Introduction
1. Background
Individuals with spinal cord injury (SCI) experience limited physical activity (PA) and face significant challenges due to nerve damage. Engaging in PA offers substantial benefits by improving health-related physical fitness, such as cardiopulmonary capacity and muscle strength [1,2], enhancing motor and sensory functions, preventing secondary complications, and developing essential skills for daily living [3]. Several studies have also demonstrated that regular PA can improve energy metabolism and promote bodily adaptation, gradually reactivating weakened motor functions [4,5]. Therefore, accurately determining appropriate PA levels, continuously monitoring progress, and promoting improvements in physical strength are crucial steps in preventing chronic diseases and premature death among individuals with SCI [6].
To measure PA, questionnaires like the Korean version of the International Physical Activity Questionnaire (K-IPAQ) [7,8] categorize activities into work (e.g., hard construction work, climbing stairs), transportation (e.g., bus, subway), housework (e.g., splitting wood, shoveling snow), leisure activities (e.g., running, fast cycling, inline skating), and sitting time. Although the K-IPAQ has been validated and is widely used in Korea, its items and examples are designed for individuals with a sedentary lifestyle and may not be applicable to most individuals with SCI. The Leisure Time Physical Activity Questionnaire for People with SCI (LTPAQ-SCI) [9] is available; however, it focuses solely on leisure activities, complicating the assessment of overall PA that includes work and transportation. Moreover, these instruments do not reflect unique cultural and lifestyle factors in Korea, such as the availability of activity support and rehabilitation services tailored to the country’s specific context.
Activity support services play a crucial role in assisting individuals with disabilities by providing tailored support based on their needs and preferences [10]. Typically, activity assistants deliver these services. However, in Korea these services tend to focus predominantly on housework and care tasks [11], which may unintentionally influence the PA levels of individuals with disabilities despite the aim of promoting independent living. Furthermore, many individuals with disabilities in Korea continue to receive rehabilitation services at medical facilities after discharge, benefiting from National Health Insurance coverage. This ongoing access to rehabilitation services may also affect their PA levels.
2. Objectives
Therefore, this study investigates the PA levels of Korean individuals with SCI using a modified version of the K-IPAQ that accounts for the specific characteristics of SCI. Since the World Health Organization (WHO) recommends moderate to vigorous intensity PA, we categorized activity levels into low, medium, and high intensities. Additionally, because daily PA (such as work, leisure, and transportation) beyond structured exercise is important, we analyzed activity levels based on individual factors (e.g., sex, injury level) that significantly influence PA. The modified questionnaire was validated through a pilot test (Suppl. 1, 2). This study aims to provide insights into the PA patterns of this demographic in Korea.
Methods
1. Ethics statement
This study was conducted following a thorough review and approval by the Korea University Institutional Review Board (IRB) (KUIRB-2023-0167-01). Informed consent was obtained from all participants and, where applicable, their legal guardians.
2. Study design
This study employed a cross-sectional survey design and was reported in accordance with the STROBE guidelines (available at https://www.strobe-statement.org/).
3. Setting
The study was conducted among individuals with SCI in metropolitan areas of Korea, recruited through the Korean Spinal Cord Injury Association. Two experienced researchers—one with a PhD and the other with a master’s degree—administered the questionnaire in a one-on-one format. Participants were contacted by phone 7 days prior to the survey, which took place between June and September 2023, and were asked to recall their PA over the previous week. Most surveys were completed in approximately 15 minutes, although some responses were incomplete. Weekly PA was quantified in metabolic equivalent task-min/week (MET-min/week) by multiplying the reported PA by the average time provided in the questionnaire. Data on general factors such as sex, job, type of paralysis, and injury level were self-reported at the beginning of the questionnaire.
4. Participants
Between April and June 2023, 185 individuals with SCI, aged 18 to 64, were recruited from metropolitan areas in Korea. Participants were selected if they had not experienced a recent SCI that could impair motor function within the past 6 months, ensuring a stable SCI status. Individuals with a previously stabilized SCI were excluded if they had a new SCI or relevant symptoms in the past 6 months. Exclusion criteria also included comorbidities such as cardiovascular disease, metabolic syndrome, and difficulties understanding the questionnaire. Recruitment was carried out through the Korean Spinal Cord Injury Association. Prior to the study, participants were visited in person and provided with a comprehensive explanation of the study’s objectives, procedures, safeguards for their rights, and details regarding the study’s reliability and stability. Those who agreed to participate provided written informed consent. Ultimately, 176 individuals with SCI who met the criteria and provided consent were included in the final analysis; 9 participants were excluded due to incomplete responses in one category or repeated participation (Figure 1).
