Advertisement

Spinal Cord Injury Functional Ambulation Profile: A Preliminary Look at Responsiveness

Kristin E. Musselman, Jaynie F. Yang
DOI: 10.2522/ptj.20130071 Published 1 February 2014

Abstract

Background The Spinal Cord Injury Functional Ambulation Profile (SCI-FAP) is a valid, reliable measure of walking skill (eg, walking while negotiating obstacles, doors, and stairs).

Objective The responsiveness of the SCI-FAP was assessed at least 7 months after spinal cord injury (SCI) and compared with that of the 10-Meter Walk Test (10MWT) and the Six-Minute Walk Test (6MWT).

Design A secondary analysis of data collected during a randomized, single-blind, crossover trial was performed.

Methods Participants had incomplete SCI and could walk at least 5 m without manual assistance. After 3 or 4 baseline assessments, participants completed 2 months of precision training (stepping over obstacles and onto targets on the ground) and 2 months of endurance training (treadmill training with body weight support, if needed). Walking function was assessed with the SCI-FAP, 10MWT, and 6MWT. Internal responsiveness was evaluated through change scores and standardized response means (SRMs). External responsiveness was gauged by correlating change scores on the SCI-FAP, 10MWT, and 6MWT. The minimal detectable change was calculated from the standard error of measurement from the baseline assessments.

Results The SCI-FAP scores improved with both interventions. The magnitude of change was greater for participants whose pretraining self-selected speed was less than 0.5 m/s. The SCI-FAP had moderate SRMs. The 10MWT (fastest speed) and 6MWT had the largest SRMs after precision training and endurance training, respectively. The minimal detectable change in the SCI-FAP was 96 points.

Limitations The convenience sample was small and all participants could ambulate independently (with devices); therefore, the generalizability of the findings is limited.

Conclusions The SCI-FAP was responsive to changes in walking ability in participants who had incomplete SCI and walked at slow speeds, but overall the 10MWT and 6MWT were more responsive.

After a motor incomplete spinal cord injury (SCI), walking training is a focus of rehabilitation. Comprehensive measures that are valid, reliable, and responsive are needed to evaluate progress in walking. In people with SCI, walking is most commonly evaluated with the 10-Meter Walk Test (10MWT) to assess speed, the Six-Minute Walk Test (6MWT) to assess endurance, and the Walking Index for Spinal Cord Injury II (WISCI II), which rates walking ability according to the assistance needed. Although these measures have good psychometric properties,1–4 all 3 assess walking on level ground. They do not encompass any change in terrain, obstacles in the walking path, or other skills known to be needed in daily walking.5

The Spinal Cord Injury Functional Ambulation Profile (SCI-FAP)6 was created to address the need for a measure of functional walking after SCI.7 The SCI-FAP involves 7 commonly encountered walking tasks: walking and carrying a bag, the Timed “Up & Go” Test, and walking while negotiating carpet, obstacles, stairs, a curb, and a door. It was shown to be valid and reliable in participants with chronic SCI,6 but its responsiveness remains unknown.

A taxonomy recommended by Beaton et al8 for describing responsiveness included specifying the type of change being studied (ie, important, detectable, or observed). Important change is the magnitude of change deemed meaningful by patients, clinicians, and society, whereas detectable change reflects measurement error.8 Observed change is the change in performance seen after participation in an intervention with known efficacy8 (also called “internal responsiveness”9). Whereas internal responsiveness focuses on the measure of interest, external responsiveness focuses on the relationship between the measure of interest and a well-established reference measure.9

Here we report the findings from a preliminary evaluation of the responsiveness of the SCI-FAP. Our 3 objectives were: (1) to assess the observed changes in the SCI-FAP, 10MWT, and 6MWT over the course of a clinical trial targeting walking (ie, internal responsiveness); (2) to relate the observed change in the SCI-FAP with changes observed simultaneously in the 10MWT and 6MWT (ie, external responsiveness); and (3) to determine the minimal detectable change (MDC) in the SCI-FAP. These indicators of responsiveness were assessed in a group of people who dwelled in the community, had incomplete SCI, and were participating in 2 walking training programs—1 performed overground and 1 performed on a treadmill. We hypothesized that the SCI-FAP would show internal responsiveness (ie, that SCI-FAP scores would improve with participation in walking training) and that change scores on the SCI-FAP would correlate with change scores on the other measures, demonstrating external responsiveness.

