Skip Navigation LinksHome > February 2006 - Volume 38 - Issue 2 > The Physical Activity Recall Assessment for People with Spin...
Medicine & Science in Sports & Exercise:
doi: 10.1249/01.mss.0000183851.94261.d2
CLINICAL SCIENCES: Clinical Case Studies

The Physical Activity Recall Assessment for People with Spinal Cord Injury: Validity

LATIMER, AMY E.1; MARTIN GINIS, KATHLEEN A.1; CRAVEN, B. CATHARINE2; HICKS, AUDREY L.1

Free Access
Article Outline
Collapse Box

Author Information

1McMaster University, Hamilton, Ontario, CANADA; and 2University of Toronto, Toronto, Ontario, CANADA

Address for correspondence: Kathleen A. Martin Ginis, McMaster University, Department of Kinesiology, Centre for Health Promotion and Rehabilitation, Hamilton, ON Canada, L8S 4K1; E-mail: martink@mcmaster.ca.

Submitted for publication March 2005.

Accepted for publication August 2005.

Collapse Box

Abstract

Purpose: This study examined the construct validity of the physical activity recall assessment for people with spinal cord injury (PARA-SCI).

Methods: First, to assess convergent validity, relationships between PARA-SCI scores and measures of aerobic fitness and muscular strength were examined among 73 men and women with SCI. Second, extreme groups analyses were conducted. PARA-SCI scores from 158 people with SCI were compared between groups differing on demographic, disability, and behavioral characteristics.

Results: Scores from the leisure time physical activity (LTPA) and cumulative activity PARA-SCI categories correlated positively with parameters of aerobic fitness and muscular strength. Scores from the lifestyle activity PARA-SCI category were not consistently associated with fitness parameters. LTPA category scores could differentiate between groups differing by age, sex, gym or sports team membership, and frequency of participation in LTPA. Lifestyle and cumulative activity scores were unable to distinguish between most groups.

Conclusion: The convergent validity study provided evidence of validity for the PARA-SCI LTPA and cumulative activity categories. The extreme groups analyses provided further evidence of the validity of the LTPA category by demonstrating differences in extreme groups. Together, these findings contribute to the accumulating evidence of the construct validity of the PARA-SCI LTPA category and its utility for assessing LTPA among individuals with SCI. These results also highlight measurement constraints of the lifestyle activity and cumulative activity categories.

Numerous self-report physical activity measures exist (10). Two key factors, however, render these measures invalid in the spinal cord injured (SCI) population: (a) the continuum of activities assessed is narrow and representative of activities commonly performed by people without SCI, and (b) the classification systems used to estimate the intensity of reported activities are based on standards set for people without SCI. Because of the constraints of the existing physical activity measures, research examining the physical activity patterns and the long-term health benefits of physical activity participation among individuals with SCI has been limited. To advance physical activity research in the SCI domain, the development of a SCI-specific physical activity measure is needed. Accordingly, the physical activity recall assessment for people with SCI (PARA-SCI (15)) was developed.

The PARA-SCI is a measure of physical activity for individuals with SCI who use a wheelchair as their primary mode of mobility. Martin Ginis et al. (15) provide a detailed description and copy of the instrument. Briefly, the PARA-SCI is administered in a semistructured telephone interview format. It provides an estimate of total time spent performing self-reported mild, moderate, and heavy intensity leisure time physical activity (LTPA; i.e., personally chosen activities performed during free time), lifestyle activity (i.e., routine activities: personal hygiene, household chores, work-related activity, passive leisure activity), and cumulative activity (i.e., combined total of LTPA and lifestyle activity). Preliminary data provide initial indication that the PARA-SCI is a valid and reliable measure. In a sample of nine men and women with SCI, 1-d PARA-SCI scores correlated with day-long measures of oxygen consumption (V̇O2). Over repeated administrations 1 wk apart, test-retest reliability coefficients (ICC: 0.45-0.91; (15)) suggested adequate reliability for measures of all three PARA-SCI activity categories (LTPA, lifestyle activity, and cumulative activity).

Although this preliminary evidence is promising, the scale's construct validity is yet to be explored. Moreover, it is not yet known whether PARA-SCI scores are indicative of physical fitness and sensitive to variation in physical activity levels. If the PARA-SCI is to be used as a tool to examine physical activity patterns and their relation to health and fitness outcomes, the construct validity of the PARA-SCI must be established. Thus, the goal of the present study was to examine the construct validity of the PARA-SCI. The two methods used to establish construct validity were demonstrations of convergent validity and extreme groups comparisons.

The convergent validity of a physical activity scale is typically indicated by demonstrating expected relationships between the scale scores and measures of physical fitness (22). With research indicating that physical activity leads to improvements in physical fitness among people with SCI (7), it was hypothesized that PARA-SCI scores would correlate positively with physical fitness parameters.

