Physical Activity Participation

Interventions to Promote Physical Activity

Given the low rates of physical activity participation, as well as the multi-level barriers and facilitators to physical activity participation, among persons with SCI, the need for effective physical activity-enhancing interventions is urgent. The physical activity intervention literature in SCI has expanded substantially in the last decade. More research groups have begun to test interventions to promote physical activity participation among persons with SCI. This section reviews physical activity intervention studies that include a physical activity-related psychosocial variable and/or a measure of physical activity participation as study outcomes.

In the general population, three types of physical activity interventions have strong evidence of effectiveness: (1) Informational interventions that focus on delivering information to change knowledge and attitudes about the benefits of and opportunities for physical activity (e.g., a community-based media campaign, informational pamphlets), (2) Behavioural interventions that focus on teaching behavioural skills to promote physical activity participation (e.g., goal-setting, planning), and (3) Environmental and policy interventions that focus on changing the physical environment, social networks, organizational norms and policies to enable physical activity participation (Kahn et al. 2002). This section reviews informational (Table 4) and behavioural (Table 5) physical activity interventions given the lack of research on environmental interventions in the SCI population.

Table 4. Interventions Using Informational-Only Strategies to Increase Physical Activity Psychosocial Correlates and/or Behaviour among Persons with SCI.

Author

Year

Country

Research Design

Score

Total Sample Size

Methods

Outcome

Bassett-Gunter, Martin Ginis, and Latimer-Cheung (2013)

Canada

RCT

PEDro=9

N=96

Population: Age=45±12yr.; Gender: males=57, females=37, not reported=2; Level of injury: Not reported; Level of severity: Not reported; Time since injury>1yr.

Intervention: Following participant recruitment and screening, baseline measures of vulnerability, response efficacy, and intentions were electronically mailed to each participant. Once baseline measures were complete, a two-step randomization procedure was followed to test the hypotheses regarding the effects of (a) risk information on vulnerability and (b) the relative effects of gain- and loss-framed LTPA message on response efficacy, intentions, and cognitive processing.

Outcome Measures: Vulnerability, Response Efficacy, Intention, Cognitive Processing.

Psychosocial variables:

1. Post hoc tests indicated a significant increase in disease vulnerability for the experimental condition only (p<0.001).

2. In the ANOVA considering response efficacy for disease risk, significant main effects for time were observed.

3. There were no significant main effects for condition or time by condition interaction effects for response efficacy.

4. In the ANOVA considering LTPA response efficacy for psychological health risk, main effects for time were superseded by a significant time by condition interaction effect.

5. Planned comparisons for each condition indicated a significantly greater increase in LTPA response efficacy for the loss-framed condition compared with the control and gain-framed conditions.

6. There was no significant difference in the magnitude of increase in LTPA response efficacy between the gain-framed and the control conditions.

7. A significant main effect for time was superseded by significant time by condition interaction effects.

8. Planned comparisons for each condition indicated a significantly greater increase in intentions for the loss-framed condition compared with the control condition and a trend toward a greater increase compared with the gain-framed condition.

9. There was no significant difference between the gain framed and control conditions.

10. Neither change in disease vulnerability (p>0.05) nor change in psychological health vulnerability (p>0.05) was a significant predictor of change in intentions.

11. Change in LTPA response efficacy for disease risk was not a significant predictor of change in intentions (p>0.05).

12. Change in response efficacy for psychological health risk was a significant and positive predictor of change in intentions (p>0.05).

13. None of the individual cognitive processing variables differed between the gain- and loss-framed conditions at the Bonferroni adjusted value of (p<0.013): total thoughts (p=0.02); favorable thoughts (p=0.04); unfavorable thoughts (p=0.23); accurate recall (p=0.07).

Foulon et al. (2013)

Canada

RCT

PEDro=6

N=79

Population: Gender: male=52, female=27; Level of injury: Paraplegia=37, Tetraplegia=42; Level of severity: AIS A=40, AIS B=10, AIS C=13, AIS D=15. Motivational Experimental Group: Mean age= 44.06yr, Mean time since injury: 20.39yr. Motivational Control group: Mean age=46.93yr, Mean time since injury: 23.21yr. Volitional Experimental Group: Mean age=42.17yr, Mean time since injury: 16.85yr. Volitional Control Group: Mean age=44.61yr, Mean time post injury: 12.70yr.

Intervention: Based on a Health Action Process Approach (HAPA) participants were categorized as being in the motivational or volitional phase of behavior change and then randomly allocated to read an experimental vignette (EV) or a control vignette (CE). The informational portrait vignettes of the EV group were tailored to their demographic characteristics and targeted social cognitions for LTPA. The CE was not tailored and was written about a man with a SCI and did not talk about physical activity.

Outcome Measures: Risk perception, outcome expectations, Task self-efficacy, Action planning, Intentions, Coping planning, Action control, Maintenance self-efficacy, Recovery self-efficacy, Perceived similarity with vignette character.

Psychosocial variables:

1. In the motivational group, those who read the EV felt more similar to the vignette character than CV group (p<0.05) on all dimensions except age and sex.

2. In the volitional group, those in the EV group felt more similar to the character on all measured dimensions (p<0.05).

3. There were no main effects of the condition or time for any of the HAPA constructs for any of the groups.

4. There was a significant condition x time interaction for coping plans. The motivational group had a non-significant decrease in coping plans among the EV group but no change for CV group. In the volitional group, there was a non-significant increase in coping plans for the CV group but no change for EV group.

Table 5. Interventions Using Behavioural Strategies to Increase Physical Activity Psychosocial Correlates and/or Behaviour among Persons with SCI.

