Abstract
BACKGROUND:
Spinal cord injury (SCI) leads to various physical, psychological, and social challenges. Sport is a holistic physical activity that may target these challenges. No literature systematically summarizes the overall impact of sport participation for those with SCI.
OBJECTIVE:
To comprehensively report the findings of quantitative studies investigating the impact of sport on the physical, psychological, and social health of individuals with SCI.
METHODS:
Six databases were searched: APA PsycInfo, CINAHL, Embase, Emcare, Ovid Medline, and PubMed (non-Medline). Studies were included if (a) participants were adults with SCI for ≥12 months, (b) outcomes resulting from ≥3 months of sport participation were investigated, (c) sport occurred in the community setting, and (d) comparisons of sport and non-sport conditions were analyzed. Details regarding study characteristics, participants, sport(s), and outcomes were extracted. Methodological quality was assessed using the Modified Downs and Black checklist.
RESULTS:
Forty-nine studies were included. Study quality ranged from poor to moderate. Sport participation showed favourable results for outcomes including function, quality of life, and community integration. Mixed results were found for outcomes including cardiac function, depressive symptoms, and employment. No significant associations were found for postural control, resilience, and education.
CONCLUSIONS:
The review findings suggest sport may be a promising intervention for addressing some challenges associated with SCI.
Introduction
Spinal cord injury (SCI) is a life-altering event that substantially impacts affected individuals, as well as their families and society as a whole (Noonan et al., 2012). SCI is caused by traumatic (e.g., motor vehicle or work-related accident) or non-traumatic (e.g., tumour, infection) events that result in the complete or partial loss of sensory, motor, and autonomic function below the level of injury ("World Health Organization. Spinal cord injury.,” 2013). This damage results in significant physical impairments and activity limitations that ultimately lead to the restriction of participation within the community and at home (Stucki et al., 2002).
Individuals with SCI may experience challenges in their physical, psychological, and social health. For instance, 65% of individuals experience chronic pain, one-third of whom report severe chronic pain (Siddall & Loeser, 2001). Furthermore, people with SCI have been shown to engage in much less daily leisure time physical activity (LTPA) compared to able-bodied Canadians (i.e., a large portion of individuals engage in no LTPA at all), which puts them at greater risk of living sedentary lifestyles and facing the associated complications in health and well-being (K.A.M. Ginis et al., 2010). In addition to the physical challenges, individuals with SCI may experience difficulties in their psychological and social health, including: elevated depressive symptoms and anxiety levels, changes in self-identity, difficulty re-integrating into the community, and lower employment rates (Charlifue & Gerhart, 2004; Kennedy & Rogers, 2000; Lidal et al., 2007; Prout & Porter, 2017; Yoshida, 1993). Together, these physical, psychological, and social challenges affect individuals’ activity and ability to participate in society, thereby negatively influencing their quality of life (QOL) (Leduc & Lepage, 2002).
Sport participation is a potential solution that may help to address the challenges individuals with SCI experience. ‘Sport’ is defined as a type of physical activity involving exertion, skill, and/or hand-eye coordination as the primary focus, with elements of competition and formally existing through organizations (Pink, 2008). Sport is an activity that often occurs within social contexts. Given its social nature, it is believed that sport is associated with improved psychosocial health, in addition to its physical benefits (Eime et al., 2013). According to Eime and colleagues’ Health through Sport Conceptual Model, ‘physical’ refers to the body or processes of the body, ‘psychological’ refers to the mind or processes of the mind, and ‘social’ refers to the interactions or sense of interactions amongst individuals (Eime et al., 2013). Engagement in sport can manifest in the form of team sport (e.g., wheelchair rugby, wheelchair basketball) or individual sport (e.g., wheelchair racing, swimming), and participation can occur on various competitive levels (Eime et al., 2013). For the SCI population, the use of sport holds high potential as it aligns with existing literature demonstrating that ‘interventions’ addressing multiple problems at once will be more successful than those focusing on just one or two areas of concern (Block et al., 2005).
Despite the promise of sport for helping people with SCI, there is limited research to date examining the overall impact of sport participation for individuals living with SCI, specifically in relation to the three domains of physical, psychological, and social health. Existing literature has reviewed physical activity or exercise in relation to SCI, but is more limited with regard to sport (Ginis et al., 2010; Hicks et al., 2011; Williams et al., 2014). Given that ‘physical activity’ is defined as any bodily movements produced by skeletal muscles that result in energy expenditure (Caspersen et al., 1985), and that ‘exercise’ is defined as any structured or repetitive physical activity performed with the main intention of improving physical fitness (Pink, 2008), the current reviews published in the literature lack intervention specificity. For the four reviews that do examine sport in the SCI population, two of these studies are narrative reviews (Martin et al., 2012; Slater & Meade, 2004) and do not comprehensively nor systematically report sport research, one study is a systematic review that only reports on psychosocial outcomes (Prout & Porter, 2017), and one is a thematic synthesis only reporting qualitative studies (Cheung et al., 2021). Thus, this review will be the first to examine quantitative studies investigating sport participation in all three domains according to Eime et al.’s Health through Sport Conceptual Model (Eime et al., 2013). Given the broad nature of outcomes to be examined in this study, a scoping review was deemed a suitable approach for systematically mapping the existing literature (Peters et al., 2015). The objective of this study therefore was to comprehensively report the impact of sport participation on the physical, psychological, and social health of adults living with SCI.
Materials and methods
Study design
A scoping review was conducted in accordance with the Joanna Briggs Institute (JBI) guidelines for scoping reviews (Peters et al., 2020, 2015). These guidelines provide detailed instructions for various stages of the review process, such as for creating a research question, developing inclusion criteria, extracting data, and presenting the results (Peters et al., 2020, 2015). The JBI guidelines facilitate a comprehensive approach to completing scoping reviews and therefore was the authors’ guideline of choice.