Validation of a physical activity measure by extreme groups involves administering the measure to two groups who should differ in terms of their physical activity and then confirming that the groups' scores, in fact, are significantly different. Extreme groups were created on the basis of demographic, disability, and behavioral characteristics. Consistent with previous research in the general population, it was hypothesized that LTPA scores would be greater for men than women, younger participants than older participants, those belonging to a gym or sports team than those who do not, and those reporting more frequent LTPA participation than those reporting infrequent LTPA participation (3). In accordance with SCI-specific research showing that individuals with tetraplegia participate in fewer daily activities than those with paraplegia (12), it was predicted that people who were more physically disabled (i.e., complete lesion, tetraplegia, power wheelchair use) would report less heavy intensity lifestyle activity than less physically disabled people (i.e., incomplete lesion, paraplegia, manual wheelchair use). It was also hypothesized that individuals who were working or attending school would accumulate more minutes of mild intensity lifestyle activity than those who were not working or attending school (Canadian Paraplegic Association, http://www.canparaplegic.org/pdf/1/wforce.pdf3).

Back to Top | Article Outline

METHODS

Participants

Study procedures were reviewed and approved by the research ethics boards at McMaster University and the Toronto Rehabilitation Institute. Written consent was obtained from all subjects before participation. Participants were recruited using convenience-sampling methods. Study eligibility requirements were: (a) neurologic impairment secondary to SCI (i.e., traumatic or nontraumatic SCI), (b)18 to 65 yr of age, (c) wheelchair use (power or manual) as the primary mode of mobility outside of the home, (d) able to read and speak English, and (e) absence of memory deficits. For the convergent validity phase only, study volunteers were excluded if they had an injury above C5 because they would not have had adequate residual function to perform the fitness tests. Participant demographics are summarized in Table 1.

Table 1
Table 1
Image Tools

A total of 73 men and women participated in the convergent validity study. Most were community dwelling (97.3% outpatients, 2.7% inpatients) and were recruited through community physical activity programs (38.3%), outpatient medical clinics (16.4%), and the consumer databases associated with these services (31.5%). The remaining 13.7% were recruited using snowball sampling, whereby other participants in the study referred them.

The extreme groups analysis was conducted as a secondary analysis of data from 158 men and women who had participated in either the PARA-SCI reliability study (N = 90 (15)), the convergent validity study (N = 46), or both (N = 23).

Back to Top | Article Outline
Testing Protocols
PARA-SCI.

The PARA-SCI is a self-report measure of time spent doing mild, moderate, and heavy intensity physical activity for the 3-d period preceding the interview. A trained research assistant administered the PARA-SCI via telephone following a detailed interview protocol (refer to Martin Ginis et al. (15)). Before completing the interview, the research assistant reviewed the PARA-SCI definition of physical activity-any activity requiring physical exertion (2)-and the definitions of mild, moderate, and heavy intensity activity as defined by the PARA-SCI intensity classification system (15). During the interview, the research assistant recorded every reported physical activity, along with its duration and intensity level, and whether it was LTPA or lifestyle activity. LTPA included all structured and unstructured physical activities that respondents chose to do during their free time (2). All other reported physical activities were classified as lifestyle activity.

Back to Top | Article Outline
Muscle strength.

Testing was conducted using multistation, wheelchair accessible, weight training systems at two testing sites. Of the participants, 62% completed all strength testing at a community-based center for health promotion (site 1) using Equalizer Exercise Machines (Red Deer, AB, Canada). The remaining 38% completed the strength testing at an in- and outpatient rehabilitation facility (site 2) using an H480 Access Trainer (Hoist Fitness System, San Diego, CA). Both systems are wheelchair accessible, multistation apparatuses that allow users to wheel right up to a lifting station and perform strengthening exercises while remaining seated in their wheelchairs. Exercises are performed by pushing or pulling levers (e.g., the chest press is performed by pushing a lever straight forward at chest level) or grips attached to pulleys(e.g., the biceps curl is performed by stabilizing the upper arm and elbow in front of the body in an extended position, and then contracting the biceps, pulling the handgrip up to the shoulder). Wrist cuffs were used by all tetraplegic participants as they had insufficient hand function to use the grips for the biceps exercise. The cuffs were attached to the end of the pulley. The weight systems at both testing sites were calibrated using a strain gauge (Aries Instruments Limited, Toronto, ON, Canada) and amplifier (Vishay Measurement Instrument Company, Raleigh, NC), thus allowing the loads lifted at either site to be equated.

Muscle strength was assessed to determine the maximal load that could be lifted in one repetition (1RM) for chest press (unilaterally) and biceps curl (unilaterally). Tests were terminated at the participant's point of fatigue. Attainment of 1RM was confirmed by participants' indication that their last lift was "heavy intensity" activity, as defined by the PARA-SCI activity intensity classification system.

Back to Top | Article Outline
Aerobic fitness.

Aerobic fitness was assessed to determine V̇O2peak through a progressive exercise test on an arm ergometer (Monark Ergomedic 881E, Varberg, Sweden). Tetraplegic participants with insufficient hand function to maintain grip on the ergometer had their hands attached to the grips with elastic bandages. Expired gases were collected during the test using the V̇O2000 portable metabolic unit (MedGraphics, St. Paul, MN) for measurements of O2 uptake and CO2 output. The highest value recorded during the test averaged over 30 s was deemed V̇O2peak. Attainment of V̇O2peak was corroborated by participants' self-report of "heavy intensity" activity in accordance with the PARA-SCI activity intensity guidelines, and an observed respiratory exchange ratio (RER) ≥1.00 (13). The workload at which the participant was cycling when V̇O2peak was reached was used as an additional index of aerobic fitness.