Method

Setting and Participants

People with incomplete SCI were recruited through the Canadian Paraplegic Association; the Glenrose Rehabilitation Hospital in Edmonton, Alberta, Canada; the Foothills Hospital in Calgary, Alberta, Canada; and advertisements on a website. People who had a stable lesion (eg, traumatic or nonprogressive cause) between C1 and L1, were at least 7 months postinjury, could walk at least 5 m without the assistance of another person, and did not have a cognitive or musculoskeletal impairment that affected walking were included. Written informed consent was obtained. Training sessions and assessments were performed at the Clinic for Ambulatory Rehabilitation, Research and Education, University of Alberta.

Design Overview

A secondary analysis was performed on data collected during a single-blind, randomized, crossover clinical trial in which 2 approaches for walking training were compared.10 Baseline assessments of walking function were repeated 3 or 4 times over 3 to 6 weeks to ensure stable scores for each participant. Manual muscle testing of 8 lower extremity muscles also was performed during the baseline period as previously described.11 After baseline testing, participants were randomly assigned to 1 of 2 groups by selection of a piece of paper from a box (block randomization with a block size of 4). The randomization was done by the treating therapist. The first group completed 2 months of precision training overground (described below), a 2-month rest period, and 2 months of endurance training on a treadmill (Fig. 1). The second group followed the same schedule, with the exception that endurance training preceded precision training. Clinical assessments of walking were administered at monthly intervals during the training and rest periods. The assessments were performed in the evening by physical therapists who did not work at the clinic during the day. This schedule ensured that the assessors remained unaware of group assignments.

Figure 1.
Figure 1.

Study design. After a series of baseline assessments, participants were randomly assigned to 1 of 2 treatment orders: endurance, rest, precision (top) or precision, rest, endurance (bottom). Asterisks indicate time points of assessments (bimonthly during the baseline period and monthly during the training and rest periods). “Analyzed” indicates the total number of participants per phase who were included in the analysis. “Excluded” indicates the number of participants per phase who trained but whose data were excluded from the analysis. Reasons for exclusion were poor attendance at training sessions (1 for endurance training and precision training), endurance training performed overground (1 participant), and chemotherapy administered concurrently (1 participant). “Discontinued” indicates the number of participants who dropped out and either did not begin or did not complete the training phase.

Interventions

People with SCI participated in 2 programs that focused on the retraining of walking—precision training and endurance training. Each training method was experienced for about 30 minutes per session, 3 to 5 sessions per week, for 2 months. Both types of training were administered by a physical therapist and an assistant. To ensure adherence to the training protocols, the therapist completed a daily training log, which was reviewed by a researcher weekly.

For precision training, participants walked along a 15-m track while stepping over obstacles and placing their feet on visual targets on the ground. The obstacles were Styrofoam (Rona, Boucherville, Quebec, Canada) blocks of various heights and widths, and the targets were thin fabric circles whose diameter was approximately equal to the width of the participant's foot. The emphasis of precision training was on accuracy (ie, clearing the obstacles without touching them and placing the foot on the targets to completely obscure them), not on speed or endurance. The placement and size of obstacles and targets were changed daily, and the difficulty of the tasks was gradually increased (ie, increased obstacle height or width and increased target diameter). Walking aids and physical assistance were permitted.

Endurance training involved walking on a treadmill with body-weight support and physical assistance, if needed. Participants walked at a speed that matched or exceeded their self-selected overground walking speed. Body-weight support, if used, was progressively decreased. The emphasis was to walk as fast and as long as possible, minimizing rest periods. For balance while walking, participants held a handrail (placed at chest height).

Outcome Measures

The primary outcome measure was the SCI-FAP. Secondary outcome measures included the 6MWT, the 10MWT (at comfortable and fastest speeds), and the WISCI II (self-selected and maximum ratings).