This review followed the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) checklist (Tricco et al., 2018). The PRISMA-ScR checklist consists of 22 items and provides a guide for comprehensive reporting in scoping reviews (Tricco et al., 2018). The objectives, data collection procedures, and data analysis plans for this review were specified in advance and documented in a protocol registered on the Open Science Framework (https://osf.io/ezduk) on January 20, 2021. This scoping review was part of a larger search strategy that also produced a thematic synthesis aiming to synthesize qualitative studies exploring the impact of sport participation following SCI (Cheung et al., 2021).
Article identification
To develop a search strategy and establish clear inclusion and exclusion criteria, the Population, Concept, and Context (PCC) elements were considered. The Population was adults with SCI; the Concept was sport; and the Context was community settings (i.e., not a rehabilitative or laboratory setting).
From inception to January 19, 2021, six databases were searched: APA PsycInfo, CINAHL, Embase, Emcare, Ovid Medline, and PubMed (non-Medline). A search strategy was developed with the assistance of an Information Specialist (MP). There were no date or language limits applied. This search strategy used only two PCC elements –Population, which included athletes with SCI and other disabilities, and Concept, which was sports. Incorporating only two PCC elements ensured that the results of the search strategy were comprehensive and that the likelihood of missing relevant studies was minimized. An example database search strategy is available for reference in the Appendix.
Following the comprehensive database search, identified records were uploaded to EndNote, a reference management software ("EndNote X9,” 2019), where duplicates were identified and removed. Next, the records were uploaded to Covidence, a web-based systematic review management platform ("Covidence.,” 2021). Three researchers (LC, KC, and MGH) engaged in a pilot title and abstract screening trial where 20 random articles were independently screened, and the interrater reliability was calculated via percent agreement (McHugh, 2012). For the purposes of the screening trial, interrater reliability over 80% was considered an appropriate threshold to warrant proceeding to the formal screening process. After passing the pilot screening trial and refining the eligibility criteria, the three researchers completed title and abstract screening, in which two reviewers independently screened each study. Conflicts that arose during this process were resolved by a third reviewer (LC, KC, MGH, or KEM). Next, the three researchers completed full-text review of the remaining articles, in which two reviewers independently reviewed each full-text. One researcher (LC) contacted corresponding authors via email to obtain any missing information that prevented a decision to be made regarding inclusion or exclusion. Google Translate ("Google Translate.,” 2021) was used to translate the 20 articles published in languages that were not English. Conflicts that arose during full-text review were resolved through discussion until consensus was reached.
Inclusion and exclusion criteria
Articles were selected for inclusion if they: a) studied individuals over 16 years of age (e.g., ages at which individuals are participating in adult leagues) who had a SCI for at least 12 months, and were of any etiology (i.e., traumatic, atraumatic), any severity of injury (i.e., complete, incomplete), any neurological level of injury (i.e., cervical, thoracic, lumbar), and any mobility status (i.e., ambulatory, wheelchair user); b) investigated physical, psychological, and/or social outcomes resulting from organized team or individual sport participation that occurred for a minimum duration of three months (to reflect the duration of a typical sport season, and to allow for any cardiovascular changes to occur (Golbidi & Laher, 2012)); c) focused on sport participation in the community setting (i.e., outside the rehabilitative or laboratory environments) at any competitive level (i.e., recreational or competitive); and d) used within-subject or between-subject comparisons of sport and non-sport conditions using randomized, non-randomized, or descriptive quantitative study designs (Hong et al., 2018).
Articles were excluded if they studied individuals with conditions that were not SCI (including congenital causes of spinal damage such as spina bifida) or consisted of mixed populations in which outcome data specific to SCI could not be extracted. Articles studying rehabilitative physical activity, activities of daily living, occupational physical activity, and LTPA or exercise not falling under the authors’ definition of sport (e.g., leisure arm cycling) were all excluded. All types of qualitative studies, conference proceedings and abstracts, textbook chapters, notes, reviews, and unpublished literature were excluded. Studies in which there were no between- or within-person analyses comparing sport and non-sport conditions were also excluded.
Data extraction and quality assessment
Prior to the formal data extraction and quality assessment process, three researchers (LC, KC, and MGH) independently engaged in data extraction and critical appraisal of three studies, after which results were compared and discrepancies were discussed to ensure consistency between researchers. This process also allowed for the refinement of the data extraction table and clarification of the extent of detail to extract. Following completion of this step, data were extracted from each article by one of the three researchers independently.
During the data extraction process, study details including year of publication, country, study design, and study aim were extracted. Participant-specific information was also extracted, including number of sport and non-sport participants with SCI, participant age, participant sex, injury details (i.e., etiology, severity, level, duration of injury), and sport details (e.g., type of sport, duration of sport participation). Outcome-specific information was also extracted in detail, including the outcomes used, type of comparison made (i.e., sport vs non-sport, pre-sport vs post-sport), and findings or interpretations from the study.
To assess study quality, the Modified Downs and Black checklist for randomized and non-randomized studies was used (Eng et al., 2007). This tool was developed using rigorous psychometric methods and is shown to be valid and reliable (Downs & Black, 1998; Saunders et al., 2003). This checklist consists of 27 questions organized into four sections: Reporting, External validity, Internal validity (bias), and Internal validity (confounding or selection bias) (Eng et al., 2007). Each of the 27 questions were scored a 0 or 1, except for question #5 regarding the distribution of principal confounders, which was scored a 0, 1, or 2 (Eng et al., 2007). Thus, all studies were given a total score out of 28, where a higher score indicated higher methodological quality (Eng et al., 2007). Total scores above 19 indicated “good” study quality, scores between 11 and 19 indicated “moderate” quality, and scores under 11 indicated “poor” quality (Methajarunon et al., 2016).