For all participants, the first three increments of this continuous multistage test were 2 min, and the remaining increments were 1 min. Initial workload and the increment of workload increase were determined by lesion level (tetraplegia, paraplegia). Individuals with tetraplegia began cycling at a rate of 50 rpm at a workload of 0 W (4,13). In each proceeding stage, the cycling rate was held constant and the workload was increased by 5 W (11). Individuals with paraplegia began cycling at a rate of 50 rpm at a workload of 25 W (28). In each proceeding stage, the cycling rate was held constant and the workload was increased by 10 W for women and 15 W for men (8). For participants who reached the maximal resistance on the arm ergometer (100 W) before reaching their volitional point of exhaustion, the resistance was held constant (100 W) and the cycling rate was increased by 5 rpm. At 100W, an increase in 5 rpm is equivalent to a 10-W increase in workload (16). Tests were terminated at the participant's volitional point of exhaustion.

Back to Top | Article Outline
Data Collection
Convergent validity study.

Participants performed the fitness tests at one of two testing sites. Chi-square analyses revealed no significant differences in the distributions for sex and lesion characteristics (i.e., level, completeness) between testing sites (P > 0.05). Further, separate ANOVA indicated no differences between testing sites for any of the fitness parameters.

After completing the fitness tests, participants were randomly assigned to an interview date. They were given a package of study materials, including a copy of the PARA-SCI intensity classification system, an appointment reminder card, and a $15 honorarium. Interviews were scheduled for a date at least 4 d after the initial assessment to ensure that the participant's fitness assessment was not included in the 3-d recall. On the appointed date, a trained research assistant contacted the participant via telephone. Participants were required to have their copy of the intensity classification system available for reference during the interview. At the completion of the interview, participants were debriefed and thanked.

Back to Top | Article Outline
Extreme groups analyses.

The extreme groups comparisons were a secondary analysis of data collected during the reliability (15) and convergent validity studies. For participants who completed the scale on more than one occasion (e.g., as part of the test-retest reliability study), PARA-SCI scores from the first administration were analyzed. Further, those participants who completed the scale as part of the reliability study were contacted 1 wk after completing the PARA-SCI to complete a single-item frequency rating of their physical activity participation.

Back to Top | Article Outline
Data Treatment and Statistical Analyses.

All analyses were conducted using the Statistical Package for the Social Sciences (SPSS) for Windows, v12. The PARA-SCI was scored as follows. First, mean daily total LTPA and lifestyle activity scores were calculated by averaging the number of minutes reported for all intensity categories of LTPA and lifestyle activity, respectively. Second, a mean daily total cumulative activity PARA-SCI score was calculated by averaging the number of minutes of physical activity (total LTPA + total lifestyle activity) reported for all 3 d. As well, for each category of activity (i.e., LTPA, lifestyle activity, and cumulative activity), separate scores for each level of intensity were calculated.

Back to Top | Article Outline
Convergent validity study: data screening and standardization across sex and lesion level.

The sample size for the muscular strength parameters ranged from 60 to 66 because some participants could not complete all of the fitness tests, and those who classified the intensity of a particular 1RM exercise as less than heavy (i.e., did not reach their maximum) had their data removed for that exercise. Similarly, the sample size for the aerobic fitness parameters was 48 because of the exclusion of participants who did not reach their V̇O2peak (i.e., did not achieve RER ≥1.00 and self-report heavy intensity activity at peak). A chi-square analysis indicated that significantly more individuals with paraplegia (80.6%) than tetraplegia (51.5%) attained V̇O2peak, χ2 (1) = 6.91, P= 0.009. No difference in the sex distribution was seen between those who attained V̇O2peak and those who did not.

Sex (1,3) and lesion characteristics have been shown to moderate physical fitness (9). Indeed, in a series of ANOVA, it was observed that men lifted significantly more weight than women on all of the strength tests (all P< 0.05). Significant differences also emerged across lesion level for all of the fitness parameters except right biceps strength, such that people with paraplegia had greater 1RM strength, V̇O2peak and workload at V̇O2peak than those with tetraplegia (P < 0.05). None of the fitness parameters varied as a function of lesion completeness. Given the moderating effects of sex and lesion level, each fitness parameter was standardized (converted to a z-score) as necessary. Right biceps 1RM was standardized for sex only. The remaining 1RM values were standardized for sex and lesion level. V̇O2peak and workload were standardized for lesion level. One-tailed bivariate correlations were then calculated between PARA-SCI scores and each standardized fitness parameter. Level of significance was set at P < 0.05.

Back to Top | Article Outline
Extreme groups analyses: Creation of extreme groups.

PARA-SCI scores were compared across extreme groups created on the basis of demographic (i.e., age, sex, employment status), disability (i.e., lesion level and completeness, mode of mobility), and behavioral characteristics (i.e., participation frequency, gym or sports team membership). For age, extreme groups were created using a tertile split. Independent t-tests confirmed that significantly different older and younger groups were created (P < 0.05). For the behavioral characteristics, extreme groups were created using a tertile split based on responses to a single-item, continuous measure of LTPA participation frequency that was administered to participants in the reliability study as part of their second interview: "In the course of the past week, how often did you engage in leisure time physical activity for at least 30 min?" (1 = never, 7 = everyday). Similar single-item measures of physical activity have demonstrated adequate validity and reliability (5). Independent t-tests confirmed that significantly different higher and lower frequency groups were created on the basis of scale responses (P < 0.05). These participants were also classified as belonging to a gym or sports team (or not), based on their self-reports.