The SCI-FAP considers the time and assistance needed to complete 7 walking tasks. Participants were timed as they completed each task at a comfortable walking pace. The amount of assistance required (from a device, another person, or both) was evaluated with an ordinal scale. A score for each SCI-FAP task was calculated as follows: task score=(time × assistance rating)/mean able-bodied time. A total SCI-FAP score was obtained by summing all 7 task scores. An average adult who is able-bodied would score 7 points.6 The SCI-FAP has content, discriminative, and convergent validity and high interrater and test-retest reliability in people with chronic incomplete SCI.6

The 6WMT is a measure of walking endurance.12 Participants walked back and forth along a 25-m walkway, covering as much distance as possible in 6 minutes. The same walking aid was used for the 6MWT throughout the study period. The 6MWT has good validity, reliability, and responsiveness in people with SCI.2–4

The 10MWT is a measure of walking speed. Participants first walked in a straight line for 14 m at their comfortable speed (self-selected speed) and then repeated the test at their fastest pace (fast speed). Participants used the same walking device for the 10MWT throughout the study period. The time taken to traverse the middle 10 m of the walkway was used to calculate speed. The 10MWT is a valid and reliable measure in people with SCI,2,4 and it is responsive to change in walking function.3

The WISCI II is a 21-point ordinal scale that rates a person's ability to walk 10 m overground according to the braces, devices, and physical assistance needed.13 It is a valid and reliable measure,1,2,4 but its ability to detect change is less well established.3 Both self-selected (ie, WISCI II level used in the community) and maximum (ie, highest possible WISCI II level) scores were recorded.14

Data Analysis

Demographic characteristics were averaged across participants; the mean (1 standard deviation) is reported unless otherwise indicated. The following analyses were performed separately for precision training and endurance training. Within a participant, a mean baseline score was calculated for each measure by averaging the scores from the 3 or 4 baseline testing sessions (for the first phase of training) or by averaging the scores from the 2 testing sessions completed during the rest period (for the second phase of training). The change in performance on 7 outcomes (SCI-FAP score, SCI-FAP time, self-selected 10MWT, fast 10WMT, 6MWT, self-selected WISCI II, and maximum WISCI II) was calculated for each training period (score at 2 months of training minus baseline or rest period score). The SCI-FAP time was calculated by summing the times taken to complete all 7 tasks (no assistance rating included).

To assess the internal responsiveness of the SCI-FAP and to compare its responsiveness with those of the other measures, we used the standardized response mean (SRM). The SRM was calculated by dividing the mean change in a measure by the standard deviation of the mean change.9 For the SCI-FAP score and the SCI-FAP time, the SRMs were multiplied by −1 because a decrease in score on the SCI-FAP represents an improvement (unlike the scoring on the 10MWT and 6MWT). Therefore, a positive SRM was interpreted as an improvement in performance on a given measure, whereas a negative SRM implied a decline in performance.

To gauge how responsive a measure was to change, we applied the Cohen interpretation of effect size values15 to the absolute value of the SRM (ie, <0.2 was small, 0.2–0.8 was moderate, and >0.8 was large). For each SRM, a 95% confidence interval was calculated by use of a jackknife procedure.16 The jackknife procedure is a data resampling technique that was used to estimate the variance of each SRM. With this approach, the data point for 1 participant was dropped from the original sample, and the SRM was calculated. This process was repeated until each data point had been eliminated once. The result was a sample of n SRMs, where n equaled the number of participants, from which a mean SRM and a 95% confidence interval were calculated. The SRMs were then compared across measures by use of a 1-way repeated-measures analysis of variance. To evaluate the responsiveness of the individual SCI-FAP tasks, we calculated the SRMs for the 7 tasks. The Pearson correlation coefficient (r) was used to correlate change in the SCI-FAP with changes on the 6MWT and the 10MWT (self-selected and fast). An intention-to-treat analysis was not used.

The MDC in the SCI-FAP (total score, total time, and task scores), 10MWT (self-selected and fast), and 6MWT were calculated by use of the standard error of measurement (SEM) as follows: Embedded Image in this equation, sx is the baseline standard deviation and rx is the test-retest reliability of the outcome measure. To calculate the baseline standard deviation for each measure, we used the participants' mean scores from the baseline assessments. Test-retest reliability was assessed by use of a one-way random-effects intraclass correlation coefficient for absolute agreement. Scores from the first 3 baseline assessments were used. The MDC95, which is the MDC at a 95% confidence interval, was calculated from the SEM as follows17: Embedded Image To assess whether the internal responsiveness of the SCI-FAP differed with different walking proficiencies, we divided the participants into fast and slow groups on the basis of their self-selected walking speeds (as assessed with the 10MWT) before each training phase. Participants whose speed met or exceeded 0.5 m/s were considered to walk fast (fast group), and participants whose speed fell below 0.5 m/s were considered to walk slow (slow group). A speed of 0.5 m/s was chosen as the division point because it is the minimum required speed for walking in the community18,19 and approximates the values reported to distinguish between housebound walkers and community walkers for stroke (0.4 m/s20,21) and SCI (0.57 m/s22 and 0.49 m/s23). Change scores on the walking outcomes for participants in the fast and slow groups were compared by use of a Mann-Whitney U test (precision training and endurance training separately). The alpha level was set at .05. Post hoc analyses were performed with the Bonferroni correction.