Author

Year

Country
Research Design

Score
Total Sample Size

Methods

Outcome

Chemtob et al. (2019)

Canada

RCT

PEDro=7

N_Initial=24, N_Final=22

Population: Mean Age= 51.64 yr; Gender: Males=16, Females=6; Injury Etiology: Traumatic=13, Non-traumatic=9; Level of Injury: Paraplegia=22; Mean Time Since Injury=15.45 yr

Intervention: Intervention Group (n=10): The intervention group received one, 1-h counselling session per wk, for 8 wk, delivered via an online video-chat platform. The counselling sessions focused on fostering the basic psychological needs and autonomous motivation, teaching behaviour change techniques, and self-regulatory strategies; Control Group (n=12): The control group received no interventions and was asked to continue with their regular routine.

Outcome Measures:

Primary outcome measures: Psychological Needs Satisfaction in Exercise Scale, Treatment Self-Regulation Questionnaire.

Secondary outcome measures: Leisure-Time Physical Activity Questionnaire, total moderate to vigorous leisure time physical activity (MVPA), total leisure time physical activity (LTPA), The Life Satisfaction Questionnaire-11, Patient-Health Questionnaire-9, Patient-Perceived Participation in Daily Activities.

Psychosocial variables:

1. Compared to the control group, the intervention group reported greater autonomous motivation post intervention (Hedge’s g = 0.91)

2. Large to moderate effects supporting the intervention group were found for social cognitive predictors of LTPA (Hedge’s g > 0.76) post-intervention.

Physical activity participation:

1. Compared to the control group the intervention group reported greater levels of LTPA post intervention (Hedge’s g = 0.85).

Ma et al. (2019)

Canada

RCT

PEDro=5

N_Initial=32, N_Final=28

Population: Gender: males=17, females=11; Level of injury: Tetraplegia=13, Paraplegia=15. ProACTIVE SCI: Mean age: 45.79yr; Mean time since injury: 14.71yr. Controls: Mean age:45.57yr; Mean time since injury:18.14yr.

Intervention: Participants were performing <150min of moderate to vigorous PA per week and randomized to either ProACTIVE SCI or a wait list control group. ProACTIVE SCI was a 1h introductory session and 8 weekly 10-15min behavioural PA coaching sessions. Resistance bands were provided. A wrist accelerometer was worn on the non-dominant wrist.

Outcome Measures: Accelerometer-measured Physical Activity, Leisure Time Physical Activity Questionnaire for People with SCI (LTPAQ), Peak Oxygen Uptake test, Health Action Process Approach (HAPA) constructs.

Psychosocial variables:

1. Significant group × time effects were found for affective outcome expectancies, intentions, moderate and heavy aerobic self-efficacy, moderate and heavy strength self-efficacy, action planning, monitoring, social support, and knowledge in favor of the intervention condition.

Physical activity participation:

1. There was a significant large group x time effect of the intervention on LTPAQ total PA and moderate to vigorous physical activity.

2. The intervention group, on average, had almost three times more total physical activity and five times more moderate to vigorous physical activity than controls post-intervention.

3. Self-reported physical activity increased significantly over time within the intervention group (between baseline and week 4, 7, postintervention and follow-up).

Arbour-Nicitopoulos et al. (2017)

Canada

RCT

PEDro=9

N=90

Population: Guidelines Age=48.79±10.59yr.; Gender: males=29, females=13; Level of injury: paraplegia=17, quadriplegia=25; Level of severity: Not reported; Time since injury=17.88±11.62yr. ToolKit Age=47.11±10.23yr.; Gender: males=31, females=4; Level of injury: paraplegia=17, quadriplegia=18; Level of severity: Not reported; Time since injury=17.06±12.56yr.

Intervention: Participants were randomized to view the SCI Get Fit Toolkit or the Physical Activity Guidelines for adults with SCI (PAG-SCI) and outcome measures were taken at baseline, 24 hours post-baseline, 1-week post-intervention, and 1-month post-intervention.

Outcome Measures: Intentions, outcome expectancies, task self-efficacy, barrier self-efficacy, action planning, MVPA behaviour.

Psychosocial variables:

1. At 24-hour post-baseline, no condition effects on residual change of intentions, task self-efficacy, or barrier self-efficacy were evident.

2. Post hoc analysis revealed near significant positive changes in intentions (p=0.06) and barrier self-efficacy (p=0.05) at 24 hours post-baseline.

3. Post hoc analysis showed significant change in outcome expectancies (p=0.02) at 24 hours post-baseline.

4. No time effects were shown for task self-efficacy at 24 hours post-baseline.

5. At 1-week post-intervention, no condition effects were found for residual change in intentions, task self-efficacy, barrier self-efficacy or action planning.

6. At 1-month post-intervention, no condition effects were found for residual change in intentions, task self-efficacy, barrier self-efficacy or action planning.

7. Post hoc analysis reported a decrease in task self-efficacy at 1-week (p=0.03) and 1-month (p<0.001) post-intervention.

8. No other significant changes were found via post-hoc analysis.

Physical activity participation:

1. 1-week post-intervention, participants in the toolkit condition were 3.54 times more likely to participate in at least one bout of 20 min of MVPA compared to participants in the guidelines condition.

2. At 1-month post-intervention, participants in the toolkit condition were 1.82 times more likely to engage in at least 20 min of MVPA in the past week compared to participants in the guidelines condition.