Quantitative outcome synthesis
The socio-demographic and SCI characteristics, sport participation, and health outcomes differed broadly across studies. To capture these diverse characteristics, a summary of the studies, participants, and sports are discussed in the next section, followed by a comprehensive report of all investigated health outcomes. Detailed findings of each included study are also available in Tables 3–5. ‘Favourable’ results refer to primarily positive and significant associations across relevant studies; ‘mixed’ results refer to a mix of associations that did and did not reach significance; and ‘insignificant’ results refer to associations that primarily did not reach significance.
Summary of physical outcomes and relevant findings
Summary of physical outcomes and relevant findings
Summary of psychological outcomes and relevant findings
Summary of social outcomes and relevant findings
For the purposes of this scoping review, outcomes were organized into physical, psychological, and social domains as outlined by Eime et al. (2013). The Health through Sport Conceptual Model, developed by Eime and colleagues, was selected as it considers three crucial elements: (1) determinants of sport participation (e.g., policy, environmental, organizational factors); (2) characteristics of sport itself (i.e., individual, team, organized, informal sport); and (3) health outcomes of sport participation (i.e., physical, psychological, and social domains) (Eime et al., 2013). Not only is this model informed by a systematic review that sought to identify the benefits of sport participation for adults, but this model also considers factors beyond the individual level and is specific to sport (Eime et al., 2013). It is also important to acknowledge that while outcomes in this review were organized into three distinct categories for clarity, the outcomes may realistically fall into more than one domain.
Included studies
A total of 15,965 records were identified through database searching and an additional two records were identified through other sources. Following de-duplication, a pilot screening trial consisting of 20 articles was conducted between the three reviewers, in which an interrater reliability score of 95% was calculated. After passing this screening trial, all 8,470 articles were screened, in which 8,287 articles were excluded and 183 full-text articles were reviewed for eligibility. Following the full-text review stage, 133 articles were excluded. In total, 50 articles were selected for inclusion (Table 1). Two distinct articles (Huonker et al., 1996, 1998) contained the same data but were published in two different journals; thus, 49 unique studies were included. A PRISMA flow diagram (Moher et al., 2009) (Fig. 1) was created to visually depict the article selection process.
Full study characteristics of included studies (n = 49)
Full study characteristics of included studies (n = 49)
PRISMA flow diagram of article selection process.
Scores on the Modified Downs and Black checklist ranged from 7-17/28. Forty-two studies scored between 11 and 19, indicating “moderate” methodological quality. The remaining seven studies scored below 11, indicating “poor” quality. Studies were generally rated high in the Downs and Black checklist ‘Reporting’ section, as the hypotheses, participants, interventions, and findings were often clearly described (Downs & Black, 1998). The checklist items rated lowest were in the ‘External validity’ section, as participants often did not represent the entire source population (i.e., entire population of individuals with SCI) and lacked validation that the sample was representative (Downs & Black, 1998). Checklist items within the ‘Internal validity’ sections were also often rated low, as participants and assessors were not blinded, intervention compliance was unreliable, different groups (i.e., sport groups and non-sport groups) were recruited from different populations, participants were not randomized to the intervention, and sample sizes were not justified (Downs & Black, 1998; Eng et al., 2007). Modified Downs and Black checklist total scores are outlined in Table 1 and detailed scoring of each study is available for reference in the supplementary online material.
Study characteristics
The majority of studies were of cross-sectional design (n = 38, 78%). The remaining studies included cohort studies (n = 7, 14%), a case control study (n = 1, 2%), a non-randomized controlled trial (n = 1, 2%), a case report (n = 1, 2%), and a case series (n = 1, 2%). Included articles were conducted in 20 different countries, with studies most commonly originating from Brazil (n = 13, 27%), Poland (n = 6, 12%), USA (n = 5, 10%), Canada (n = 4, 8%), and Germany (n = 3, 6%). One or two studies originated from other countries including Australia, Japan, Russia, and Italy.
The characteristics of SCI were diverse across studies. Regarding level of injury, studies comprised participants with quadriplegia only (n = 16, 33%), participants with paraplegia only (n = 18, 37%), mixes of participants with quadriplegia or paraplegia (n = 14, 28%), or unreported level of injury (n = 1, 2%). Mean duration of injury for sport participants ranged from 2.1 to 26.5 years, and mean duration of injury for non-sport participants ranged from 1.8 to 25.2 years. The age of most participants did not exceed 50 years; mean ages of sport participants ranged from 22.2 to 49.1 years and the mean ages of non-sport participants ranged from 24.0 to 49.8 years. Studies consisted of mixes of male and female participants (n = 32, 65%), only male participants (n = 14, 29%), or unreported participant sex (n = 3, 6%); there were no studies with only female participants. Gender was not specified in any of the studies. ‘Non-sport’ participants included individuals who did not meet the study’s definition of ‘sport’ or ‘athlete’, or individuals who were inactive.
Regarding sport characteristics, variation in both team and individual sports was observed –common sports included wheelchair rugby, wheelchair basketball, wheelchair racing, and swimming. Mean duration of sport participation ranged from 0.7 to 9.4 years. Many studies did not specify the precise competitive level of sport participants; however, of the studies that did comment (n = 21, 43%), participants competed on international, national, regional, and recreational levels. Full study characteristics are summarized in Table 1. An overview of results across all studies are outlined in Table 2 and a summary of physical, psychological, and social outcomes extracted from each study are available in Tables 3–5. Detailed data extraction entries are available in the supplementary online material.
