the 19-week follow-up assessment with reasons for dropouts included.

Perceived exercise ability was assessed using the participant-reported Global Rating of Change (GRC) scale at post-programme assessment. The scale was created in line with recent research recommendations and asked, “How would you describe your ability to exercise compared to when the group classes began?”⁵⁰ The participants responded on an 11-point Likert scale ranging from –5 (a great deal worse) to 5 (a great deal better).

Exploratory outcomes

Social support

Social support from family was assessed using the Sallis Social Support for Exercise Survey.⁵¹ The 10-question survey (with an optional 3 additional questions) assesses how often, in the previous 3 months, family members have engaged in behaviours that supported exercise. Examples include “exercised with me,” “gave me encouragement to stick with my exercise programme,” and “talked about how much they like to exercise.” Participants rank how often these acts occurred on a scale ranging from 1 (none) to 5 (very often), but they also have the option to report 0 (does not apply), which is scored as 1. A higher score represents more social support, up to a maximum of 50. The survey has acceptable test–retest reliability (r = 0.55 to r = 0.86) and internal consistency (α = 0.61 to α = 0.91) in adults attempting behavioural change.⁵¹ All measures (of both feasibility and social support) were assessed by non-blinded examiners (i.e., the instructors).

Participant-reported exercise behaviour

To capture a subjective report of exercise behaviour, at all assessments participants were asked, “Over the last week, how much time have you spent exercising?” Participants were provided with the definition of exercise: planned, structured, repetitive, and purposeful physical activity with the goal of improving health. Responses could be given in minutes or hours.

Data analysis

Descriptive statistics (means, medians, percentages) were calculated for the participants’ demographics and feasibility measures. The Shapiro–Wilk test was used to assess normality,⁵² with medians and minimums and maximums reported for non-parametric data.

Social support for the entire sample was compared across the three assessment periods using the Wilcoxon signed-rank test (two-tailed); it was also used to compare social support at different times within sub-groups (i.e., people with stroke with care partners, people with stroke without care partners, care partners). A Mann–Whitney U-test (two-tailed) was used to compare familial social support levels between people with stroke with and without care partners pre- and post-programme. Spearman’s rank-order correlations determined the association between social support and YMCA attendance for both the 8-week and the 19-week portions of the programme.

Weekly attendance rates for both portions were compared using a paired Student’s t-test (two-tailed). Participant-reported exercise behaviour was compared at three time points (pre-programme, after the 8-week portion, and after the 19-week portion) using the Wilcoxon signed-rank test (two-tailed). Spearman’s correlations determined the association between reported exercise behaviour and YMCA attendance. A Mann–Whitney U-test (two-tailed) compared GRC scores between people with stroke with and without care partners. An α ≥ 0.05 was used to test for significance. All p-values were two-tailed to allow for bidirectional relationships (e.g., social support may have been associated with either higher or lower attendance levels).

In the event of missing data (e.g., a participant did not attend the post-programme or follow-up assessment), the average was calculated from available data and imputed. If participant data were available only for the pre-programme assessment, a last-observation-carried-forward approach was used.⁵³ All analyses were performed using IBM SPSS Statistics for Macintosh, Version 25.0 (IBM Corporation, Armonk, NY).

Results

The participants’ characteristics are displayed in Table 4. A total of 12 of the 15 people with stroke (80%) were men, and five of the six (83%) care partners were women; all care partners were family members of the people with stroke. A total of 17 of 21 (80%) of the participants were aged older than 60 years.

Table 4 .