Kooijmans et al. (2017)
Netherlands
RCTPEDro=6

N=64

Population: Gender: males=45, females=19; Level of injury: tetraplegia=22; Level of severity: Complete=50. Intervention group: Mean age: 48yr; Mean time since injury: 21yr. Control group: Mean age: 49yr; Mean time since injury: 23yr.
Intervention: Participants were randomized to either a 16wk self-management intervention (HABITS) or the control group that received information about an active lifestyle. The HABITS intervention targeted optimizing intentions toward a healthier lifestyle and improved perceived behavioral control. The intervention group received 1 home visit, 5 individual and 5 group sessions. Assessments were done pre and post intervention and at 42wk.
Outcome Measures: Amount of self-propelled wheelchair driving, Physical Activity Scale for Individuals with Physical Disabilities, SCI exercise self-efficacy scale, Utrecht Proactive Coping Competence scale, University of Rhode Island Continuous measure, Exercise Decisional Balance.

Psychosocial variables:

1. No overall intervention effect or between-group differences were shown for perceived behavioral control.

Physical activity participation:

1. No overall intervention effects were found on the amount of self-propelled wheelchair driving and self-reported physical activity.

Nooijen et al. (2016)
Netherlands
RCT
PEDro=6
N=45

Population: Mean age: 44yr; Gender: males=33, females=12; Level of injury: tetraplegia=13; Level of severity: complete=24; Mean time since injury: intervention=139d, control=161d.
Intervention: Participants were stratified based on lesion level and completeness and then randomized to either the intervention or control group. All participants completed a structured handcycle training program during their last 8wk of inpatient rehabilitation. The intervention group also had a behavioral component which was 13 individual face-to-face sessions with a coach to promote a physically active lifestyle.
Outcome Measures: Fatigue Severity Scale, The Center for Epidemiological Studies Depression Scale, Pain Intensity Score, Illness Cognition Questionnaire, Exercise Self-Efficacy Scale, Utrecht Proactive Coping Competence Scale, Social Support for Exercise Behavior Scale, Objectively Measured Wheeled Physical Activity.

Psychosocial variables:

1. There was no direct significant intervention effect for fatigue, exercise self-efficacy, proactive coping, social support family, or social support friends.

2. The intervention effect on physical activity was mediated separately by >10% by pain, disability, helplessness, exercise self-efficacy and proactive coping.

Thomas et al. (2011)

USA

RCT

PEDro=5

N=21

Population: Mean age: 43.6yr; Gender: male=10, female=11; Level of injury: C1-C7=9, T1-T5=6, Below T5=6. Mean time since injury: 12.3yr.

Intervention: Participants had not engaged in an exercise program in the previous 6 months. Participants kept a daily activity log for three months. After the first 3 months, participants were randomized to the basic intervention (BI) group or the enhanced intervention (EI) group (3 months). The BI group’s video contained education on benefits of physical activity and specific exercises that could be done. The same was given to the EI group in addition to individualized instruction in an in-home physical activity program, provided exercise supplies and given telephone check-ins periodically. Participants were evaluated at baseline, 3mo, 6mo (post intervention) and 9mo follow-up.

Outcome Measures: Transtheoretical model of health behavior change (TTM) questionnaire, Borg Rating of Perceived Exertion Scale, self-reported physical activity log.

Physical activity participation:

1. There were no significant between group differences in terms of mean self-reported days per week with a minimum of 10 minutes of continuous moderate physical activity at any assessment point.

2. The number of physical activity minutes significantly increased in the BI group at 3 months (p<0.05), 6 months (p<0.01) and 9 months (p<0.05) compared to baseline.

3. The number of physical activity minutes in the EI group increased significantly at 6 months and 9 months compared to baseline (p<0.05).

4. No significant between group differences were found in terms of number of physical activity minutes.

Wise et al. (2009)

USA

RCT

PEDro=7

N=21

Population: Basic Intervention: Age=43.3±13.1yr.; Gender: males=5, females=6; Level of injury: C1-C7=5, T1-T5=2, Below T5=4; Level of severity: Not reported; Time since injury=11.6±8.5yr.

Enhanced Intervention: Age=44.0±16.1yr.; Gender: males=5, females=5; Level of injury: C1-C7=4, T1-T5=4, Below T5=2; Level of severity: Not reported; Time since injury=13.0±10.3yr.

Intervention: Participants were instructed by a physical therapist to document their daily physical activity over 3 months from time point 1 (T1) to T2. At T2 participants were randomized to a Basic Intervention group (BIG; n=11) in which they received a brochure and a DVD/videotape explaining the benefits of physical activity and giving specific examples of appropriate exercises for individuals with SCI, or an Enhanced Intervention group (EIG; n=10) in which they received the same brochure and DVD given to participants in the BIG, as well as, individualized instruction in an in-home physical activity program, along with a varied array of exercise supplies.

Outcome Measures: Range of Motion (ROM), Upper Extremity Manual Muscle Testing (UE MMT), Self-Reported Physical Activity (min/wk).

Physical activity participation:

1. Improvement in physical activity was significant at T2 (p<0.05), T3 (p<0.01), and T4 (p<0.05) when compared to baseline value for BIG.

2. Improvement in EIG physical activity was significant at T3 and T4 (p<0.05).

3. Improvement in physical activity was not significant for between group comparison (p>0.05).

4. When the groups were combined, the degree of improvement in physical activity was significant for each assessment visit (T2, p<0.01; T3, p<0.001; T4, p<0.01) when compared to baseline value.

Arbour-Nicitopoulos et al. (2009)

Canada

RCT

PEDRo=7

N_Initial=44, N_Final=38

Population: ACP condition group: Mean age: 49.00±12.93yr; Mean time post-injury: 18.01±14.16yr; Gender: males=15, females=7; APO condition group: Mean age: 50.41±12.76yr; Mean time post-injury: 11.75±9.82yr; Gender: males=15, females=7.