Before conducting the between-groups analyses, all of the PARA-SCI scores were submitted to either square root or log transformations (as dictated by the skewness and kurtosis of each variable; (27)) to remedy their nonnormal distributions. A series of between-groups comparisons were then performed using the GLM MANOVA procedure in SPSS. For each PARA-SCI activity category (i.e., cumulative activity, LTPA, lifestyle activity), a MANOVA was performed with mild, moderate, and heavy intensity activity as the outcome variables and an extreme group characteristic as the between-groups factor. Separate MANOVA were conducted for each extreme group characteristic. In a few of these analyses, Box's M tests were significant indicating differences in the scores' covariance matrices. To account for these cases of significance, Pillai's trace was used as the criterion for testing for multivariate effects.

Significant multivariate effects were followed-up with ANOVA. They were conducted also to compare total scores for each PARA-SCI category (i.e., total cumulative activity, total LTPA and total lifestyle activity) by extreme groups. Because education can be a determinant of the nature of an individual's work (blue collar vs white collar; (21)), education was entered as a covariate in all of the analyses in which employment status was the independent variable. The P value for all analyses was set at P < 0.05, and effect sizes were calculated using Cohen's d. Because of missing data, a tertile split on some of the variables, and the administration of the participation frequency measure only in the reliability study, the sample size fluctuates between each of the analyses, ranging from 83 to 158.

Back to Top | Article Outline

RESULTS

Convergent Validity
PARA-SCI correlations with muscle strength.

Biceps 1RM correlated positively with total, moderate, and heavy intensity LTPA (r ≥ 0.21, P < 0.05; refer to Table 2). Thus, as hypothesized, greater time spent engaging in LTPA, particularly moderate and heavy intensity LTPA, was related to greater 1RM biceps strength. Biceps strength also correlated with heavy intensity lifestyle and cumulative activity (r ≥ 0.23, P<0.05), indicating that greater time spent engaging in all modes of heavy intensity physical activity was associated with greater biceps 1RM. Left chest 1RM correlated only with moderate intensity LTPA (r = 0.23, P = 0.03) such that individuals reporting more moderate intensity LTPA had greater left chest strength. Right chest 1RM was not related to any of the PARA-SCI scores.

Table 2
Table 2
Image Tools
Back to Top | Article Outline
PARA-SCI correlations with aerobic fitness.

As hypothesized, V̇O2peak correlated with heavy intensity LTPA and moderate and heavy intensity cumulative activity (r ≥ 0.26, P < 0.04; Table 2). Individuals spending more time engaged in heavy intensity LTPA and moderate and heavy intensity cumulative activity had greater aerobic endurance. Workload was significantly related to moderate, heavy, and total LTPA as well as heavy intensity cumulative activity (r ≥ 0.28, P < 0.02). Therefore, as predicted, individuals reporting more minutes of physical activity, particularly moderate and heavy intensity LTPA, attained a greater workload at V̇O2peak.

Back to Top | Article Outline
Extreme Groups Analyses
Between-groups differences in leisure time physical activity.

Table 3 presents the LTPA scores as a function of each grouping variable. With regard to the demographic characteristics, the MANOVA on mild, moderate, and heavy intensity LTPA scores was significant for age, F (3,106) = 3.94, Pillai's trace = 0.10, P = 0.01. As predicted, younger respondents engaged in more moderate intensity LTPA than older participants, F (1,108) = 10.68, P = 0.001, d = 0.62. For total LTPA, separate ANOVA indicated a main effect for age, F (1,108) = 11.18, P = 0.001, d = 0.64 and sex F (1,156) = 4.51, P = 0.04, d = 0.36. Consistent with trends in the general population, men and younger participants reported more total LTPA compared with women and older participants, respectively. None of the other demographic characteristics differentiated LTPA scores.

Table 3
Table 3
Image Tools

For the behavioral characteristics, LTPA scores varied across groups. In a MANOVA comparing mild, moderate, and heavy activity for those who did versus those who did not belong to a gym or sports team, a significant multivariate effect emerged, F (3,129) = 7.01, Pillai's trace= 0.14, P < 0.001. Those who belonged to a gym or team reported more moderate, F (1,131) = 4.93, P = 0.03, d = 0.39, and heavy intensity LTPA, F (1,131) = 19.93, P < 0.001, d = 0.79, and, in a separate ANOVA, they also reported more total LTPA, F (1,131) = 12.87, P < 0.001, d=0.62. Likewise, a significant multivariate effect emerged when participants were categorized as high or low active using the single-item activity frequency measure, F (3,80) = 3.65, Pillai's trace = 0.12, P = 0.02. High active participants reported more minutes of mild, F (1,82) = 4.76, P = 0.03, d = 0.47, moderate, F (1,82) = 5.17, P = 0.03, d = 0.50, and heavy intensity activity, F (1,82) = 6.20, P = 0.02, d = 0.53. In a separate ANOVA, they also reported more total LTPA, F (1,82) = 12.64, P = 0.001, d = 0.77. LTPA scores did not differ across any disability characteristics.

Back to Top | Article Outline
Between-groups differences in lifestyle activity.