5. Variables
The primary outcome was the level of PA measured in MET-min/week. Secondary outcomes included PA values categorized into 5 types and 3 intensity levels.
6. Data sources/measurement
The K-IPAQ was selected as the base measurement tool and modified through an advisory meeting to better capture the specific characteristics of SCI. These modifications aimed to simplify PA recording and facilitate the calculation of overall PA. The questionnaire was divided into 2 sections: one for collecting general information (e.g., sex, job, type of paralysis, injury level, and severity) and another for recalling and self-reporting PA over the past week. The original K-IPAQ PA items (work, transportation, housework, leisure time, and sitting time) were reclassified into 5 categories (work, leisure, physical therapy, housework, and transportation using a manual wheelchair) to better reflect the experiences of individuals with SCI, with each item assigned an intensity level. Research data are available in Data Set 1.
A pilot test was conducted with 10 individuals with SCI to validate the modified questionnaire (Suppl. 1, 2). Participants were instructed to report the days and hours they engaged in PA for at least 10 minutes in each of the 5 categories over the previous week. The results showed a statistically significant positive correlation (r=0.710, P=0.000) between the high-intensity PA levels measured by the questionnaire and those recorded by an accelerometer (GENEActiv; Activinsights Ltd., Cambridgeshire, UK), demonstrating the reliability of the questionnaire [12].
7. Bias
Sampling was non-randomized, as only individuals who agreed to participate were included. Although there was no researcher-induced selection bias, the sampling method may have introduced bias.
8. Study size
The sample size was not predetermined, as the study aimed to include the entire eligible population that met the inclusion and exclusion criteria and consented to participate in the survey.
9. Statistical methods
Data were analyzed using IBM SPSS for Windows ver. 27.0 (IBM Corp., Armonk, NY, USA). The Shapiro-Wilk test was initially performed to assess data distribution. Since the sample data did not meet the normality assumption due to varying participant numbers across factors, nonparametric methods were used for subsequent analyses. PA participation rates were compared using the chi-square test, while PA amounts were analyzed using the Wilcoxon signed-rank test. Analyses by the 5 categories and 3 intensity levels were conducted only for the non-athlete group. Comparisons between athletes and non-athletes were performed using the independent samples t-test. All tests were two-sided with a significance level set at P<0.05, and results are expressed as mean±standard error of the mean.
Results
1. Participants
The general characteristics of the participants, including their injury type and level, are presented in Table 1.
2. Main results
1) The PA participation rate by group
(1) Sex: The analysis of the questionnaire data revealed significant sex differences in participation rates for housework (P=0.010) and transportation (P=0.005) among the 5 PA categories (Table 2).
(2) Injury level: Injury levels were categorized as cervical, thoracic, and lumbar to reflect differences in paralysis types and their impact on wheelchair selection. Due to the limited number of participants with lumbar injuries and the similarity in lower extremity paralysis between lumbar and thoracic injuries, the lumbar group was combined with the thoracic group. The analysis focused on injury level rather than the type of paralysis, as self-reported data occasionally resulted in individuals with cervical injuries mistakenly classifying themselves as paraplegic. Among the 5 PA categories, significant differences in participation rates for housework (P=0.001) and transportation (P=0.000) were observed between participants with thoracic and cervical injuries (Table 2).
(3) The use of activity support services: Regarding activity support services, significant differences were observed in participation rates for transportation (P=0.006). Specifically, 87.3% (n=55) of participants who did not use activity support services engaged in transportation, compared to 67.3% (n=76) of those who did use such services, indicating higher transportation participation among non-users (Table 2).
2) The amount of PA by category and intensity
Of the 176 participants, 50 were athletes—a proportion significantly higher than that typically observed in the general SCI population. Consequently, the PA survey results from this study may not be representative of the overall SCI population. As a result, we compared the total PA of the athlete group (ATH group) and the non-athlete group (NATH group).