Intervention: Participants were randomly divided into either an action planning group (APO) or action coping planning (ACP) group. Informational, instructional and other materials to assist with exercise were provided to participants prior to initiating a 10wk program. Both groups were facilitated in completing an action plan and the ACP group also developed a coping plan intended to assist in overcoming potential barriers.

Outcome Measures: Leisure time physical activity (LTPA) participation as measured by a short version of the PARA-SCI, Intentions (2 Likert type questions), Coping self-efficacy, General barriers self-efficacy, Facility barriers self-efficacy, Scheduling self-efficacy, Health-related break from LTPA. Most measures were collected pre and post 10wk intervention as well as mid-point (5wk).

Psychosocial variables:

1. No difference was found in the frequency with which participants altered their original action plans over the 10-week period between ACP and APO condition groups.

2. Participants in the APO condition did not spontaneously form coping plans over the 10 weeks.

3. LTPA intentions decreased for both conditions over weeks 2 to 10. No significant main effect for condition or time and condition interaction was found.

4. A significant medium-sized effect for time for general barriers self-efficacy was observed.

5. Confidence to schedule moderate to heavy LTPA decreased for both groups over weeks1 to 10. However, significant medium-large sized effects for condition were found for all 3 types of coping self-efficacy.

6. Participants in the ACP condition group had greater confidence to schedule and overcome LTPA-related barriers compared to the APO condition group.

7. The APO condition group had greater confidence to overcome facility-related barriers than did those in the ACP condition.

8. For the intervention– coping self-efficacy relationship, the ACP condition group had greater scheduling and barrier self-efficacy, and lower facility related barriers than the APO condition group.

Physical activity participation:

1. LTPA participation was significantly greater at weeks 5 and 10 for the ACP condition in comparison with the APO condition group. The main effect for time or the time and condition interaction was not significant.

Latimer et al. (2006)

Canada

RCT

PEDro=4

N_Initial=54, N_Final=37

Population: Chronic SCI; Mean age: 40.61yr; Gender: males=16, females=21; Level of injury: paraplegia (35), tetraplegia (19); Mean time post-injury: 19.34yr

Intervention: Intervention group: Subjects and researchers created implementation intentions over the telephone, for 30min of physical activity 3d/wk, for 4wk. A 4wk calendar and daily log book was emailed to the subject. After 4wk, implementation intentions and calendars were updated for subsequent 4 wks. Control group: Subjects were advised by an interventionist to engage in 30 min of physical activity 3d/wk, for 4 wks. Subjects verbally recited activities they would perform, and these were put into a calendar and emailed with a daily log book. After 4wk, verbal recitation occurred again and a new calendar and daily log was received for a subsequent 4wk.

Outcome Measures: Intentions- 2 statements used: 1) “I will try to do at least 30 min of moderate to heavy physical activity 3d/wk over the next 4 wks” (1= definitely false; 7= definitely true);

2) “I intend to do at least 30 min of moderate to heavy physical activity 3d/wk in the forthcoming month (1=extremely unlikely; 7=extremely likely); Physical Activity: Physical Activity Recall Assessment for Individuals with Spinal Cord Injury (PARA-SCI); Perceptions of control (perceived behavioural control, PBC; scheduling self-efficacy; barrier self-efficacy).

Psychosocial variables:

1. Scheduling self-efficacy: ↑ at week 5 when implementation intentions were utilized (p=0.04).

2. PBC and barrier self-efficacy did not differ between groups.

Physical activity participation:

1. Minutes of daily physical activity were higher when implementation intentions were utilized (p=0.04).

2. The overall number of days subjects participated in ≥ 30 min of physical activity was not affected by intention implementation.

3. The intentions-behavior relationship was significantly stronger in the intervention group (p=0.03), as compared to the control group.

Zemper et al. (2003)

USA

RCT

PEDro=4

N_Initial=67, N_Final=43

Population: SCI: Mean age: 47yr (range 22-80); Gender: males=30, females=13; Level of injury: paraplegia (18), tetraplegia (17), ambulatory (8); Mean time post-injury: 14yr (range 1-49)

Intervention: Intervention group: 6 – 4hr workshop sessions over 3mo, which included lifestyle management, physical activity, nutrition, preventing secondary conditions, 3 individual coaching sessions, and 2 follow-up calls within 4 mos. after workshop. Control group: no intervention.

Outcome Measures: Health Promoting Lifestyle Profile II; Secondary Conditions Scale; Self-rated Abilities for Health Practices scale (SAHP); Perceived Stress Scale; Physical activities with disabilities (PADS); Arm crank ergometer testing; neurologic exam; Body Mass Index (BMI); all at baseline and post-study.

Psychosocial variables:

1. When compared to control group, the intervention group showed statistically significant improvements in the following:

● Health practice abilities (SAHP, p<0.05);

● Health promoting lifestyle (HPLP- II, p<0.001);

● ↑ of stress management techniques, ↓ perceived stress (HPLP-II subscale, p=0.001).

Physical activity participation:

1. Physical Activity (HPLP-II): ↑ physical activity and improved physical fitness (p=0.001); however, no improvement on the PADs or physical fitness measures.

Jeske et al. (2020)

Canada

Pre-Post

N=9

Population: Median age: 39yr; Gender: males=8, females=1; Level of injury: paraplegia=2, tetraplegia=7; Median time since injury: 18yr.

Intervention: Videoconference intervention using group-mediated cognitive behavioral counseling focused on adding 20min of LTPA per week. Intervention was four, 60-min, weekly skype sessions led by a facilitator trained in behavior change techniques and group mediation. Session themes included: group unity, self-monitoring, goal setting and problem solving. An online survey was conducted at baseline, post-sessions and 24hr post-intervention.