Overview of results with number of studies contributing to each outcome in brackets
Overview of results with number of studies contributing to each outcome in brackets
Cardiac dimensions and function
Cardiac outcomes at rest were assessed in eight studies (Huonker et al., 1996, 1998; Martins et al., 2018; Matos-Souza et al., 2016; Otsuka et al., 2008; Phillips et al., 2017; Schumacher et al., 2009; Ternovoy et al., 2012; Zamuner et al., 2013); however, findings were generally mixed. Echocardiography (Huonker et al., 1996, 1998; Schumacher et al., 2009), duplex sonography (Huonker et al., 1996, 1998), carotid ultrasonography (Matos-Souza et al., 2016), electrocardiograms (Martins et al., 2018; Otsuka et al., 2008; Zamuner et al., 2013), spiroarteriocardiorhythmography (Ternovoy et al., 2012), and transcranial doppler (Phillips et al., 2017) were used to examine outcomes including cardiac dimensions, stability, and autonomic response. Most studies reported that participants of sport (e.g., wheelchair racing, basketball, rugby) showed favorable outcomes in at least one cardiovascular dimension or volume in sport participants compared to non-sport participants; however, findings between studies conflicted (Huonker et al., 1996, 1998; Schumacher et al., 2009). No consistent differences between sport and non-sport conditions in variables reflecting carotid atherosclerosis (Matos-Souza et al., 2016), cardiac autonomic response (Martins et al., 2018; Otsuka et al., 2008; Zamuner et al., 2013), or cerebrovascular function (Phillips et al., 2017) were seen across studies. Findings from one cross-sectional study however suggested that sport elevates sensitivity of arterial baroreflex and attenuates the increase in diastolic arterial pressure following SCI (Ternovoy et al., 2012).
Respiratory function
Four studies investigated respiratory variables using spirometry (Moreno et al., 2013; Zwiren & Bar-Or, 1975), Stange’s test (Ternovoy et al., 2012), Genchii’s test (Ternovoy et al., 2012), respiratory muscle strength tests (Moreno et al., 2012), and thoracic mobility tests (Moreno et al., 2012), revealing mixed results on the impact of wheelchair rugby or basketball participation. Inconsistent results between sport and non-sport groups were observed for variables including maximal voluntary ventilation, forced vital capacity, and forced expiratory volume after one second (Moreno et al., 2013; Ternovoy et al., 2012; Zwiren & Bar-Or, 1975), however time for Stange’s test and Genchii’s test were higher in sport participants compared to non-sport participants, indicating better respiratory function (Ternovoy et al., 2012). Respiratory muscle strength and thoracic mobility were significantly better in sport participants with quadriplegia compared to non-sport participants, however not for participants with paraplegia (Moreno et al., 2012).
Exercise performance
Eight studies (Bhambhani et al., 1995; Dallmeijer et al., 1997; Davis & Shephard, 1988; Eriksson et al., 1988; Huonker et al., 1996, 1998; Schumacher et al., 2009; Segers et al., 1977; Zwiren & Bar-Or, 1975) showed favourable results regarding sport (e.g., wheelchair racing, basketball, track, swimming) for improving exercise performance. Peak oxygen uptake (̇O2peak mL·kg–1·min–1) during maximal exercise tests of varying design was the most frequently cited outcome. Six studies (Bhambhani et al., 1995; Davis & Shephard, 1988; Eriksson et al., 1988; Huonker et al., 1996; Huonker et al., 1998; Schumacher et al., 2009; Zwiren & Bar-Or, 1975) revealed higher ̇O2peak in sport participants compared to non-sport participants. One cohort study found no significant changes in VO2peak before and after rugby training (once per week for six months), nor any significant differences in other measured variables between sport and non-sport groups (Dallmeijer, Hopman, Angenot, et al., 1997).
In two studies, endurance time (i.e., time until volitional exhaustion) was found to be significantly higher in sport-trained individuals compared to untrained participants (Bhambhani et al., 1995; Davis & Shephard, 1988). Other variables examined during exercise tests, including vital capacity, ventilation, maximal heart rate, stroke volume, power output, and lactate concentrations, revealed mixed results.
Serum profile
The findings of six studies (Dallmeijer, Hopman, & van der Woude, 1997; Hubner-Wozniak et al., 2010; Hubner-Wozniak et al., 2012; Matos-Souza et al., 2016; Sadowska-Krepa et al., 2016; Schumacher et al., 2009) revealed generally favourable results with respect to serum lipid, antioxidant, and haemoglobin profile. Sports included wheelchair rugby, wheelchair racing, and jiu-jitsu. Serum lipid concentrations (Dallmeijer, Hopman, & van der Woude, 1997; Hubner-Wozniak et al., 2010; Matos-Souza et al., 2016; Sadowska-Krepa et al., 2016), blood antioxidant concentrations (Hubner-Wozniak et al., 2012; Sadowska-Krepa et al., 2016), brain-derived neurotrophic factor levels (BDNF) (Sadowska-Krepa et al., 2016), oxidative stress (Sadowska-Krepa et al., 2016), and haemoglobin (Schumacher et al., 2009) were examined. Two studies found significantly higher high-density lipoprotein cholesterol concentrations in sport participants compared to non-sport participants (Dallmeijer, Hopman, & van der Woude, 1997; Hubner-Wozniak et al., 2010); however, two other studies found no group differences (Matos-Souza et al., 2016; Sadowska-Krepa et al., 2016). No consistent findings were seen for other lipid variables (e.g., triglyceride, low-density lipoprotein cholesterol). Mixed results were observed for blood antioxidants including catalase, glutathione reductase, and superoxide dismutase. Lower oxidative stress (Sadowska-Krepa et al., 2016) and higher total haemoglobin mass and concentration (Schumacher et al., 2009) were found in sport participants compared to non-sport participants, but no significant differences were seen in BDNF levels (Sadowska-Krepa et al., 2016).