Participants’ Characteristics at Pre-Programme Assessment

Descriptor	Median (range, min-max)*
	Total (n = 21)	Care partners (n = 6)	People with stroke with care partners (n = 6)	People with stroke without care partners (n = 9)
Age, y, mean (SD)	67.6 (11.6)	73.7 (8.7)	67.8 (16)	63.4 (9.1)
Women, no. (%)	8 (38.1)	5 (83.3)	1 (16.7)	2 (22.2)
Relationship to person with stroke	–	5 spouses; 1 mother	–	–
Roles of care partners, no. (%)
Transportation	–	3 (50.0)	–	–
Communication	–	3 (50.0)	–	–
Safety with mobility	–	2 (33.3)	–	–
Modified Caregiver Strain Index, no.†	–	10 (2–19)	–	–
Time since stroke, mo	–	–	83 (17–141)	32 (8–73)
Assistive device use	5 SPC; 1 hemi walker	–	3 SPC	2 SPC; 1 hemi walker‡
No. of reported falls, previous 12 mo	0 (0–2)	0	0 (0–2)	0 (0–2)
Usual walking speed, m/s	1.2 (0.1–1.6)	1.2 (1.0–1.4)	1.2 (0.7–1.4)	1.2 (0.1–1.6)
Five Times Sit to Stand, s	12.3 (7.8–46.7)	12.1 (10.9–18.3)	9.4 (7.8–25.2)§	13.7 (10.5–46.7)
2-minute walk test, m	169.3 (12.5–205.7)	169.4 (130.5–197.5)	175.9 (55.5–204.2)	152.4 (12.5–205.7)
ABC Scale,¶ %	85.0 (3.8–99.9)	94.4 (81.3–98.8)	85.4 (50.6–99.9)	81.3 (3.8–96.6)

Note: Dashes indicate not applicable.

^*Unless otherwise indicated.

^†A higher score indicates greater perceived care partner strain; max = 26.

^‡Participant was a primary power chair user but used the hemi walker for household ambulation.

^§One participant was unable to complete the task without use of their hands; thus, time was not recorded.

^¶A higher score indicates greater perceived balance confidence; max score = 100%.

SPC = single-point cane; ABC = Activities-specific Balance Confidence Scale.

Feasibility outcomes

Recruitment

A flow diagram of recruitment and retention is displayed in Figure 1. After the initial phone screening of 37 potential participations, 9 declined to participate and 1 was excluded because they had no transportation. Of the remaining 27 eligible participants, 6 did not attend the pre-programme assessment and thus were considered dropouts. Thus, 21 of the initial 37 (57%; target rate of > 50%) screened participants and of the 27 (78%) eligible participants were successfully recruited (attended pre-programme assessment and at least one class). The participants were recruited through the University of South Carolina’s research database (n = 12), word of mouth (n = 7), and a talk given at a stroke support group (n = 2).

Figure 1 .

Flow diagram of recruitment and retention.

Adherence

Weekly attendance rates for both the 8-week and the 19-week portions of the programme are displayed in Figure 2. Overall, participants attended 12.3 of the 15 (82%; target rate of > 80%) available classes during the 8-week portion. During the transition from supervised (8-week) to self-directed (19-week) physical activity, the NExT instructors were available to meet with the participants on request. Eight participants met with the instructors for a total of 15 unique YMCA visits. The instructors provided individualized resistance training recommendations for two people with stroke and one care partner, balance exercises for five participants (including one care partner), mobility and flexibility exercises for four participants, and upper extremity resistance training machine set-up assistance for one person with stroke. All 15 instructor-facilitated YMCA visits occurred within the first 4 weeks of the 19-week portion of the programme; there were no further participant requests for instructor assistance.

Figure 2 .

YMCA attendance rates for the NExT programme.

NExT = Neurological Exercise Training.

Paired t-tests (two-tailed) were used to compare attendance rates for the entire sample and each sub-group for the 8-week and 19-week portions of the programme. The error bars indicate the standard error of the means. For both the entire sample and each sub-group, attendance rates were higher (p < 0.05) during the 8-week portion than the 19-week portion.

Mean weekly attendance rates were significantly higher during the 8-week portion than during the 19-week portion for the entire sample (mean difference = 0.95; p < 0.001; 95% CI: 0.71, 1.19 visits), care partners (mean difference = 1.15; p = 0.002; 95% CI: 0.66, 1.64), people with stroke with care partners (mean difference = 1.05; p = 0.007; 95% CI: 0.44, 1.66), and people with stroke without care partners (mean difference = 0.75; p = 0.002; 95% CI: 0.36, 1.15).