Outcome Measures: Leisure Time Physical Activity Questionnaire for Adults with SCI.

Psychosocial variables:

1. 78% of participants (n=7) either increased or maintained their level of intention to add an additional 20min of moderate to heavy leisure time physical activity per week.

Physical activity participation:

1. 44% (n=4) added at least one, 20min bout of mild or moderate-heavy intensity leisure time physical activity during the week following the intervention.

Hiremath et al. (2019)

USA

Observational

N_Initial=20, N_Final=16

Population: Mean age: 39.4±12.8yr; Mean time since injury: 12.4±12.5yr; males=16; Level of injury: paraplegia=16, Level of severity: complete=12.

Intervention: The first, second, and third phases of the study, each 1mo long, involved collecting baseline physical activity (PA) levels, providing near-real-time feedback on PA level (PA Feedback), and providing PA Feedback with just-in-time-adaptive intervention (JITAI), respectively. A smartwatch and a wheel rotation monitor streamed data to the smartphone. Individuals received six audio/vibration prompts once/2hr to answer questions on the smartphone.

Outcome Measures: Leisure Time Physical Activity Questionnaire for people with SCI (LTPAQ-SCI), Fatigue Severity Scale (FSS), Wheelchair User’s Shoulder Pain Index (WUSPI).

Physical activity participation:

1. Participants reported 26.0±17.8 min/day of light intensity physical activity, 17.7±13.8 min/day of moderate intensity physical activity, and 11.7±15.5 min/day of vigorous physical activity at baseline.

2. After the PA Feedback phase, participants reported 28.2±23.8 min/day of light intensity physical activity, 23.3±19.8 min/day of moderate intensity physical activity, and 13.2±17.1 min/day of vigorous physical activity.

3. After the PA Feedback with JITAI phase, participants reported 25.8±22.9 min/day of light intensity physical activity,17.5±21.6 min/day of moderate intensity physical activity, and 10.6±13.5 min/day of vigorous intensity physical activity.

4. A smaller number of participants had a considerable decrease in their light- and/or moderate-intensity PA during PA Feedback with JITAI.

5. Compared to the PA Feedback with JITAI phase a smaller number of participants were able to considerably increase their light- and/or moderate-intensity PA during the PA Feedback phase.

6. Most of the participants indicated that they were performing a higher level of light- and/or moderate-intensity PA during the PA Feedback and PA Feedback with JITAI phases, but few participants indicated that chronic pain, being busy at work, weather, hospitalization not related to the study, and lack of accessible resources led to a decrease in PA levels.

Tomasone et al. (2018)

Canada

Pre-Post

N_initial=46

N_final=25

Population: Age=51.46±12.36yr.; Gender: males=23, females=22, not reported=1; Level of injury: paraplegia=23, tetraplegia=21, not reported=2; Level of severity: Not reported; Time since injury=17.00±17.59yr.

Intervention: Participants completed informational/behavioural phone call counselling sessions to explore the implementation correlates of change in leisure time physical activity (LTPA) intentions and behavior in the second phase of Get In Motion (GIM).

Outcome Measures: LTPA Intentions, LTPA Behaviours, Counselling Session Checklist, Client Reflection.

Psychosocial variables:

1. Client’s baseline intentions for engaging in aerobic, strength-training, and total LTPA were high and did not change over the course of the 6-month service (p≥0.24).

Physical activity participation:

1. Significant time effects were seen for changes in time spent in strength-training and total MVPA over the 6-month period (p≤0.03).

2. No significant changes in time spent in strength-training or total MVPA were seen between 2 and 6 months (p≥0.23).

de Oliveira et al. (2016)

Australia

PCT

N=64

Population: Inactive Group: Mean age: 48.9yr; Gender: males=51%, females=49%, Level of injury: C5-C8, A: 21.5%, C5-C8, B or C: 30%, T1–S4 to S5, A: 21.5%, T1–S4 to S5, B or C: 27%; Injury etiology: traumatic: 73%, non-traumatic: 27%; Mean time post injury: 9yr.

Active group: Mean age=48.2yr; Gender: males=89%, females=11%; Level of injury: C5-C8, A: 11%, C5-C8, B or C: 30%, T1–S4 to S5, A: 37%, T1–S4 to S5, B or C: 22%; Injury etiology: traumatic: 93%, non-traumatic: 7%; Mean time post injury: 10yr.

Intervention: Participants took part in the Spinal Cord Injury and Physical Activity in the Community (SCIPA Com), which involved supervised physical activity programs 2x/wk for 30-60min for 8-12wk.

Outcome Measures: Physical Activity Recall Assessment for Individuals with Spinal Cord Injury (PARA-SCI), Patient-Specific Functional Scale (SFS), Rosenberg Self-Esteem Scale (RSS), World Health Organization Quality of Life Scale – BREF (WHOQOL-BREF)

Physical activity participation:

1. Participants showed a significant improvement in leisure-time physical activity (LTPA) levels compared to baseline (P<0.001),

2. Participants showed a significant improvement in functional goal achievement compared to baseline (p<0.001).

3. Over time, LTPA participation was greater among the active than the inactive group, although LTPA levels among the inactive improved compared with baseline (p<0.05).

Arbour-Nicitopoulos et al. (2014)

Canada

Pre-post

N=65

Population: Mean age: 50.42yr; Gender: male=37, female=27; Level of injury: Paraplegia=30, Tetraplegia=29; Mean time since injury: 14.46yr.