The analysis showed that the ATH group (n=50) had a PA level of 5,661.6±385.4 MET-min/week (Suppl. 3, Figure 2), while the NATH group (n=126) had a PA level of 2,907.2±194.8 MET-min/week. This significant difference (P=0.000) indicates a higher PA level in the ATH group compared to the NATH group. Therefore, the ATH group was excluded from further analysis to obtain a more accurate representation of the PA levels among individuals with SCI in Korea. Suppl. 3 presents the rate of PA participation among athletes, the PA amounts by the 5 categories, and the PA amounts by group/intensity.
The PA values, categorized into 5 types, were as follows: work (n=17), 1,948.0±377.4 MET-min/week; leisure (n=96), 1,747.6±141.0 MET-min/week; physical therapy (n=33), 661.8±78.2 MET-min/week; housework (n=51), 869.6±58.7 MET-min/week; and transportation (n=89), 918.8±50.7 MET-min/week. The number of participants varied across categories because those who did not engage in a particular category were excluded from the analysis. For example, among participants with jobs, only those who engaged in low-, medium-, or high-intensity PA in their occupational activities were included. Among these categories, PA during work was the highest, followed by leisure, transportation, housework, and physical therapy (Figure 2A). Additionally, 35.7% (n=45) of participants engaged in muscle exercise during leisure activities, achieving a PA level of 212.0±17.2 MET-min/week for muscle exercise.
The PA values, categorized by 3 intensities, were as follows: high-intensity activity (work and leisure, n=20), 2,760.0±280.5 MET-min/week; moderate-intensity activity (work, leisure, and physical therapy; n=75), 1,148.8±89.0 MET-min/week; and low-intensity activity (work, leisure, housework, and transportation; n=111), 1,869.4±110.3 MET-min/week. The highest PA amount was observed for high-intensity activities, followed by low-intensity and then moderate-intensity activities (Figure 2B).
3) The amount of PA by group
(1) Sex: Significant sex differences were found in the housework category (P=0.007). Men recorded 739.2±74.7 MET-min/week in housework, while women recorded 1,056.0±80.1 MET-min/week, indicating higher PA levels for women in this category (Figure 3A). Although men demonstrated a higher amount of PA in work-related activities than women, this difference was not statistically significant—likely due to the small number of female participants engaged in work. No significant differences were observed in the other categories. The PA levels were as follows: work (men, 2,063.6±419.1; women, 1,089±297.0; MET-min/week; P=0.618), leisure (men, 1,790.1±165.4; women, 1,626.7±273.3; MET-min/week; P=0.553), physical therapy (men, 720.0±114.7; women, 560.0±74.3; MET-min/week; P=0.839), and transportation (men, 933.4±58.4; women, 864.9±102.9 MET-min/week; P=0.704).
(2) Injury level: Although the cervical injury group exhibited higher PA levels than the thoracic group in the work, housework, and transportation categories, these differences were not statistically significant. The PA levels were as follows: work (cervical, 2,049.0±541.3; thoracic, 1,894.4±521.0; MET-min/week; P=0.660), housework (cervical, 940.5±144.4; thoracic, 847.8±63.4; MET-min/week; P=0.404), and transportation (cervical, 931.3±87.7; thoracic, 913.4±62.5; MET-min/week; P=0.918). In the leisure and physical therapy categories, the thoracic group demonstrated higher PA levels than the cervical group; however, these differences were also not statistically significant. Specifically, leisure PA was 1,560.6±197.6 MET-min/week for cervical injuries versus 1,875.5±194.8 MET-min/week for thoracic injuries (P=0.374), and physical therapy PA was 573.3±101.1 MET-min/week for cervical injuries versus 768.0±119.9 MET-min/week for thoracic injuries (P=0.100) (Figure 3B).