Table 4 presents the lifestyle activity scores as a function of each categorical variable. The multivariate analysis of covariance was significant for work status, F (3,151) = 3.21, Pillai's trace = 0.06 P = 0.03. Consistent with SCI research (Canadian Paraplegic Association, http://www.canparaplegic.org/pdf/1/wforce.pdf3), people who were employed or attending school accumulated more minutes of mild intensity lifestyle activity than those who did not, F (1,153) = 4.94, P = 0.03, d = 0.37. The lifestyle activity scores did not differ along any of the other categorical variables.

Table 4
Table 4
Image Tools
Back to Top | Article Outline
Between-groups differences in cumulative activity.

Table 5 presents the scores from the cumulative activity category as a function of each independent variable. Gym or sports team membership was the only variable that produced a significant multivariate effect, F(3,129) = 3.14, Pillai's trace = 0.07, P = 0.03. Specifically, ANOVA indicated that heavy intensity cumulative activity, F (1,131) = 8.60, P = 0.004, d = 0.29, differentiated between people who belonged to a gym or sports team and those who did not. No between-groups differences were found for any of the other extreme group characteristics.

Table 5
Table 5
Image Tools
Back to Top | Article Outline

DISCUSSION

Although construct validation is an ongoing process (26), the present findings provide preliminary evidence of the PARA-SCI's construct validity. In an examination of the scale's convergent validity, study participants who reported more minutes of moderate and heavy intensity LTPA and cumulative activity exhibited better physical fitness than those who reported fewer minutes of such activities. Furthermore, in a series of between-groups comparisons, LTPA scores differed significantly in expected directions. These findings have important implications for understanding the strengths and limitations of the PARA-SCI.

An important strength of the PARA-SCI is that accumulating evidence supports the validity of the LTPA category as a measure of LTPA performed by people with SCI. The results of the extreme groups analyses paralleled well-documented activity trends in the general population (3) and among people with a physical disability (20,29). Specifically, compared with their counterparts, greater LTPA was reported by men, younger respondents, people who self-classified as being frequently active, and those who belonged to a gym or sports team. Likewise, the convergent validity study produced a pattern of results seen in previous studies whereby significant relationships emerged between fitness and self-reports of heavier, but not lighter activity (24). The magnitude of the significant correlations (ranging from 0.21 to 0.36) was comparable to values found in validation studies of other self-report physical activity surveys for people with and without disabilities. For example, Rimmer et al. (20) found correlations between physical activity and disability survey (PADS) LTPA and exercise subscale scores and indices of physical fitness ranging from 0.22 to 0.28. In a validity assessment of the 7-d physical activity recall, correlations with V̇O2 max were 0.38 and 0.28 for moderate and hard self-reported activity, respectively (22). Together, the findings from the validation studies suggest that the construct validity of PARA-SCI LTPA category is just as good as that reported for other accepted measures of physical activity.

In contrast, a limitation of the PARA-SCI is that scores for lifestyle activity were not correlated with physical fitness and did not discriminate between people with different disability characteristics (lesion level and completeness, mode of mobility) who would be expected to differ in their physical conditioning and time spent on physical activity. These null findings may result from the PARA-SCI's broad definition of lifestyle activity, which encompasses a range of activities for which the direction of the relationship with fitness may vary. Indeed, among people with physical disabilities, time spent on some lifestyle activities may be positively associated with fitness, whereas time spent on others may be negatively correlated with fitness (20). For example, a poorly conditioned person with SCI who reports that driving a car for 20 min is moderate intensity activity will have the same PARA-SCI lifestyle activity score as a person who is conditioned and wheels around the grocery store for 20 min at a moderate intensity. Of course, fitness should be negatively correlated with time spent driving a car and positively correlated with time spent wheeling.

On the one hand, failure to distinguish between more and less impaired or fit respondents may be a strength of the PARA-SCI, because it indicates that the survey meets the important objective of capturing all types of physical activity (e.g., low-level activities that individuals with a high degree of impairment find physically demanding and have been overlooked on other scales) performed by individuals with SCI, regardless of their physical capacity. On the other hand, the equal scores may be viewed as problematic if a researcher is trying to distinguish between those who participate in lifestyle activity that produces fitness benefits from those who do not. A possible solution to this measurement conundrum is to create separate lifestyle activity subcategories representing activities for which high scores represent better versus worse fitness or less versus more skill. Creation of these subcategories would require an extensive examination of the relationship of each lifestyle activity with fitness and skill in those with SCI.

Another explanation for the lack of association between lifestyle activity and aerobic fitness is that the intensity and duration of commonly performed lifestyle activities are below the minimal threshold to increase aerobic fitness (9). It should be noted, however, that very sedentary people (e.g., those with SCI) may derive health benefits from very low levels of activity that do not increase aerobic fitness (23). Further work is needed to determine whether PARA-SCI lifestyle activity scores are related to other health-related outcomes in those with SCI (e.g., range of motion, frequency of pressure sores, or psychological well-being). It may be possible to validate this category as a measure of activity yielding non-fitness-related benefits (e.g., improved health or mobility) to people with SCI.

In a similar vein, the PARA-SCI lifestyle scores may be more strongly related to activities of daily living (ADL) and lifestyle activity skill level than fitness level (25). For instance, two people can be equally fit, but if one has poor transfer skills, that person would report more daily minutes spent transferring (i.e., more minutes of lifestyle activity) and greater exertion when transferring, than a person with excellent transfer skills. As such, investigators may find that the PARA-SCI measure of lifestyle activity is related to ADL proficiency.