Intervention: Get in Motion participants were given two elastic resistance bands, instructional guide, safety sheet and strategies for meeting LTPA goals. Participants received telephone-based counseling (10-15min) by exercise counselor trained in motivational interviewing and behavior change theory. Get in Motion service utilized the Health Action Process Approach (HAPA) model. Participants received calls weekly for first 2 months, bi-weekly for months 2-4 and monthly for months 4-6.

Outcome Measures: Intentions, self-report LTPA Questionnaire for people with SCI (LTPAQ-SCI)

Psychosocial variables:

1. Clients’ intentions for engaging in regular LTPA were high at baseline and were sustained through the 6-month period (p=0.44).

Physical activity participation:

1. There was a non-significant increase in the percentage of clients who were regularly active at baseline compared to 4 months (p=0.13) and 6 months (p=0.09).

Pelletier et al. (2014)

Canada

Pre-Post

N=17

Population: Mean age: 42.1yr; Gender: male=13, female=4; Level of injury: C3-T12; Level of severity: AIS A-C; Mean time since injury: 8.4mo.

Intervention: Participants were categorized based on discharge program (inpatient, n=9 or outpatient, n=8) and received a referral from their PT for physical activity (PA; twice per wk). The PA could be completed as unstructured LTPA or part of a structured community program. Participants also received continuous PA counselling and support for 16wk post discharge (every 4 wk). Those who did not want to participate in counselling were monitored for adherence to referral only.

Outcome Measures: Exercise beliefs questionnaire (outcome value, outcome expectation, scheduling self-efficacy, task self-efficacy), adherence (i.e., attendance or self-report).

Psychosocial variables:

1. No significant differences were found on any of the constructs (outcome value, outcome expectation, scheduling and task self-efficacy) measured between groups.

2. No significant correlations were found between any of the constructs and adherence rates.

Physical activity participation:

1. Participants attended an average of 17.4 exercise sessions out of a possible 32 (54.4% adherence rate).

Brawley et al. (2013)

Canada

Pre-Post

N=10

Population: Mean age: 57.0yr; Gender: male=5, female=5.

Intervention: Participants were recruited from a supervised leisure time physical activity program that met twice weekly and offered strength and aerobic regimens. Participants completed a group-mediated cognitive-behavioral training intervention (9wk) for increasing self-managed leisure time physical activity (LTPA). 60 min face-to-face sessions were held weekly for 7 weeks. A structured individual telephone counselling session occurred in week 9 and assessments were done at week 10.

Outcome Measures: Self-regulatory efficacy, Action plan agreement, modified version of LTPAQ-SCI, Likelihood of physically meaningful outcomes.

Psychosocial variables:

1. A significant increase in participants’ perceived likelihood of obtaining important physical outcomes consistent with their self-managed LTPA (p=0.04).

2. Self-regulatory efficacy for scheduling and planning an extra day of self-managed LTPA in the upcoming weeks was almost at the ceiling at baseline (M = 86.20 out of a maximum of 100, SD=10.49), and remained high at the end of the intervention (M= 89.43, SD=10.23).

3. Action planning showed a marginally significant increase from pre- to post-intervention (p=0.06).

Physical activity participation:

1. There was a significant increase in weekly minutes of moderate to heavy self-managed LTPA from pre to post intervention (p<0.02).

2. There was no significant difference in structured LTPA minutes.

Latimer-Cheung et al. (2013)

Canada

Pre-post

Study 1 N=7,

Study 2 N=12

Population: Study 1 (n=7): Mean age: 51.86yr; Gender: male=4, female=3; Level of injury: Paraplegia=6; Severity: Complete=4, Incomplete=3; Mean time since injury: 28.76yr.

Intervention: a single, 30min counseling session using motivational interviewing principles to strengthen social cognitions associated with LTPA. Participants were assessed the next day.

Outcome Measures: Leisure Time Physical Activity Questionnaire for People with SCI.

Population: Study 2 (n=12): Mean age: 42.92yr; Gender: male=5, female=7; Level of injury: Paraplegia=12; Severity: Complete=7, Incomplete=5; Mean time since injury: 23.21yr.

Intervention: A home visit by a certified personal trainer and a peer with paraplegia. Education about strength training, identified existing resources in the home that could be used for strength training and had exercises modelled for them that they could try while the trainer reinforced participants’ performance and past mastery experiences. Participants were assessed pre intervention, post intervention 1 week later and follow-up (5wk later).

Outcome Measures: modified Leisure Time Physical Activity Questionnaire for People with SCI, social-cognitive variables (self-efficacy, intentions, action planning).

Study 1

Psychosocial variables:

1. Significant medium to large sized increases in goal setting self-efficacy (d=0.72) and intention strength (d=1.01) (p<0.032) from pre to post intervention.

2. Small to medium sized effects emerged for intentions and action planning but they were not significant.

Study 2

Psychosocial variables:

1. Significant medium to large sized increase for task frequency self-efficacy (d=0.52), barrier self-efficacy (d=0.87), intentions (d=0.60), and action planning (d=1.14) (p<0.28).

2. There were no significant increases in task duration self-efficacy, goal setting self-efficacy, or scheduling self-efficacy.

Physical activity participation:

3. Number of bouts of strength training, duration and total min per week of strength training increased significantly (p<0.024).

4. At follow-up, 9 of 11 participants were strength training at least twice per week.

Dolbow et al. (2012)

USA

Pre-Post

N=17

Population: Mean age: 45.8±13.8yr; Gender: males=15, females=2; Level of injury: cervical=11, thoracic=6; Severity of injury: AIS A=5, AIS B=9, AIS C=3; Time since injury: 12.0±13.3yr.

Intervention: Home-based functional electrical stimulation cycling program 40-60min sessions, 3 times/wk for 16wk.

Outcome Measures: Exercise adherence.