Body composition and bone mineral density
Three studies showed some favourable results in terms of body composition and bone mineral density, more specifically in the upper body of participants, measured with Dual Energy X-ray Absorptiometry (DEXA) (Goktepe et al., 2004; Gorla et al., 2016), anthropometry (Zwiren & Bar-Or, 1975), and skinfold measurements (Zwiren & Bar-Or, 1975). Sports included basketball, swimming, javelin, and wheelchair rugby. Bone mineral density demonstrated inconsistent results across studies, however bone mineral content and bone area were found to be significantly increased in the sport condition (Goktepe et al., 2004; Gorla et al., 2016). Reductions in adiposity and increases in fat-free mass were observed in various body parts across studies (Gorla et al., 2016; Zwiren & Bar-Or, 1975).
Pain
Two studies demonstrated favourable results regarding the impact of sport (e.g., wheelchair basketball, table tennis, archery) on pain. Sport participants had fewer reports of pain (Kljajic et al., 2016) and lower Chronic Pain Grade Questionnaire scores (Kosmidou et al., 2017) than non-sport participants.
Strength and power
Four studies demonstrated mixed results for sport (e.g., rock climbing, basketball, swimming, javelin) regarding muscle strength and power (Dallmeijer, Hopman, Angenot, et al., 1997; DelGrande et al., 2020; Freitas et al., 2019; Zwiren & Bar-Or, 1975). Significantly higher isometric strength was found longitudinally but not cross-sectionally (Dallmeijer, Hopman, Angenot, et al., 1997). Torque, work, and power of the rotator cuff, measured by a Biodex, were higher in sport individuals compared to non-participating individuals (Freitas et al., 2019). Hand-held dynamometers revealed minimal changes in grip strength following sport participation (DelGrande et al., 2020; Zwiren & Bar-Or, 1975).
Balance and postural control
Two studies revealed insignificant results with respect to balance and postural control (DelGrande et al., 2020; Magnani et al., 2016), measured with the Mini-Balance Evaluation Systems Test (Mini-BESTest) (DelGrande et al., 2020) and electromyography (Magnani et al., 2016). Sports included rock climbing, basketball, and tennis. No significant differences in Mini-BESTest scores or trunk muscle activation were found between sport and non-sport groups.
Physical activity
One study used the Godwin Shephard Leisure Time Physical Activity Questionnaire to investigate physical activity levels. Individuals playing wheelchair rugby had significantly higher physical activity levels than non-sport players (Phillips et al., 2017).
Function and independence
Six studies (Curtis et al., 1986; da Silva et al., 2005; Dos Passos Porto et al., 2016; Furmaniuk et al., 2010; Hanson et al., 2001; Tlili et al., 2008) generally showed favourable results regarding sport for improving function and independence, measured with the Functional Independence Measure (FIM) (da Silva et al., 2005; Dos Passos Porto et al., 2016; Tlili et al., 2008), Modified Barthel Index (Curtis et al., 1986), Wheelchair Skills Test (Furmaniuk et al., 2010), and Craig Handicap Assessment and Reporting Technique (CHART) (Hanson et al., 2001). Results from most measures showed significantly higher function and independence in the sport condition compared to the non-sport condition, however specific sub-sections of the FIM revealed mixed findings.
Healthcare utilization
Findings from three studies demonstrated mixed results with respect to the impact of sport (e.g. basketball, swimming, table tennis) on healthcare utilization (Curtis et al., 1986; Kljajic et al., 2016; Stotts, 1986). Mean number of hospitalizations was found to be significantly lower in sport participants than non-sport participants for one study (Stotts, 1986), but no significant differences were found in a second study (Curtis et al., 1986). Sport participants had significantly fewer physician visits (Curtis et al., 1986) and hospitalizations due to skin breakdown compared to non-sport participants (Stotts, 1986). Presence of complications (e.g., pressure ulcers, urinary infections) and number of hours of attendant care were not different between groups (Curtis et al., 1986; Kljajic et al., 2016).
Nerve or soft tissue injuries
Four studies (Akbar et al., 2015; Antonietti et al., 2008; Dozono et al., 1995; Medina et al., 2015) demonstrated mixed results regarding whether there was a difference in nerve or soft tissue injuries in those who participated in sport (i.e., wheelchair basketball, wheelchair racing, wheelchair rugby) compared to those who did not. Magnetic resonance imaging (Akbar et al., 2015; Medina et al., 2015), nerve conduction studies (Dozono et al., 1995), and the Inquiry of Referred Mobility (Antonietti et al., 2008) were used. Relative risk and prevalence of shoulder injuries yielded inconsistent results across studies (Akbar et al., 2015; Antonietti et al., 2008; Medina et al., 2015). Fewer sport participants had upper extremity peripheral neuropathies or abnormal nerve conduction results compared to non-sport participants (Dozono et al., 1995).
Psychological domain
Quality of life
The impact of sport engagement on quality of life (QOL) was investigated in six studies (Antonietti et al., 2008; DelGrande et al., 2020; Kljajic et al., 2016; McVeigh et al., 2009; Medola et al., 2011; Tlili et al., 2008). Although several different questionnaires were used across studies (i.e., World Health Organizational Quality of Life –Field Trial Version (WHOQOL-BREF) (Antonietti et al., 2008), QOL Profile for Adults with Physical Disabilities (DelGrande et al., 2020), Spinal Cord Injury QOL Questionnaire (Kljajic et al., 2016), Reintegration to Normal Living Index (McVeigh et al., 2009), and 36-Item Short Form Health Survey (SF-36) (Medola et al., 2011; Tlili et al., 2008)), participation in sport (e.g., wheelchair basketball, rock climbing, table tennis, archery, and table tennis) generally showed favourable results with respect to QOL. Results across studies showed significantly higher total QOL scores in the sport condition compared to the non-sport condition, however specific sub-sections of the WHOQOL-BREF and SF-36 revealed mixed findings.