Role of the Funding Source

The Canadian Institutes of Health Research, the Christopher and Dana Paralysis Foundation, the Alberta Paraplegic Foundation, and the Rick Hansen Foundation funded this project.

Results

Participants

Recruitment of participants began in October 2007, and training of the last participant was completed in July 2012. Seventy people with SCI were assessed for eligibility; 37 did not meet the eligibility criteria, 1 declined to participate, and 10 were unable to participate for other reasons (eg, the distance to travel to and from the training sessions was too great). Twenty-two people with SCI participated in endurance training, precision training, or both. Data for 1 participant were excluded from the analysis because of poor attendance at training sessions. Another participant discontinued training during the first phase because walking with a walker was aggravating pain in her wrists. No other adverse events resulted from either training program. Therefore, 20 participants (14 men), all of whom lived at home, were included. The trial was stopped at this point because an interim calculation of sample size, based on the mean change scores on the SCI-FAP, revealed that a large number of participants (360) would be needed to show significant differences between training methods.

All 20 participants completed 2 months of precision training; 17 participants completed the endurance phase. Participant 17 withdrew from the study after completing precision training. The endurance phases for 2 participants were excluded from the analysis because chemotherapy was administered concurrently (participant 16) and the training was done overground rather than on the treadmill (participant 18).

Demographic characteristics are shown in Table 1. The mean age was 46.0 years (SD=13.6), and the participants were 5.4 years (SD=8.8) postinjury. Walking ability before any training varied in the participants; the lowest WISCI II score was 9 (ambulates 10 m with walker, braces, and no physical assistance), and 4 participants achieved the highest score of 20 (ambulates 10 m with no devices, no braces, and no physical assistance). Nine participants had a self-selected walking speed of at least 0.50 m/s before precision training and, therefore, formed the fast group for precision training. There were 8 participants in the fast group for endurance training.

View this table:
Table 1.

Participant Characteristicsa

Changes in Walking Outcomes With Training

The changes in the SCI-FAP scores with 2 months of precision training varied considerably in the participants; scores ranged from −322 (improvement) to +11 (small decline). When the participants were grouped on the basis of pretraining self-selected speed (ie, <0.50 m/s and ≥0.50 m/s), those in the fast group showed, on average, no change in the SCI-FAP score, whereas those in the slow group showed a significantly greater (P=.005) mean improvement of −77 points (Tab. 2). Likewise, the changes in the SCI-FAP scores after 2 months of endurance training varied from −434 (improvement) to +2, and the participants in the slow group showed a significantly greater (P=.0001) change in the SCI-FAP score than those in the fast group, whose mean change was negligible (Tab. 2). The change in the SCI-FAP time mirrored the change in the total SCI-FAP score.

View this table:
Table 2.

Change Scores on Walking Measuresa

The mean changes in the 10MWT, 6WMT, and WISCI II scores are shown in Table 2. For the 10MWT (fast and self-selected) and 6MWT, participants in the fast and slow groups showed statistically similar improvements (P≥.05). There was minimal change in the WISCI II scores (self-selected and maximum) with either training type, regardless of pretraining walking speed.

Internal Responsiveness of Walking Outcomes With Precision Training

The SRMs for the measures used during precision training are shown in Figure 2A. Since few participants showed any change in the WISCI II scores (self-selected or maximum), we did not include this measure in further assessments of responsiveness. The SRM magnitude varied according to the measure (P<.001); however, all values were moderate. Post hoc analyses revealed that the fast 10MWT had a significantly larger SRM (0.79) than the other 4 measures (P<.005) and that the SRM of the self-selected 10MWT (0.64) was significantly larger than those of the 6MWT, SCI-FAP score, and SCI-FAP time.