The PARA-SCI cumulative activity category demonstrated good convergent validity. Corroborating the preliminary findings from an assessment of criterion validity (i.e., day-long oxygen consumption (V̇O2) correlated strongly with moderate and heavy intensity PARA-SCI cumulative activity scores (15), aerobic fitness was associated with PARA-SCI moderate and heavy intensity cumulative activity scores. Additionally, biceps strength was associated with heavy intensity cumulative activity. In the extreme groups analysis, however, cumulative activity only distinguished between people who belonged to a gym or sports team and those who did not. Although the convergent validity data are encouraging, because the cumulative activity category is a function of the lifestyle activity category, it faces similar limitations in terms of differentiating between people with low physical functioning who report fitness nonenhancing lifestyle activity as mild, moderate, or heavy intensity versus those with high functioning who report more vigorous forms of lifestyle activity at these intensities. As the lifestyle activity category is further refined and validated, the cumulative activity category should also begin to emerge as a more valid measure of physical activity for people with SCI.

Providing evidence of the construct validity of the PARA-SCI represents a potentially important opportunity for advancing the study of physical activity in those with SCI. A limitation, however, is that the PARA-SCI has been validated for telephone use only. An instrument valid for telephone-based interviewing is ideal for large population studies, but may have limited use in clinical settings where face-to-face interviews are preferable. Evidence from a validation study of another self-report measure of physical activity suggests that recall surveys can be administered in person and over the telephone interchangeably (6). Such versatility in administration mode likely generalizes to the PARA-SCI. Nevertheless, further validation studies are needed to confirm this prediction. Another potential limitation is that data for the validation studies were obtained from a sample in which most of the respondents were men. The similarity of the sample's demographics to the population demographics of those with SCI is a strength of the present research. The extent to which these findings can be generalized to women, however, is unknown.

Despite these limitations, the study findings contribute to the accumulating evidence indicating the validity of the PARA-SCI as a measure of LTPA. Given that the validity of a measure cannot be established in a single study but rather requires ongoing evaluation (26), continued research is needed to further establish the validity of the PARA-SCI. Ongoing PARA-SCI research will have tremendous value for those with SCI. People with SCI have been understudied in terms of the relationship between physical activity, health, and fitness (14), but a valid measure of LTPA will enable researchers to move forward and investigate these relationships. Furthermore, self-report measures have been widely used to establish the association between LTPA and improved health in the general population (17,18) and to develop public health activity guidelines. Having demonstrated that PARA-SCI LTPA scores are indicative of physical fitness and sensitive to variation in physical activity levels holds promise for the use of this measure in similar health outcome studies in those with SCI. This suggestion is timely given the emerging need to establish the health benefits of physical activity for people with SCI (19). Finally, by developing a valid tool for tracking activity patterns, researchers will have a means to examine determinants of activity in those with SCI. This possibility represents an important step toward developing activity-enhancing interventions and pursuing a research agenda aimed at promoting the health of people with SCI through physical activity.

Research supported by Canadian Institutes of Health Research (CIHR) Grant # MOP-57778. Amy Latimer is currently a Postdoctoral Fellow in the Health, Emotion and Behavior Lab at Yale University.

Back to Top | Article Outline

REFERENCES

1. American College of Sports Medicine. American College of Sports Medicine Position Stand. The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness, and flexibility in healthy adults. Med. Sci. Sports Exerc. 30:975-991, 1998.

2. Bouchard, C., and R. J. Shephard. Physical activity, fitness, and health: the model and key concepts. In: Physical Activity, Fitness, and Health: International Proceedings and Consensus Statement, C. Bouchard, R. J. Shephard, T. Stephens (Eds.). Champaign, IL: Human Kinetics, 1994, pp. 77-88.

3. Craig C. L., and C. Cameron. Increasing physical activity: assessing trends from 1998-2003. Ottawa, ON: Canadian Fitness and Lifestyle Research Institute, 2004, pp. 20-28.

4. DiCarlo, S. E. Effect of arm ergometry training on wheelchair propulsion endurance of individuals with quadriplegia. Phys. Ther. 68:40-44, 1988.

5. Godin, G., and R. J. Shephard. A simple method to assess exercise behavior in the community. Can. J. Appl. Sports Sci. 10:141-146, 1985.

6. Hayden-Wade, H. A., K. J. Coleman, J. F. Sallis, and C. Armstrong. Validation of the telephone and in-person interview versions of the 7-day PAR. Med. Sci. Sports Exerc. 35:801-809, 2003.

7. Hicks, A. L., K. A. Martin, D. S. Ditor, et al. Long-term exercise training in persons with spinal cord injury: effects on strength, arm ergometry performance and psychological well-being. Spinal Cord 41:34-43, 2003.

8. Jacobs, P. L., M. S. Nash, K. J. Klose, R. S. Guest, B. M. Needham-Shropshire, and B. A. Green. Evaluation of a training program for persons with SCI paraplegia using the Parastep 1 ambulation system: Part 2. Effects on physiological responses to peak arm ergometry. Arch. Phys. Med. Rehabil. 78:794-798, 1997.

9. Janssen, T. W., C. A. van Oers, H. E. Veeger, A. P. Hollander, L. H. van der Woude, and R. H. Rozendal. Relationship between physical strain during standardized ADL tasks and physical capacity in men with spinal cord injuries. Med. Sci. Sports Exerc. 32:844-859, 1994.