Physical activity participation:

1. There was no significant decline in adherence over the study period.

2. The odds of adhering to the exercise program were greater for younger versus older participants, those without pain versus those with pain, and for those who were active versus inactive prior to the study (p<0.05 for all).

3. Level of injury, time since injury and history of depression had no effect on rate of adherence.

Warms et al. (2004)

USA

Pre-Post

N_Initial=17, N_Final=16

Population: Mean age: 43.2yr; Gender: 13 males, 3 females; Mean time post-injury: 14.4yr.

Intervention: “Be Active in Life” program: included educational materials (2 pamphlets, 2 handouts), a home visit with a nurse (90 min scripted motivational interview, goal and personal action plan establishment), and follow up calls at day 4, 7, 11 & 28 (approx. 8min each). Program lasted for 6wk, and had a final follow up 2wk post-completion.

Outcome Measures: Physical activity (wrist-worn actigraph); Self-rated Abilities for Health Practices Scale (includes Exercise Self-efficacy subscale); Self-rated Health Scale (SRHS); Centre for Epidemiologic Studies Depression Scale (CES-D); @ baseline, 6wk completion; 2wk post-completion.

Psychosocial variables:

1. There was no significant change in self-rated abilities for health practices from pre- to post-intervention.

2. Exercise self-efficacy significantly increased from pre- to post-intervention (p=0.05).

Physical activity participation

1. Counts/day increased in 60% of subjects, and self-reported activity increased in 69% of subjects, but both were not significant.

Discussion

Over the past decade, there has been a burgeoning amount of research exploring informational and behavioural interventions to increase leisure-time physical activity psychosocial variables and behaviour among persons with SCI. All interventions have been developed and evaluated in high-income countries including Canada, the United States, the Netherlands and Australia. Future intervention research is required to test the efficacy of physical activity-enhancing interventions for persons with SCI in low- and middle-income countries.

Recognizing the importance of offering evidence-based information about p (Williams et al. 2017), informational strategies (e.g., offering information about the benefits of physical activity or risks of physical inactivity, examples of exercises that can be performed) are sometimes used independently in interventions. Of the two interventions that used informational-only strategies, one RCT (Bassett-Gunter et al. 2013) showed positive changes, whereas one RCT (Foulon & Ginis 2013) demonstrated no change, in physical activity-related psychosocial variables. Changes in physical activity participation were not assessed in either study.

Most intervention studies used behavioural strategies. Of the 22 studies that used behavioural strategies, six RCTs (Arbour-Nicitopoulos, Ginis et al. 2009; Arbour-Nicitopoulos et al. 2017; Chemtob et al. 2019; Latimer et al. 2006; Ma et al. 2019; Zemper et al. 2003) and four pre-post studies (Brawley et al. 2013; Jeske et al. 2020; Latimer-Cheung et al. 2013; Warms et al. 2004) highlighted increases in physical activity-related psychosocial variables, whereas two RCTs (Kooijmans et al. 2017; Nooijen et al. 2016) and three pre-post studies (Latimer-Cheung et al. 2013; Pelletier et al. 2014; Tomasone, Arbour-Nicitopoulos et al. 2018) demonstrated no change in physical activity-related psychosocial variables.

Eight RCTs (Arbour-Nicitopoulos, Ginis et al. 2009; Arbour-Nicitopoulos et al. 2017; Chemtob et al. 2019; Latimer et al. 2006; Ma et al. 2019; Thomas et al. 2011; Wise et al. 2009; Zemper et al. 2003), one prospective controlled trial (De Oliveira et al. 2016), and five pre-post studies (Brawley et al. 2013; Hiremath et al. 2019; Jeske et al. 2020; Latimer-Cheung et al. 2013; Tomasone, Arbour-Nicitopoulos et al. 2018) reported changes in physical activity participation following the intervention. One RCT (Kooijmans et al. 2017) and three pre-post studies (Arbour-Nicitopoulos et al. 2014; Dolbow et al. 2012; Warms et al. 2004) reported no change in physical activity behaviour following the intervention. Of note, four interventions combined both informational and behavioural strategies (Arbour-Nicitopoulos et al. 2017; Latimer-Cheung et al. 2013; Tomasone, Arbour-Nicitopoulos et al. 2018; Wise et al. 2009). Also noteworthy is that behavioural strategies were implemented with varying degrees of intensity, from offering information about how to engage in behavioural strategies (Arbour-Nicitopoulos et al. 2014) to having one-on-one tailored interventionist support for engaging in behavioural strategies (Tomasone, Arbour-Nicitopoulos et al. 2018).

The use of multiple strategies across behavioural interventions makes it challenging to tease apart the isolated impact of individual intervention strategies. However, in a review that extracted behaviour change techniques (or “active ingredients” of behavioral interventions) (Michie et al. 2013) used in physical activity interventions for persons with SCI (Tomasone, Flood et al. 2018), the following strategies were associated with positive LTPA outcomes and can be considered in future interventions that aim to increase physical activity-related psychosocial variables and/or behaviour: goal setting (i.e., setting a level of physical activity to be achieved), problem-solving (i.e., analysis of factors influencing physical activity behaviour and selecting strategies that overcome barriers and/or increase facilitators to participation), action planning (i.e., setting a detailed plan of what, when, where and how physical activity will be performed) and social support (i.e., providing non-contingent praise and/or emotional support for the performance of the behaviour) (Michie et al. 2013).