Figure 2.
Figure 2.

Internal responsiveness of walking measures assessed after 2 months of precision training. (A) Standardized response means (SRM) for the 5 measures studied (horizontal axis). **Significant difference (P<.005) between given measure and other 4 measures. (B) SRMs for the 7 Spinal Cord Injury Functional Ambulation Profile (SCI-FAP) tasks (horizontal axis). Error bars represent the 95% confidence interval of the SRM (calculated by use of the jackknife procedure). TUG=Timed “Up & Go” Test, 10MWT=10-Meter Walk Test, 6MWT=Six-Minute Walk Test.

Figure 2B shows the SRMs of the 7 SCI-FAP tasks after precision training. The values were moderate for most tasks (range=0.37–0.65). The obstacle and Timed “Up & Go” Test tasks had the largest SRMs.

Internal Responsiveness of Walking Outcomes With Endurance Training

The SRMs of the SCI-FAP, 6MWT, and 10MWT with endurance training are shown in Figure 3A. The SRMs were moderate, with the exception of that of the 6MWT, which was large (0.88). Like the findings for precision training, the SRM magnitude varied with the measure (P<.001). Post hoc analyses showed that the 6MWT had the largest SRM (P<.001 for comparisons with the SRMs of the other 4 measures); that the SRM of the fast 10MWT (0.75), although significantly smaller than that of the 6MWT, was significantly larger than the SRMs of the other 3 measures (P<.005); and that the SCI-FAP score and self-selected 10MWT had similar SRMs (0.59 and 0.62, respectively).

Figure 3.
Figure 3.

Internal responsiveness of walking measures assessed after 2 months of endurance training. (A) Standardized response means (SRM) for the 5 measures studied (horizontal axis). ns=nonsignificant (P>.005) differences between the measures indicated; all other comparisons were significant. (B) SRMs for the 7 Spinal Cord Injury Functional Ambulation Profile (SCI-FAP) tasks (horizontal axis). Error bars represent the 95% confidence interval of the SRM (calculated by use of the jackknife procedure). TUG=Timed “Up & Go” Test, 10MWT=10-Meter Walk Test, 6MWT=Six-Minute Walk Test.

The SRMs of the 7 SCI-FAP tasks after endurance training are shown in Figure 3B. All were moderate. Overall, the SCI-FAP tasks had similar SRMs, with the exception of the step task, which had the largest SRM (0.65).

External Responsiveness of the SCI-FAP

The change in the SCI-FAP did not correlate with the changes in the self-selected 10MWT (r=−.09, P=.70), fast 10MWT (r=−.24, P=.39), or 6MWT (r=−.29, P=.21) during precision training. Likewise, the change in the SCI-FAP did not correlate with the changes in the self-selected 10MWT (r=.05, P=.84), fast 10MWT (r=.07, P=.80), or 6MWT (r=.17, P=.52) during endurance training.

Minimal Detectable Change in the Walking Measures

The intraclass correlation coefficients for the SCI-FAP total score, total time, and task scores ranged from .943 to .984. The intraclass correlation coefficients for the 6MWT, self-selected 10MWT, and fast 10MWT were .989, .981, and .977, respectively. The SEM and MDC95 of the SCI-FAP, self-selected 10MWT, fast 10MWT, and 6MWT are shown in Table 3.

View this table:
Table 3.

Minimal Detectable Changes in Walking Measures in Participants With Spinal Cord Injurya

Discussion

The focus of the SCI-FAP on real-world walking tasks, such as walking while opening a door, make it unique among the measures commonly used to assess walking after SCI. Here we showed that the SCI-FAP was responsive to a change in walking ability—a crucial quality for any useful measure—in participants who could walk independently (ie, could walk 5 m without manual assistance). However, the SCI-FAP was not more responsive than the 10MWT or the 6MWT and, in fact, might have been less responsive than these measures in certain situations (eg, depending on the type of intervention). Furthermore, the responsiveness of the SCI-FAP appeared to be dependent on walking proficiency. Changes in the SCI-FAP scores with training were small for participants with high proficiency in walking, suggesting that its use as a measure of change was limited in participants who walked fast (ie, could walk at ≥0.5 m/s). The 6MWT and 10MWT were more appropriate for measuring changes in participants with high proficiency in walking.