(3) The use of activity support services: Based on the use of activity support services, there was a significant difference in PA levels in the housework category (P=0.010). Participants who used the service (n=28) engaged in housework activities at a level of 1,004.1±73.4 MET-min/week, whereas those who did not use the service (n=23) recorded 705.9±84.3 MET-min/week, indicating greater engagement in housework among service users (Figure 3C). No significant differences were observed in other categories. The PA levels were as follows: work (service users, 1,484.2±378.0; non-users, 2,801.0±736.9; MET-min/week; P=0.149), leisure (service users, 1,722.7±190.8; non-users, 1,789.0±203.8; MET-min/week; P=0.515), physical therapy (service users, 540.0±79.6; non-users, 849.2±145.4; MET-min/week; P=0.087), and transportation (service users, 1,006.2±63.5; non-users, 811.8±79.5; MET-min/week; P=0.078). In the work category, although PA was higher among those not using the service (n=6) compared to those using the service (n=11), this difference was not statistically significant—likely due to the small sample size in the non-user subgroup.
Discussion
1. Key results
The analysis demonstrated that the PA level for the ATH group (n=50) was 5,661.6±385.4 MET-min/week (Suppl. 3, Figure 2), while the NATH group (n=126) recorded 2,907.2±194.8 MET-min/week. After excluding the ATH group, among the 5 categories, work-related PA was the highest, followed by leisure, transportation, housework, and physical therapy. Furthermore, when categorized by intensity, the PA levels of individuals with SCI were as follows: high-intensity activity (work and leisure) was 2,760.0±280.5 MET-min/week, moderate-intensity activity (work, leisure, and physical therapy) was 1,148.8±89.0 MET-min/week, and low-intensity activity (housework and transportation) was 1,869.4±110.3 MET-min/week.
2. Interpretation/Comparison with previous studies
Our findings indicate that the PA levels of individuals with SCI in Korea appear higher than those reported in surveys from other countries. For example, Rosenberg et al. [13] reported that individuals with SCI in the United States (n=411) had a PA level of 2,340.0±2,980.0 MET-min/week, while Bjørgen [14] found that Norwegian SCI participants (n=185) had a PA level of 2,274.88±2,819.33 MET-min/week. In contrast, our study determined that Korean SCI participants had a PA level of 2,907.2±194.8 MET-min/week.
Several previous studies have reported that males with disabilities generally exhibit higher PA levels than females [15,16] and that individuals with thoracic injuries typically engage in more PA than those with cervical injuries [17]. In contrast, our study found no significant differences in PA based on sex or between cervical and thoracic injuries. This result is consistent with Bjørgen [14], which reported no significant differences in PA by sex or paralysis type (tetraplegia versus paraplegia), and with de Groot et al. [18], who observed no significant PA difference between cervical (48.7±20.4 MET time/day) and thoracic injuries (56.0±31.0 MET time/day). These discrepancies suggest that participant demographics, survey methodology, and the timing of questionnaire administration may influence PA outcomes.
Although previous research has typically focused on overall sex differences in PA, our study examined PA across 5 distinct categories. We observed that females engaged more frequently in housework, resulting in higher PA levels in that category compared to males. Conversely, males demonstrated greater activity in transportation using a manual wheelchair. Psychosocial factors such as differing interests and motivations may underlie these sex disparities in PA [19]. Glauber [20] noted that men often prioritize external work, whereas women tend to focus on family-related tasks. Our findings align with these observations, suggesting that social factors may play a more significant role than disability-related factors in determining PA levels, similar to patterns observed in studies of non-disabled individuals.
Conversely, no significant differences were observed in the categories of work, leisure, and physical therapy. This finding is in line with Saebu and Sørensen [21], who found no association between sex and PA levels among young individuals with disabilities (including cerebral palsy, SCI, and muscle disease; n=998). They suggested that activity levels are primarily constrained by mobility restrictions and environmental factors that limit PA opportunities for individuals with disabilities, leading to no significant sex difference [21]. These results underscore the challenges individuals with disabilities face when engaging in activities that require movement outside the home.
While Bjørgen [14] and de Groot et al. [18] reported no differences in PA levels based on injury level or paralysis type, our study provided a more detailed evaluation by categorizing PA participation into 5 groups. Unlike previous studies that grouped participants by type of paralysis, we classified them by the level of injury. Our analysis revealed that individuals with thoracic injuries engaged significantly more in housework and transportation than those with cervical injuries. Nevertheless, the overall PA levels did not differ significantly between the 2 groups. Although individuals with cervical injuries participated less frequently in these specific activities, their activity levels were comparable to those of individuals with thoracic injuries when they did participate.