Cognitive function
Three studies revealed mixed results regarding the impact of sport on cognitive function, as measured by the Stroop test (Phillips et al., 2017) or Schulte’s test (Segers et al., 1977; Skucas et al., 2014). The Stroop test was used to assess selective attention and the ability to control conflict interference (Phillips et al., 2017), while the Schulte’s test was used to assess attention span (Segers et al., 1977). Sports included wheelchair rugby, basketball, table tennis, and archery. One study showed no significant differences in cognitive function between groups (Phillips et al., 2017), whereas the other two revealed significantly higher indices of cognitive function in the sport condition compared to the non-sport condition (Segers et al., 1977; Skucas et al., 2014).
Depressive symptoms
Three cross-sectional studies investigated the impact of sport (e.g., basketball, rugby, tennis, fencing, swimming) on depressive symptoms, as measured with the Centre for Epidemiological Studies Depression Scale (CES-D) (Foreman et al., 1997), Questionnaire for Depression (QD) (Gioia et al., 2006), and Middlesex Hospital Questionnaire (MHQ) (Monnazzi, 1982). The results were mixed. One study that used the CES-D to measure depressive symptoms showed no significant differences between groups (Foreman et al., 1997); however, two studies that used the QD and depression sub-section of the MHQ revealed lower levels of depressive symptoms in sport participants than non-sport participants (Gioia et al., 2006; Monnazzi, 1982).
Anxiety
The same three studies discussed above also demonstrated support for sport’s influence on reducing anxiety, measured by the State Trait Anxiety Inventory (Foreman et al., 1997; Gioia et al., 2006) and MHQ (Monnazzi, 1982). Two studies revealed lower anxiety in the sport group compared to the non-sport group (Gioia et al., 2006; Monnazzi, 1982), and one study showed a trend towards lower anxiety in sport participants (Foreman et al., 1997).
Life satisfaction
One cross-sectional study revealed somewhat favourable results in terms of the impact of wheelchair rugby on life satisfaction, measured with the Life Satisfaction Questionnaire (LiSat-9) (Zwierzchowska et al., 2017). High-point athletes (i.e., those in sport classes 2.0–3.5) showed significantly greater satisfaction with the self-care abilities component of the LiSat-9 compared to non-sport participants; however, no significant differences were found in the other eight components (Zwierzchowska et al., 2017).
Self-efficacy
The results of one cross-sectional study examined the influence of sport on self-efficacy (Adnan et al., 2001). Individuals participating in wheelchair rugby had higher scores on the Transferring sub-section of the Self-Efficacy Scale for Activities of Daily Living (SEADL) and higher scores for all items of the Self-Efficacy Scale for Quad Rugby Skills, in comparison to non-sport participants. No significant differences were observed in other sub-sections of the SEADL (e.g., dressing, eating, navigating ramps).
Sexual adjustment and self-esteem
The results of one study showed a positive association between sport and sexual adjustment and self-esteem, measured with the Physical and Sexual Esteem Questionnaire (Dos Passos Porto et al., 2016). Sport participants who engaged in wheelchair basketball or handball demonstrated significantly higher sexual adjustment scores and higher sexual and body esteem scores in comparison to those who did not participate (Dos Passos Porto et al., 2016).
Resilience
The same study as described above did not show results favouring the use of sport for improving resilience (Dos Passos Porto et al., 2016). Use of the Resilience Scale revealed no significant difference in scores between sport and non-sport groups (Dos Passos Porto et al., 2016).
Social domain
Employment
The results of four cross-sectional studies (Blauwet et al., 2013; Curtis et al., 1986; Hanson et al., 2001; Tasiemski et al., 2000) investigating the impact of sport participation (e.g., basketball, tennis, rowing, swimming, wheelchair racing, and archery) on employment were inconsistent, as measured using frequency counts (Blauwet et al., 2013; Tasiemski et al., 2000), hours per week of involvement (Curtis et al., 1986), and the CHART (Hanson et al., 2001). One study (Blauwet et al., 2013) revealed significantly higher employment in sport participants than their non-sport counterparts through multivariable logistic regression; however, results in the univariable model were insignificant. One study showed no significant correlation (Tasiemski et al., 2000), while another showed a trend towards more weekly involvement in employment in sport participants compared to non-sport participants (Curtis et al., 1986). The CHART revealed inconsistent findings in the Occupation Dimension and the Economic Self-Sufficiency Dimension (Hanson et al., 2001).
Education
Two studies (Curtis et al., 1986; Tasiemski et al., 2000) examined the association between sport (e.g., basketball, swimming, wheelchair racing, and archery) and education status. Hours per week of involvement (Curtis et al., 1986) and frequency counts (Tasiemski et al., 2000) were used to assess this outcome. Neither study demonstrated that sport participation significantly improves education status.
Community integration
According to two studies (Hanson et al., 2001; McVeigh et al., 2009), sport participation was positively associated with community integration. The Community Integration Questionnaire (CIQ) (McVeigh et al., 2009) and Social Integration Dimension of the CHART (Hanson et al., 2001) were used. Participants who engaged in sport (e.g., in wheelchair racing or wheelchair basketball) had significantly higher community integration scores than those who did not play sport.
Intensity of psychosocial problems
One study revealed favourable results regarding sport participation on the intensity of subjectively experienced psychosocial problems (Morgulec-Adamowicz et al., 2011). The Witkowski Psychosocial Problems Spinal Cord Injury Scale (PP-SCI) involved participants ranking items grouped within personality, family, social contacts, and occupational sub-sections, using a 6-point scale (“it does not concern me” to “the problem intensity is very high”). Those who played wheelchair rugby demonstrated lower mean global and sub-section scores on the PP-SCI than those who did not play, signifying lower intensity of psychosocial problems (Morgulec-Adamowicz et al., 2011).