Internal Responsiveness of the SCI-FAP

The total SCI-FAP score was responsive to changes in walking function with both overground training and treadmill training. A change in the SCI-FAP score with training was consistently seen in participants who had initial walking speeds of less than 0.5 m/s, whereas participants with initial speeds of at least 0.5 m/s showed little change. The SCI-FAP has a ceiling effect,6 which may explain the reduced responsiveness in participants in the fast group.

Individual SCI-FAP tasks also were responsive. For precision training, the obstacle and Timed “Up & Go” Test tasks were most responsive to change. This result was not surprising because these tasks measure skills that were practiced in precision training (eg, obstacle clearance, turning 180°, and transferring between sitting and standing for breaks). Surprisingly, the step task was most responsive to change with endurance training. This task simulates stepping up and down a curb. Endurance training did not involve practice of this skill or any similar skills, such as stair climbing. However, the step task is believed to be the most challenging task on the SCI-FAP6; therefore, there is likely more room for improvement on this task than on other tasks. Indeed, the mean change in the step task was −25 points, whereas the mean changes in the other 6 tasks ranged from −7 to −11 points. Since the SRM is a ratio of the mean change to the variability of that change, larger change scores can lead to larger SRMs.

The responsiveness of the SCI-FAP time was, for the most part, equivalent to that of the total SCI-FAP score. The difference between the SCI-FAP score and the SCI-FAP time is the inclusion of an assistance rating in the score. This finding may suggest that the assistance rating is unnecessary. Further work is needed to examine how the time and assistance scores contribute to the total score and how the SCI-FAP can be scored to optimize its validity, reliability, and responsiveness.

Internal Responsiveness of the SCI-FAP Versus the 10MWT and the 6MWT

In participants who had SCI and could walk without manual assistance, the responsiveness of the SCI-FAP was found to be comparable to that of the 10MWT and the 6MWT. The 6MWT showed the greatest responsiveness with endurance training. This result is not surprising since endurance training aimed to increase the distance walked. More surprising was that fast walking speed showed the greatest responsiveness with precision training, and it was also one of the most responsive measures for endurance training. The psychometric properties of the fast 10MWT have not been well researched for people with SCI. It has high test-retest reliability in people with traumatic brain injury,25 chronic stroke,26 and Parkinson disease.27 A similar measure, the 50-Foot Walk Test, has been performed at maximum speed by people with SCI.4,28 Its reliability and validity have not been formally assessed in this population, nor has its responsiveness; however, the latter is speculated to be limited by a floor effect.29

External Responsiveness of the SCI-FAP

The SCI-FAP change scores did not correlate with the 10MWT and 6MWT change scores. This finding is contrary to what we expected, as previously we found moderately strong correlations between SCI-FAP scores and 10MWT and 6MWT scores.6 The present findings suggest that a change induced in the SCI-FAP by walking training does not necessarily parallel changes in the other 2 measures. Indeed, a few participants in the present study showed large improvements on the SCI-FAP but little change in self-selected walking speed (self-selected 10MWT), and 1 participant showed the opposite. There are considerable differences in the scoring of these measures. The SCI-FAP score reflects speed and the amount of assistance required to complete each task, whereas the 10MWT and 6MWT scores are dependent solely on speed. Therefore, if speed does not change but the amount of assistance required is reduced, then improvement on the SCI-FAP only may be seen. Another possible explanation for the lack of correlation is that the SCI-FAP measures aspects of walking that are not reflected by measures of speed and endurance on a flat surface, as suggested previously.6 If the SCI-FAP and 10MWT/6MWT measure different constructs, their change scores would not be expected to be related. Recently, however, the 10MWT was found to be a good estimator of several functional walking skills, such as walking along a curve, on a compliant surface, and over obstacles.30 Therefore, the lack of correlation between change scores is more likely the result of differences in scoring.

Detectable Change in the Walking Measures Studied

The MDC95 of the SCI-FAP for the participants in the present study was 96 points. This magnitude of improvement was not possible for most of the participants in the fast group, as their pretraining scores were lower than 96 points—further evidence that the SCI-FAP is not a responsive measure for people with a higher level of functioning. The MDC95 of the SCI-FAP tasks ranged from 9 to 21 points, with the exception of the step task (36 points). The large MDC95 for the step task reflected greater variability in baseline performance of all participants.