Safety

No adverse events occurred during the 8 weeks of supervised classes or at the YMCA at any time. Vital signs taken before and after each supervised class revealed no contraindications or adverse response to exercise. At post-programme assessment, one participant reported having four falls outside of class (one in which he lost control of his power chair, one when he was picking up an object from the floor, one when going down a ramp in his power chair, and one unspecified). None of the resulting falls required medical attention. This participant did not attend the follow-up assessment and was unable to be reached by phone. No other falls were reported.

Retention

Of the 21 participants who completed the pre-programme assessment and attended at least one class during the 8-week portion of the programme, 19 (91%) completed the post-programme and follow-up assessments (see Figure 1). Two participants did not complete the post-programme assessment, but one returned at follow-up. One additional participant did not complete the follow-up assessment but was present at the post-programme assessment.

Perceived exercise ability

GRC scores are depicted in Figure 3, with 17 of 20 (85%; target rate of ³ 50%) participants reporting a positive increase in the ability to exercise. The light grey horizontal bar indicates the range of scores for the entire sample, and the three vertical bars indicate the median reported change by sub-group (e.g., care partners). The GRC was assessed after the 8 weeks of supervised classes (post-programme).

Figure 3 .

Global Rating of Change.

* One participant (person with stroke) did not attend post-programme assessment; thus, no GRC was obtained.

Comparing perceived exercise ability scores between people with stroke with and without care partners revealed no statistically significant difference (U = 35.5; p = 0.14).

Exploratory outcomes

Social support

Perceived social support for the entire sample (people with stroke and their care partners) and each sub-group is displayed in Figure 4. For the entire sample, familial social support increased from pre- to post-programme (z = –2.222; p = 0.026) but decreased from post-programme to follow-up (z = –2.113; p = 0.035). For care partners, social support increased from pre- to post-programme (z = –2.201; p = 0.028) but did not significantly change from post-programme to follow-up (z = –1.725; p = 0.08). Comparing pre-programme data with follow-up data revealed no difference in the perceived levels of social support for the care partners (z = –1.687; p = 0.09). Social support was higher among people with stroke who had care partners than among those without care partners only at post-programme assessment (U = 44; p = 0.05). The error bars indicate inter-quartile range (IQR).

Figure 4 .

Perceived social support for exercise from family.

Note: Error bars indicate inter-quartile range.

* Significant difference at p < 0.05.

For the entire sample, there were no statistically significant correlations between social support at any assessment (i.e., pre-programme, post-programme, or follow-up) and attendance for either the 8-week or the 19-week portions of the programme (p > 0.05). Attendance during the 8-week portion was positively associated with attendance during the 19-week portion (r_s = 0.573; p = 0.007).

Participant-reported exercise behaviour

At pre-assessment, participants reported a median weekly exercise amount of zero hours (IQR 2). Their reported exercise time increased to 3 hours (IQR 2.15; z = –3.163; p = 0.002) at post-assessment and remained stable at 2.5 hours (IQR 4) at follow-up (z = –0.233; p = 0.82). Reported exercise behaviour at follow-up was higher than that at pre-assessment (z = –2.763, p = 0.006). The only significant correlation between YMCA attendance and reported exercise was attendance during the 19-week portion and reported exercise behaviour at follow-up, which indicated a positive association (r_s = 0.488; p = 0.025).

Discussion

In this study, we explored the feasibility of actively involving the care partners of people with stroke in a physical activity intervention. The main results of this study are that care partners can safely participate in physical activity alongside people with stroke, with high levels of recruitment, adherence, and retention. In addition, although social support improved for the care partners after 8 weeks of supervised physical activity, these improvements were not associated with future YMCA attendance and were not maintained in the following 19 weeks, during which participants were responsible for self-directing their own physical activity.