Time since injury (TSI) represents the duration since an SCI occurred, and along with age, it plays a critical role in influencing PA levels among individuals with SCI. Generally, as age and TSI increase, both PA levels and wheelchair-related capabilities tend to decrease [22]. TSI appears to have a more pronounced effect on individuals with paraplegia than on those with tetraplegia [22]. Perrier et al. [23] studied SCI individuals in Canada (average age, 46.8±13.4 years; average TSI, 15.2±11.1 years) and observed that individuals with injuries at C5–C7 engaged in more PA (142.7±158.2 minutes) than those with injuries at T1–S5 (114.0±137.7 minutes) [23]. This indicates that age and TSI significantly impact PA levels, with variations depending on injury level and type.
In our study, individuals with cervical injuries had an average age of 50.7±1.8 years and an average TSI of 17.7±1.4 years, whereas those with thoracic injuries had an average age of 55.3±1.4 years and an average TSI of 22.4±1.5 years. These findings indicate that participants with thoracic injuries were generally older and had a longer duration since injury than those with cervical injuries. Moreover, the average age of our participants far exceeds the 30-year threshold recommended by de Groot et al. [18] for evaluating PA levels. It is plausible that if individuals with thoracic injuries were younger and had a shorter TSI, their PA levels might be higher than those with cervical injuries. Given that individuals with thoracic injuries typically enjoy better mobility, future studies should carefully consider the effects of age and TSI when assessing PA levels among individuals with SCI.
In Korea, activity support services assist individuals with disabilities in living independently by providing support ranging from 2 hours to full-time care based on disability levels. Our analysis indicates that those receiving these services engaged in significantly more housework-related PA than those who did not receive support, whereas individuals without these services participated more in transportation-related activities. No significant differences were observed in PA levels for work, leisure, or physical therapy. This pattern may be related to our earlier sex-specific analysis, which revealed higher participation in housework among females. In our study, 54.4% (n=49) of male participants and 80.5% (n=29) of female participants utilized activity support services. The lower usage rate among men may be attributed to a tendency to prioritize external work [20] and to rely on family members for housework, reflecting social roles and expectations.
Finally, the PA levels of individuals with SCI in Korea were evaluated against WHO recommendations, which suggest a PA level between 600 and 1200 MET-min/week. Our findings indicate that 48.4% (n=61) of participants exceeded the minimum recommended level of 600 MET-min/week by incorporating high- and moderate-intensity activities. Additionally, 37.3% (n=47) surpassed the upper threshold of 1200 MET-min/week, demonstrating that approximately half of the participants achieved an adequate PA level. Despite this, less than half of individuals with SCI in Korea participate at the level recommended by the WHO, highlighting an urgent need for interventions—such as establishing rehabilitation facilities and developing tailored exercise programs—to help these individuals achieve and maintain appropriate PA levels.
3. Limitations
First, although the questionnaire was validated through a pilot test, the relatively small sample size limits the full assertion of its validity, necessitating further research to confirm its reliability. Second, classifying activities into high, moderate, and low-intensity based on the Rating of Perceived Exertion introduces subjectivity, which could affect the accuracy of the activity level assessments. Third, the survey’s geographic limitation and the small number of participants with lumbar injuries may result in only a partial representation of the PA levels of all individuals with SCI in Korea. Fourth, athletes were excluded from the analysis because their PA levels were deemed unrepresentative of the general SCI population. Finally, reliance on participants’ recall of activities over the past 7 days may have led to an overestimation of PA duration and intensity. These limitations should be considered when interpreting the study’s results.
4. Generalizability
Because this study was conducted exclusively in metropolitan areas, generalizing the PA levels to all individuals with SCI in Korea would be challenging. Nonetheless, the findings provide valuable baseline data.
5. Suggestions for further studies
Future research should concentrate on more clearly establishing the reliability and validity of the questionnaires, as well as expanding the study to include a nationwide sample.
This study on PA among individuals with SCI in metropolitan Korea revealed that their average daily PA levels were higher than those reported in studies from other countries and exceeded the WHO’s recommendations. However, the fact that only about half of the participants met the recommended PA levels highlights the need for more proactive measures. Developing targeted programs to help individuals with SCI initiate and sustain adequate levels of PA is essential.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement1