Discussion
This scoping review systematically summarized existing literature on the impact of sport participation on the physical, psychological, and social health of individuals living with SCI. According to the findings of this review, regular participation in sport was found to have a positive impact on various aspects of physical health (e.g., exercise performance, body composition, function and independence), psychological health (e.g., QOL, life satisfaction, anxiety levels), and social health (e.g., community integration, intensity of psychosocial problems). Inconsistent results were found with respect to sport’s impact on cardiac or respiratory function, strength and power, healthcare utilization, nerve or soft tissue injuries, cognitive function, depressive symptoms, and employment status. Results from this scoping review did not reveal significant associations between sport participation and balance and postural control, resilience, or education status. Many different types of sport were investigated across studies (e.g., wheelchair rugby, wheelchair basketball, swimming, wheelchair racing, archery) and participants engaged in sport at varying competitive levels (e.g., recreational, competitive national). Quality of studies ranged from poor to moderate. This review provides preliminary evidence to support the use of sport as a lifestyle component as well as a potential intervention to address some of the challenges that people with SCI experience.
Participation in sport offers unique benefits for individuals living with SCI. For instance, literature examining the effects of exercise (i.e., non-sport LTPA) for people with SCI have investigated similar physical outcomes as the studies investigating sport participation (Hicks et al., 2011; van der Scheer et al., 2017). These reviews showed improvements in overall fitness, alike to the findings for sport in this scoping review; however, there was insufficient evidence to conclude that exercise improves function (Hicks et al., 2011; van der Scheer et al., 2017). Conversely, this scoping review, demonstrates that sport may be a useful intervention for increasing function and independence (Curtis et al., 1986; da Silva et al., 2005; Dos Passos Porto et al., 2016; Furmaniuk et al., 2010; Hanson et al., 2001; Tlili et al., 2008). Sport may offer greater improvements in function as it is an activity that occurs in a less controlled environment and requires more spontaneity and adaptation, which is characteristic of many functional tasks. In a systematic review investigating the impact of LTPA on QOL, almost all studies defining LTPA as sport participation showed favourable results for QOL, whereas the other interventions falling under LTPA (i.e., treadmill training, arm ergometry, resistance training) revealed mixed results (Tomasone et al., 2013). Given the social nature of sport, it is not surprising that individuals engaging in sport are more consistently experiencing improvements in QOL. Finally, this scoping review of quantitative studies complements the findings of a qualitative thematic synthesis exploring sport participation for those living with SCI (Cheung et al., 2021). Greater emphasis on psychological and social impacts of sport were highlighted in the qualitative synthesis, whereas most studies in this scoping review investigated physical outcomes of sport participation (Cheung et al., 2021). Interestingly, the qualitative synthesis reported that sport facilitated the resilience process for participants (Cheung et al., 2021), whereas this quantitative synthesis showed no impact of sport on resilience, although this was demonstrated by only one study (Dos Passos Porto et al., 2016).
Modified Health through Sport Conceptual Model, adapted from Eime and colleagues (Eime et al., 2013).
The findings from this review may be used to enhance the existing Health through Sport Conceptual Model (Eime et al., 2013). Strengths of the current model included its relatively recent development (i.e., within the past ten years), its detailed consideration of various relevant factors (e.g., health systems, policy, environment), and its specificity to sport. The purpose of the systematic review, which informed the model, also paralleled that of the current scoping review (i.e., to report outcomes of sport participation). However, several limitations of this model must be acknowledged. Three key considerations were excluded from the systematic review and resulting conceptual model: (1) adapted sport, (2) elite sport, and (3) challenges of sport. Firstly, this scoping review highlighted many physical, psychological, and social outcomes resulting from sport participation specifically for individuals living with SCI (i.e., adapted sport participants, individuals with disabilities). The refined conceptual model should consider adapted sport in addition to non-adapted sport. Furthermore, the findings from this review demonstrated the impact of sport across competitive levels –ranging from recreational- to elite-level sport. Thus, the enhanced model should consider recreational as well as elite sport. Lastly, the systematic review informing the Health through Sport Conceptual Model only reported the benefits of sport participation, whereas this scoping review reported outcomes that included benefits and challenges associated with sport. The updated model, therefore, should consider both the benefits and challenges of sport participation. These three considerations have been captured in the refined Health through Sport Conceptual Model (Fig. 2), which may be applied to a broader range of individuals, including those with disabilities.
The mean age of participants in all studies included in this scoping review were under 50 years old. This is a notable finding, as the average age of onset of SCI in Canada is increasing (i.e., the majority of individuals receiving rehabilitation are over 50 years of age) (Bickenbach et al., 2013; Craven et al., 2012; Noonan et al., 2012), meaning the impact of sport on a large portion of the SCI population has not been studied. Furthermore, the average life expectancy of living with SCI has also been increasing (McColl et al., 1997; “SPINAL CORD INJURY FACTS AND STATISTICS,” 2005). Thus, more individuals are ageing with SCI and experiencing the health challenges associated with ageing and SCI (Groah et al., 2012; Hitzig et al., 2011). Despite the existing literature supporting sport participation for the older able-bodied adult (Gayman et al., 2017; Jenkin et al., 2017; Stenner et al., 2020), no research to date has examined sport engagement for ageing adults with SCI. A recommendation for future research therefore will be to explore the experiences and perceptions of individuals aged 50 years or greater with SCI on sport.