To our knowledge, this study is the first to actively involve and target the physical health of both people with stroke and their care partners. Care partners of people with stroke experience burden and strain that may have a negative effect on their emotional and physical health.^10,11 Our sample of six care partners had MCSI scores ranging from a minimum of 2 to a maximum of 19, indicating a wide range of perceived strain associated with caregiving. A recent review of interventions involving people with stroke and their care partners showed that the physical health of care partners has been neglected in research studies.⁵⁴

Interventions to date have focused on the emotional or physical health of people with stroke, the emotional health of the care partner, or a combination of the two. Interventions for care partners of those with other diagnoses, including people with dementia, have previously demonstrated feasibility and improvements in physical activity of and burden on the care partner.^55–58 These interventions, however, were home based and targeted only care partners, whereas the current intervention involved both people with stroke and care partners. Care partners in the present study demonstrated good recruitment rates, adherence during the supervised portion of the intervention, and retention, with no adverse events.

Of note is that the median walking speed for the care partners in our study (1.2 m/s) was lower than the reported normative values for women of a similar age (1.3 m/s)⁵⁹ and similar to the walking speeds of both groups of people with stroke (see Table 4). The similar walking speeds observed among the care partners could be a result of between-groups sex differences (i.e., the majority of care partners were women, and the majority of people with stroke were men), but they may also represent sub-optimal physical health compared with adults who do not provide care for others. These findings, in combination with previous findings of poor cardiovascular health and low physical activity levels among care partners,^12,21 suggest that future physical activity interventions should strongly consider involving the active care partner.

The relationship between familial social support and physical activity remains unclear. In this study, social support was not associated with YMCA attendance during either the supervised or the self-directed portions of the programme. Although the care partners reported increased familial social support after the 8-week supervised portion, social support returned to pre-assessment levels after the 19-week portion; this finding suggests that continued support in the form of supervised classes may be necessary for care partners to continue to feel supported to exercise.

Although social support is a reported facilitator of physical activity after stroke,^22,24 we found no association between social support and YMCA attendance in our study. Similarly, Alzahrani and colleagues found that spousal support was not associated with accelerometer-assessed physical activity levels.⁶⁰ The lack of an association can likely be attributed to the multi-factorial nature of physical activity behaviour. Many factors, including walking endurance,⁶¹ balance,⁶² self-reported physical function,⁶³ pre-stroke physical activity levels,⁶⁴ and mood,⁶⁵ have been found to be associated with physical activity levels after stroke. Overall, more research is needed to help determine what contributes to physical activity behaviour in both people with stroke and care partners of various abilities. In turn, tailored physical activity interventions could be developed and implemented for both groups.

At pre-programme assessment, familial social support of people with stroke with care partners did not differ from those without care partners. Immediately after the 8-week supervised portion (post-programme), people with stroke with care partners reported higher social support than those without care partners, a finding that parallels the increase in social support reported among care partners. For people with stroke and their care partners, the group format of the classes may have fostered feelings of support as they engaged in the same physical activity together.

We aimed to remove commonly reported barriers to physical activity and foster social support, thus enabling people with stroke and care partners to transition from supervised to self-directed physical activity. At the end of the 8-week supervised portion of classes, the participants reported an increase in their ability to exercise; all sub-groups (i.e., people with stroke with care partners, people with stroke without care partners, care partners) exceeded the minimal clinically important change of 2 points on the 11-point GRC scale.⁶⁶ No difference was found between the GRC scores of the eight people with stroke with care partners and those of the six without care partners; this finding suggests that the presence of a care partner did not lead to greater exercise self-efficacy. When dividing participants into sub-groups, the small sample sizes limited our ability to detect a difference. Further studies of the role that care partners play in promoting self-efficacy should be conducted.

At pre-assessment, the participants reported a median of zero hours spent exercising over the previous week. Although the increases in reported exercise to 3 hours at post-assessment and 2.5 hours at follow-up are encouraging, reported exercise at follow-up was only weakly correlated with YMCA attendance during the 19-week portion of the programme. This finding could suggest that the participants were obtaining exercise outside the YMCA, or perhaps it reflects known discrepancies between participant-reported activity levels and those measured more objectively.