10. Kriska, A., and C. J. Casperson. Introduction to a collection of physical activity questionnaires. Med. Sci. Sports Exerc. 29:S5-S9, 1997.

11. Lasko-McCarthey, P., and J. A. Davis. Protocol dependency of V̇O2max during arm cycle ergometry in males with quadriplegia. Med. Sci. Sports Exerc. 23:1097-1101, 1991.

12. MacDonald, M. R., W. R. Nielson, and M. G. Cameron. Depression and activity patterns of spinal cord injured persons living in the community. Arch. Phys. Med. Rehabil. 68:339-343, 1987.

13. Manns, P. J., and K. E. Chad. Determining the relation between quality of life, handicap, fitness, and physical activity for persons with spinal cord injury. Arch. Phys. Med. Rehabil. 80:1566-1571, 1999.

14. Martin Ginis, K. A., and A. L. Hicks. Exercise research issues in the spinal cord injured population. Exerc. Sport Sci. Rev. 33:49-53, 2005.

15. Martin Ginis, K. A., A. E. Latimer, A. L. Hicks, and B. C. Craven. Development and preliminary evaluation of an activity measure for people with spinal cord injury. Med. Sci. Sports Exerc. 37:1099-1111, 2005.

16. Monark Exercise AB. Monark Exercise AB Manual: Rehab Trainer 881E. Valberg, SE: Monark Exercise AB, 2004, pp. 4-12.

17. Paffenbarger, R. S., R. T. Hyde, A. L. Wing, and C. C. Hsieh. Physical activity, all-cause mortality, and longevity of college alumni. N. Engl. J. Med. 314:605-613, 1986.

18. Paffenbarger, R. S., R. T. Hyde, A. L. Wing, I. M. Lee, D. L. Jung, and J. B. Kampert. The association of changes in physical-activity level and other lifestyle characteristics with mortality among men. N. Engl. J. Med. 328:538-545, 1993.

19. Rimmer, J. H., D. Braddock, and K. H. Pitetti. Research on physical activity and disability: An emerging national priority. Med. Sci. Sports Exerc. 28:1366-1372, 1996.

20. Rimmer, J. H., B. B. Riley, and S. S. Rubin. A new measure for assessing the physical activity behaviors of persons with disabilities and chronic health conditions: The Physical Activity and Disability Survey. Am. J. Health Promot. 16:34-45, 2001.

21. Sallis J. F., and N. Owen. Physical Activity and Behavioral Medicine. Thousand Oaks: CA: Sage Publications Ltd., 1998, pp. 110-134.

22. Sallis, J. F., and B. E. Saelens. Assessment of physical activity by self-report: status, limitations, and future directions. Res. Q. Exerc. Sport. 71:S1-S14, 2000.

23. Shephard, R. J. Limits to the measurement of habitual physical activity by questionnaires. Br. J. Sports Med. 37:197-206, 2003.

24. Singh, P. N., S. Tonstad, D. E. Abbey, and G. E. Fraser. Validity of selected physical activity questions in white Seventh-day Adventists and non-Adventists. Med. Sci. Sports Exerc. 28:1026-1037, 1996.

25. Stewart, M. W., S. L. Melton-Rogers, S. Morrison, and S. F. Figoni. The measurement properties of fitness measures and health status for persons with spinal cord injuries. Arch. Phys. Med. Rehabil. 81:394-400, 2000.

26. Streiner D. L., and G. R. Norman. Health Measurement Scales: A Practical Guide to Their Development and Use. 2nd ed. New York, NY: Oxford University Press Inc., 1999, pp. 54-162.

27. Tabachnick, B. G., and L. S. Fidell. Using Multivariate Statistics. 4th ed. Boston, MA: Allyn and Bacon, 2001, pp. 56-110.

28. Van Loan, M. D., S. McCluer, J. M. Loftin, and R. A. Boileau. Comparison of physiological responses to maximal arm exercise among able-bodied, paraplegics and quadriplegics. Paraplegia 25:397-405, 1987.

29. Washburn, R. A., W. Zhu, E. McAuley, M. Frogley, and S. F. Figoni. The physical activity scale for individuals with physical disabilities: Development and evaluation. Arch. Phys. Med. Rehabil. 83:193-200, 2002.

Cited By:

This article has been cited 15 time(s).