Studies in this review varied greatly with respect to the types of sport investigated. Sports included consisted of team sport (e.g., wheelchair basketball, wheelchair rugby, softball, curling) and individual sport (e.g., wheelchair racing, swimming, fencing, archery), and ranged across competitive levels (e.g., recreational, competitive regional, competitive national, competitive international). As per our definition of the concept “sport”, all sports were types of physical activity, involved exertion, skill, and/or hand-eye coordination, contained elements of competition, and existed through organizations (Pink, 2008). A noteworthy element of this definition is that for an article to have been included, the sport(s) must have incorporated some aspect of competition. This was a critical and unique element of consideration, as published literature in able-bodied sport has revealed that competition in sport may be associated with greater motivation and commitment in participants, but also potentially increased risks of burnout and injury, including acute cardiovascular events (Clancy et al., 2016; Corrado et al., 2011; Ruwald et al., 2015). In this scoping review, the benefits of sport were evident across competitive levels, signifying that participation in sport, on any competitive level, can result in positive outcomes for those with SCI. However, given the broad variation in sport types and competitive levels observed across articles, in addition to the fact that many articles (n = 18) investigated multiple sports within one given study, we were not able report outcomes specific to the type of sport nor competitive level. An interesting direction for future research, therefore, would be to conduct sport-specific research and analyze data regarding the competitive level of adapted sport participants.
Studies included in this review ranged from poor to moderate quality. External validity and internal validity were rated lower than the reporting of the studies. It is important to acknowledge that the lower the quality of included studies, the greater the likelihood for bias in the findings of each individual study, regardless of the rigour used in conducting the review (Elkins et al., 2010). Given that the majority of articles were of moderate quality, readers are encouraged to keep this consideration in mind when interpreting and contextualizing the results from this scoping review. To improve the external validity of future studies, it is recommended that researchers recruit participants who are representative of the entire SCI population (e.g., representative of age and sex distributions in SCI) and comment on the staff and facility where sport engagement took place (e.g., consider whether the facility is representative of all adaptive sport facilities) (Eng et al., 2007). To improve internal validity, it is recommended to blind assessors measuring main outcomes, track compliance of participants with sport participation, account for participants lost to follow-up, and justify participant sample size (Eng et al., 2007).
Included studies were limited to sports occurring for a minimum duration of three months; therefore, it is possible that studies with significant results satisfying all other eligibility criteria were excluded from this review. Although the number of studies included in this review is likely lower for this reason, we believe this exclusion criterion was appropriate and will help justify dosage recommendations for future sport interventions, as three months is the approximate length of a typical sport season and is the minimum timeframe during which measurable cardiovascular adaptations in the outcomes included occur (Golbidi & Laher, 2012). Additionally, we did not include studies consisting of mixed populations where data specific to SCI were unable to be extracted. Although we may have missed reporting outcomes that were studied, at least in part, with the SCI population, we were comfortable with this decision as our objective was to report the impact of sport participation specifically for those living with SCI. Finally, due to the heterogeneity of sport characteristics between and within studies, we were not able to examine the impact of specific sports or competitive levels on outcomes, nor able to examine the associations between dosage of sport engagement and health (e.g., correlation between duration of sport participation and QOL). Sport-related dosage data was not available for 94% of studies and therefore setting a clear comparison between sport and non-sport conditions was used to help state plainly the differences between sport involvement and non-sport involvement.
Conclusions
This scoping review revealed how sport participation may play a role in improving physical, psychological, and social health in individuals living with SCI. This synthesis provided a comprehensive report of outcomes resulting from regular participation in sport outside the laboratory or rehabilitative settings. The findings presented in this review are useful to inform improved quality of research in this specific area, particularly future studies investigating sport for those with SCI.
Conflict of interest
The authors report no conflicts of interest.
Funding
This research was funded by the Ontario Graduate Scholarship, the Toronto Rehabilitation Institute Student Scholarship, and the Unilever/Lipton Graduate Fellowship to LC, as well as funding from the Canadian Institutes of Health Research and an Ontario Early Researcher Award to KEM.
Supplementary materials
The supplementary materials are available from https://dx-doi-org-s.web.bisu.edu.cn/10.3233/NRE-220037
Footnotes
Appendix: Search strategy for Medline.
| # Search Strategy: Ovid MEDLINE ALL <1946 to January 19, 2021> |
| 1 [Population: SCI &Physically Disability] |
| 2 exp Spinal Cord Injuries/ (49481) |
| 3 exp Paraplegia/ (13102) |
| 4 exp Quadriplegia/ (8107) |
| 5 Wheelchairs/ (4788) |
| 6 disabled persons/ and (physical* disabled or physical* disabilit* or wheelchair*).tw,kw. (3260) |
| 7 para-athletes/ (24) |
| 8 (spinal cord adj3 (injur* or contusion* or trauma* or transection* or lacerat* or compression* or lesion* or paraly*)).tw,kw. (54340) |
| 9 (paraplegi* or quadriplegi* or tetraplegi*).tw,kw. (24310) |
| 10 (myelopath* adj2 (traumatic or post-traumatic or post traumatic or compressive)).tw,kw. (719) |
| 11 ((spinal cord injur* or SCI) adj3 post*).tw,kw. (1604) |
| 12 ((disabled or disabilit* or handicap* or physically challenged) and (wheelchair* or athlete*)).tw,kw. (2690) |
| 13 (para athlete* or para-athlete* or paralympi* or wheelchair athlete*).tw,kw. (957) |
| 14 (spinal adj3 paraly*).tw,kw. (713) |
| 15 wheelchair*.tw,kw. (7549) |
| 16 or/2-15 (103148) |
| 17 [Concept: Sports] |
| 18 exp sports/ (187691) |
| 19 sports for persons with disabilities/ (214) |
| 20 Sports Medicine/ (10958) |
| 21 Psychology, Sports/ (43) |
| 22 (wheelchair* adj3 (tournament* or basketball or racing or road race* or tennis or curling or rugby or athele* or participat* or participant* or sport*)).tw,kw. (787) |
| 23 (sport* adj4 (community* or competiti* or individual* or participat* or participant* or group* or organiz* or involve* or activ* or team* or club* or adaptiv* or para* or wheelchair* or disabilit* or disabled)).tw,kw. (25961) |
| 24 or/18–23 (208753) |
| 25 16 and 24 (4720) |
| 26 remove duplicates from 25 (4707) |