The use of theory has been encouraged for SCI physical activity research (Best et al. 2017). Many of the included studies used an established theoretical framework to guide intervention content, intervention evaluation, and/or interpret findings. The included studies reported using theories and/or theoretical constructs from the Health Action Process Approach Model (Schwarzer et al. 2011), the Theory of Planned Behaviour (Ajzen 1991), the Transtheoretical Model (Marcus & Simkin 1994), Self-Efficacy Theory (Bandura 2004), and Self-Determination Theory (Ryan & Deci 2017). A theory is an abstract set of interrelated concepts, definitions and relationships that can predict or explain how certain phenomena, events or behaviour occur (Glanz & Bishop 2010). When considering theory use, it is important to consider how theories can be used in intervention studies. Theory can be used to (1) guide the design of the intervention (i.e., select intervention strategies that will target a theory’s constructs); (2) explore mediators or moderators of the behaviour or effects of the intervention; or (3) offer a post hoc/retrospective explanation of study findings (i.e., the theory has been introduced once the intervention is executed) (Davies et al. 2010). Studies may also vary in the degree to which theory is employed; that is, intervention studies may (1) be explicitly theory-based, wherein the intervention and evaluation of the intervention are based on a named theory, and the study offers a direct test of one or more hypotheses deduced from a named theory (i.e., to determine whether the intervention findings can be explained by the theoretical base); (2) have some conceptual basis in a theory, wherein theory is employed in the design of the intervention or evaluation, but tests of hypotheses deduced from theory are not conducted; or (3) use or examine some theoretical constructs from a theory without use of the entire theory (Davies et al. 2010). However, theory use varies in physical activity interventions for persons with SCI to date; some studies included in this chapter were explicitly theory-based whereas others did not use theory (or offered a poor reporting of theory). When theories are explicitly used to develop an intervention, it is more likely that important determinants of physical activity behaviour are targeted in the intervention, which should hypothetically increase intervention effectiveness (Glanz & Bishop 2010). Future intervention research should consider the extent to which theory is used in intervention design and evaluation if we want to fully grasp the impact of theory in physical activity-enhancing interventions in the SCI community.

While it is encouraging that theory use is expanding in this field, theory use alone cannot fully account for the effectiveness and maintenance of physical activity interventions. Other intervention features, such as intervention tailoring, dose, delivery mode, and provider, can also influence intervention effectiveness (Tomasone, Flood et al. 2018). For example, support from health and fitness professionals has been touted as important for enhancing physical activity participation among persons with SCI (Giouridis et al. 2021; Williams et al. 2017). Among included studies, the integration of health and fitness professionals was seen in different delivery formats. Several interventions included coach-counselling as a component and the counselling was delivered by a health or fitness professional (Arbour-Nicitopoulos et al. 2014; Chemtob et al. 2019; Ma et al. 2019; Nooijen et al. 2016; Tomasone, Arbour-Nicitopoulos et al. 2018; Zemper et al. 2003); or a trained peer (Latimer-Cheung et al. 2013). Of note, two interventions utilized a group-mediated cognitive behavioural intervention that was delivered by a health and fitness professional but harnessed the power of group-based sessions (Brawley et al. 2013; Jeske et al. 2020). Several interventions included structured and supervised physical activity programs where persons with SCI would exercise with supervision from a health or fitness professional (De Oliveira et al. 2016; Kooijmans et al. 2017; Pelletier et al. 2014). Other interventions offered home-based physical activity support by a health and fitness professional (Dolbow et al. 2012; Thomas et al. 2011; Warms et al. 2004; Wise et al. 2009) and one study included both a health and fitness professional along with a peer (Latimer-Cheung et al. 2013). Variety in intervention tailoring, dose and delivery mode was also evident. Researchers are encouraged to explore these additional aspects of intervention design and fully report all intervention details, not just strategies employed in interventions so that future syntheses can make recommendations. Using reporting guidelines, such as the TIDieR checklist (Hoffmann et al. 2014) will facilitate complete reporting of intervention descriptions.

Finally, the synthesis of the included interventions points to several additional areas for future research. One intervention aimed to enhance physical activity behaviour alongside other health behaviours among persons with SCI (Zemper et al. 2003). The utility of multiple behaviour change interventions among persons with SCI remains unknown and is a fruitful avenue for future research. While most behavioural interventions integrated behavioural strategies, few if any studies were explicit about providing training to participants with SCI about independent use of the strategies for self-management of physical activity beyond the intervention period. Future interventions should seek to train participants in how to use behavioural strategies (e.g., goal setting, action planning) without guidance from interventionists, with a goal to foster long-term behaviour change. Finally, and stemming from this point, is that most included studies examined the impact of interventions immediately following the intervention period, and most interventions occur over a relatively short period. Given physical activity behaviour requires sustained effort over a person’s lifetime, interventionists need to consider designing interventions that foster long-term change in psychosocial variables and physical activity participation. Extending intervention studies by including a follow-up period would begin to establish this needed evidence base.

Conclusion

There is level 1b evidence from one RCT that informational interventions are effective for increasing physical activity-related psychosocial variables among persons with SCI.

There is level 1a evidence from three RCTs, as well as support from three lower quality RCTs and four additional studies, that behavioural interventions are effective for increasing physical activity-related psychosocial variables among persons with SCI.

There is level 1a evidence from four RCTs, as well as support from four lower quality RCTs, one prospective controlled trial, and five additional studies, that behavioural interventions are effective for increasing physical activity behaviour among persons with SCI.

Future research should seek to fully employ behavioural theory throughout intervention design and evaluation, conduct a process evaluation to consider additional intervention components that influence effectiveness (e.g., dose, tailoring, delivery mode, provider), and design interventions that foster and evaluate long-term changes in LTPA psychosocial variables and participation.

Chapter Downloads

Physical Activity Participation Module With Tables

Patient Handouts

Outcome Measures

Active Clinical Trials