Participant-reported physical activity levels are often higher than those measured with devices such as accelerometers.^67,68 Another possible reason for the reported increase in exercise is social desirability and wanting to please the examiners. Ideally, future interventions would capture both subjective and objective measures of physical activity because these likely measure different constructs but are nonetheless important for understanding the complex behaviour of physical activity.

Increases in participant-reported exercise behaviour and positive changes in reported exercise ability did not appear to influence YMCA attendance: participants attended the 8-week supervised portion at a higher weekly rate than they did the 19-week self-directed portion. Decreased attendance during the 19-week portion could have been a result of inadequate duration or content of the 8-week supervised portion. People with stroke and their care partners may require additional time and instruction beyond 8 weeks and may require a more specific, theory-driven intervention to elicit a change in their behaviour.^69,70 Another possibility is that people with stroke may prefer supervised and structured programmes,⁷¹ with structured programmes potentially addressing the physical activity knowledge and self-efficacy barriers commonly reported after stroke.^23,24

Moreover, being physically active alongside other individuals with stroke has been reported as a facilitator of exercise,²² and it may be important to foster continued physical activity. Although removing commonly reported barriers to physical activity and targeting facilitators is an essential component of promoting physical activity, people affected by stroke may need ongoing, continuous support from knowledgeable providers to adhere to physical activity in the long term. Although group physical activity classes for people with stroke are prevalent throughout Canada,^72–74 the United States does not currently offer such programmes. Rather than creating new programmes, as was attempted in this study, a more pragmatic solution in the United States may be to use an existing programme with proven feasibility. Future longitudinal research on existing community programmes is needed to determine their value in improving the health of people with stroke.

The results of this study should be interpreted with a consideration of its limitations. First, our study used a single-group design, and without randomization or a comparison group, the findings may have resulted from non-intervention phenomena such as changes in time. Second, because this was a feasibility study, we recruited a small convenience sample, which largely consisted of high-functioning men with stroke, with median walking speeds of 1.19–1.24 m per second. Thus, the study findings cannot be generalized to people with stroke with lower walking speeds, women, or those who do not express interest in a physical activity intervention.

Third, other barriers to physical activity, including transportation, were not addressed, and lack of transportation resulted in one individual not participating in the study (see Figure 1). Fourth, examiner bias may have been present because the intervention instructors also conducted assessments and were not blinded to the intervention. Future studies would benefit from objectively assessing physical activity before, during, and after interventions while concurrently assessing social support. Finally, although social support is a reported facilitator of physical activity after stroke,²⁴ this study assessed only social support coming from family and may have missed social support coming from others, such as class instructors or fellow participants.

Conclusion

This study actively involved people with stroke and their care partners in a physical activity intervention. To date, care partners involved in physical activity interventions have largely played a passive role, and their physical health has not been adequately addressed. Recruiting and retaining these care partners was feasible and safe, with no adverse events experienced. After 8 weeks of supervised physical activity, the care partners reported increased familial social support, but this improvement was lost after 19 weeks of self-directed activity. Similarly, although class attendance during the 8-week supervised portion was high, the participants decreased their use of the YMCA during the 19-week self-directed portion.

Collectively, our findings suggest that ongoing support for exercise, perhaps in the form of structured group classes, may be necessary to promote long-term physical activity for both people with stroke and their care partners. Further investigation of interventions will be needed to maximize adherence and outcomes for both people with stroke and their care partners.

Key Messages

What is already known on this topic

Despite the benefits of physical activity, people with stroke are largely inactive; this predisposes them to future cardiovascular disease. The care partners of those with stroke are also less active than the general population and as a result are at an increased risk for cardiovascular disease. Lack of available physical activity programmes is a reported barrier to activity after stroke, whereas social support is a reported facilitator. Past interventions have primarily addressed the physical health of people with stroke or the psychological health of care partners, often neglecting to address the physical health of care partners.

What this study adds

Care partners can feasibly and safely participate in a physical activity intervention alongside people with stroke. A group-based, supervised physical activity programme can increase perceived social support in care partners and people with stroke, at least in the short term. People with stroke and their care partners may require additional, ongoing supervision to continue to be active participants in physical activity.