For the 6MWT, we found a SEM of 12 m. This SEM is similar to the SEMs reported for people with acute stroke and older people (20 m),31 patients with chronic stroke (18.6 m),26 and people with acute SCI (16.5 m).2 We calculated the MDC95 of the 6MWT to be 34 m, which is smaller than previous estimates (smallest real difference=45.8 m7). The SEM calculated for the self-selected 10MWT, 0.05 m/s, was similar to values previously reported for people with SCI7,32 as well as people with stroke and older people.31 Likewise, the MDC95 reported here for the self-selected 10MWT (0.15 m/s) was consistent with values previously reported for people with SCI (0.13 m/s)7 and Parkinson disease (0.18 m/s).27 The SEM and the MDC95 for the fast 10MWT—which have not been reported for people with SCI—were similar to those for the self-selected version of the test. In contrast, the MDC for a fast walking speed in people with Parkinson disease (0.25 m/s) exceeded the corresponding value for the self-selected walking speed.27

Limitations

The greatest limitation of the present study was the small number of participants. A sample of 20 is considered poor according to the guidelines of the Consensus-Based Standards for the Selection of Health Status Measurement Instruments.33,34 The small sample size affected the validity of the methods and results. For example, the accuracy of the SRM estimated with the jackknife procedure likely is not as high with a small sample. The findings need to be confirmed with a larger group of participants.

A second limitation concerns the efficacy of the interventions studied. An intervention used to examine an observed change must be effective.8,9 Although the efficacy of body-weight–supported treadmill training (ie, endurance training in the present study) has been established,28,35 the effectiveness of precision training has not. However, precision training has many similarities to other overground walking programs28,36 with which gains in walking similar to those achieved with body-weight–supported treadmill training have been reported. Mean group changes in self-selected walking speed and 6MWT distance were relatively modest in the present study, in particular, for precision training (Tab. 2). However, the change scores fell within the range of change scores reported in a recent review of gait training after SCI.35 Therefore, we have greater confidence in the evaluation of responsiveness with endurance training.

Finally, to be included in the present study, participants must have been able to walk at least 5 m without the assistance of another person (devices and braces were permitted). Although this eligibility criterion was necessary to ensure that participants could take part in precision training, it did limit the generalizability of the findings. The results presented here cannot be used to comment on the responsiveness of the SCI-FAP in participants who require manual assistance for overground walking. Likewise, responsiveness depends on the intervention received. The findings cannot be generalized to situations involving methods of walking training not studied here.

Conclusion

The SCI-FAP provides a unique dimension to walking assessment after SCI. It evaluates performance on a variety of common walking skills, providing clinicians with richer information concerning a client's walking ability, and therefore compliments the assessments of speed (10MWT) and endurance (6MWT) currently in use. The SCI-FAP is responsive to changes in walking function in people who walk at slow speeds (ie, <0.5 m/s) but without manual assistance. Establishing the MDC95 for the SCI-FAP aids in the interpretation of performance on the SCI-FAP; however, the MDC95 needs to be confirmed with a larger sample of participants divided into more homogeneous subgroups. Likewise, future work with a larger sample size should focus on confirming the responsiveness of the SCI-FAP in people with a lower level of walking function and on identifying the magnitude of change that is deemed clinically relevant by people with SCI.

Footnotes

  • Both authors provided concept/idea/research design, writing, data collection and analysis, and fund procurement. Dr Yang provided project management and facilities/equipment. Dr Musselman provided consultation (including review of manuscript before submission). The authors thank therapy staff members Kelly Brunton, Gregory Hendricks, Donna Livingstone, Sarah Pletch, Tina Carter, Michelle Krappala, and Katelyn Brown. They also thank Dr Tania Lam for comments on an earlier version of the article.

  • Ethics approval was granted by the Human Research Ethics Review Board, University of Alberta.

  • Some findings from this project were presented at the Canadian Physiotherapy Association National Congress; July 22–25, 2010, St John's, Newfoundland, Canada.

  • The Canadian Institutes of Health Research, the Christopher and Dana Paralysis Foundation, the Alberta Paraplegic Foundation, and the Rick Hansen Foundation funded this project.