Rehabilitation Psychology
Developing Physical Activity Interventions for Adults With Spinal Cord Injury. Part 1: A Comparison of Social Cognitions Across Actors, Intenders, and Nonintenders
Ginis, KAM; Tomasone, JR; Latimer-Cheung, AE; Arbour-Nicitopoulos, KP; Bassett-Gunter, RL; Wolfe, DL
Rehabilitation Psychology, 58(3): 299-306.
10.1037/a0032815
CrossRef
Rehabilitation Psychology
The efficacy of an implementation intention intervention for promoting physical activity among individuals with spinal cord injury: A randomized controlled trial
Latimer, AE; Ginis, KAM; Arbour, KP
Rehabilitation Psychology, 51(4): 273-280.
10.1037/0090-5550.51.4.273
CrossRef
Spinal Cord
Activities of daily living performed by individuals with SCI: relationships with physical fitness and leisure time physical activity
Hetz, SP; Latimer, AE; Ginis, KAM
Spinal Cord, 47(7): 550-554.
10.1038/sc.2008.160
CrossRef
Spinal Cord
Physical activity is related to lower levels of pain, fatigue and depression in individuals with spinal-cord injury: a correlational study
Tawashy, AE; Eng, JJ; Lin, KH; Tang, PF; Hung, C
Spinal Cord, 47(4): 301-306.
10.1038/sc.2008.120
CrossRef
Spinal Cord
A review of instruments assessing participation in persons with spinal cord injury
Noonan, VK; Miller, WC; Noreau, L
Spinal Cord, 47(6): 435-446.
10.1038/sc.2008.171
CrossRef
Archives of Physical Medicine and Rehabilitation
Increased Participation in Activities of Daily Living Is Associated With Lower Cholesterol Levels in People With Spinal Cord Injury
Hetz, SP; Latimer, AE; Buchholz, AC; Ginis, KAM
Archives of Physical Medicine and Rehabilitation, 90(): 1755-1759.
10.1016/j.apmr.2009.04.021
CrossRef
Applied Physiology Nutrition and Metabolism-Physiologie Appliquee Nutrition Et Metabolisme
Considerations for the development of a physical activity guide for Canadians with physical disabilities
Ginis, KAM; Hicks, AL
Applied Physiology Nutrition and Metabolism-Physiologie Appliquee Nutrition Et Metabolisme, 32(): S135-S147.
10.1139/H07-108
CrossRef
Applied Physiology Nutrition and Metabolism-Physiologie Appliquee Nutrition Et Metabolisme
Greater daily leisure time physical activity is associated with lower chronic disease risk in adults with spinal cord injury
Buchholz, AC; Ginis, KAM; Bray, SR; Craven, BC; Hicks, AL; Hayes, KC; Latimer, AE; McColl, MA; Potter, PJ; Wolfe, DL
Applied Physiology Nutrition and Metabolism-Physiologie Appliquee Nutrition Et Metabolisme, 34(4): 640-647.
10.1139/H09-050
CrossRef
Archives of Physical Medicine and Rehabilitation
Leisure Time Physical Activity in a Population-Based Sample of People With Spinal Cord Injury Part II: Activity Types, Intensities, and Durations
Ginis, KAM; Arbour-Nicitopoulos, KP; Latimer, AE; Buchholz, AC; Bray, SR; Craven, BC; Hayes, KC; Hicks, AL; McColl, MA; Potter, PJ; Smith, K; Wolfe, DL
Archives of Physical Medicine and Rehabilitation, 91(5): 729-733.
10.1016/j.apmr.2009.12.028
CrossRef
Adapted Physical Activity Quarterly
Physical Activity Measurement Among Individuals With Disabilities: A Literature Review
Cervantes, CM; Porretta, DL
Adapted Physical Activity Quarterly, 27(3): 173-190.

Archives of Physical Medicine and Rehabilitation
Leisure Time Physical Activity in a Population-Based Sample of People With Spinal Cord Injury Part I: Demographic and Injury-Related Correlates
Ginis, KAM; Latimer, AE; Arbour-Nicitopoulos, KP; Buchholz, AC; Bray, SR; Craven, BC; Hayes, KC; Hicks, AL; McColl, MA; Potter, PJ; Smith, K; Wolfe, DL
Archives of Physical Medicine and Rehabilitation, 91(5): 722-728.
10.1016/j.apmr.2009.12.027
CrossRef
Spinal Cord
Physical activity and subjective well-being among people with spinal cord injury: a meta-analysis
Ginis, KAM; Jetha, A; Mack, DE; Hetz, S
Spinal Cord, 48(1): 65-72.
10.1038/sc.2009.87
CrossRef
Spinal Cord
Establishing evidence-based physical activity guidelines: methods for the Study of Health and Activity in People with Spinal Cord Injury (SHAPE SCI)
Ginis, KAM; Latimer, AE; Buchholz, AC; Bray, SR; Craven, BC; Hayes, KC; Hicks, AL; McColl, MA; Potter, PJ; Smith, K; Wolfe, DL
Spinal Cord, 46(3): 216-221.
10.1038/sj.sc.3102103
CrossRef
2012 Annual International Conference of the Ieee Engineering in Medicine and Biology Society (Embc)
Physical Activity Monitoring and Sharing Platform for Manual Wheelchair Users
Ding, D; Ayubi, S; Hiremath, S; Parmanto, B
2012 Annual International Conference of the Ieee Engineering in Medicine and Biology Society (Embc), (): 5833-5836.

Journal of Rehabilitation Research and Development
Activity and participation after spinal cord injury: State-of-the-art report
Ullrich, PM; Spungen, AM; Atkinson, D; Bombardier, CH; Chen, YY; Erosa, NA; Groer, S; Ottomanelli, L; Tulsky, DS
Journal of Rehabilitation Research and Development, 49(1): 155-173.
10.1682/JRRD.2010.06.0108
CrossRef
Back to Top | Article Outline
Keywords:

DISABILITY; MEASUREMENT; PARAPLEGIA; TETRAPLEGIA; FITNESS; LEISURE-TIME PHYSICAL ACTIVITY

©2006The American College of Sports Medicine

Login

Article Tools

Images

Share

Search for Similar Articles
You may search for similar articles that contain these same keywords or you may modify the keyword list to augment your search.

Connect With Us