  • The trial is listed in ClinicalTrials.gov (NCT01765153).

  • Received February 20, 2013.
  • Accepted October 4, 2013.

References

  1. ↵
    1. Ditunno JF Jr,
    2. Ditunno PL,
    3. Graziani V,
    4. et al
    . Walking Index for Spinal Cord Injury (WISCI): an international multicenter validity and reliability study. Spinal Cord. 2000;38:234–243.
    OpenUrlCrossRefPubMedWeb of Science
  2. ↵
    1. van Hedel HJ,
    2. Wirz M,
    3. Dietz V
    . Assessing walking ability in subjects with spinal cord injury: validity and reliability of 3 walking tests. Arch Phys Med Rehabil. 2005;86:190–196.
    OpenUrlCrossRefPubMedWeb of Science
  3. ↵
    1. van Hedel HJ,
    2. Wirz M,
    3. Curt A
    . Improving walking assessment in subjects with an incomplete spinal cord injury: responsiveness. Spinal Cord. 2006;44:352–356.
    OpenUrlCrossRefPubMedWeb of Science
  4. ↵
    1. Ditunno JF Jr,
    2. Barbeau H,
    3. Dobkin BH,
    4. et al
    . Validity of the Walking Scale for Spinal Cord Injury and other domains of function in a multicenter clinical trial. Neurorehabil Neural Repair. 2007;21:539–550.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Musselman KE,
    2. Yang JF
    . Walking tasks encountered by urban-dwelling adults and persons with incomplete spinal cord injuries. J Rehabil Med. 2007;39:567–574.
    OpenUrlCrossRefPubMedWeb of Science
  6. ↵
    1. Musselman KE,
    2. Brunton K,
    3. Lam T,
    4. Yang J
    . Spinal Cord Injury Functional Ambulation Profile: a new measure of walking ability. Neurorehabil Neural Repair. 2011;25:285–293.
    OpenUrlAbstract/FREE Full Text
  7. ↵
    1. Lam T,
    2. Noonan VK,
    3. Eng JJ,
    4. et al
    . A systematic review of functional ambulation outcome measures in spinal cord injury. Spinal Cord. 2008;46:246–254.
    OpenUrlCrossRefPubMedWeb of Science
  8. ↵
    1. Beaton DE,
    2. Bombardier C,
    3. Katz JN,
    4. Wright JG
    . A taxonomy for responsiveness. J Clin Epidemiol. 2001;54:1204–1217.
    OpenUrlCrossRefPubMedWeb of Science
  9. ↵
    1. Husted JA,
    2. Cook RJ,
    3. Farewell VT,
    4. Gladman DD
    . Methods for assessing responsiveness: a critical review and recommendations. J Clin Epidemiol. 2000;53:459–468.
    OpenUrlCrossRefPubMedWeb of Science
  10. ↵
    1. Yang JF,
    2. Musselman KE,
    3. Livingstone D,
    4. et al
    . Repetitive mass practice or focused precise practice for retraining walking after incomplete spinal cord injury? A randomized controlled trial. Neurorehabil Neural Repair. 2013 Nov 8 [Epub ahead of print]. doi:10.1177/1545968313508473.
  11. ↵
    1. Yang JF,
    2. Norton J,
    3. Nevett-Duchcherer J,
    4. et al
    . Volitional muscle strength in the legs predicts changes in walking speed following locomotor training in people with chronic spinal cord injury. Phys Ther. 2011;91:931–943.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Guyatt GH,
    2. Sullivan MJ,
    3. Thompson PJ,
    4. et al
    . The 6-minute walk: a new measure of exercise capacity in patients with chronic heart failure. Can Med Assoc J. 1985;132:919–923.
    OpenUrlAbstract
  13. ↵
    1. Ditunno PL,
    2. Ditunno JF
    . Walking Index for Spinal Cord Injury (WISCI II): scale revision. Spinal Cord. 2001;39:654–666.
    OpenUrlCrossRefPubMedWeb of Science
  14. ↵
    1. Kim MO,
    2. Burns AS,
    3. Ditunno JF Jr,
    4. et al
    . The assessment of walking capacity using the Walking Index for Spinal Cord Injury: self-selected versus maximal levels. Arch Phys Med Rehabil. 2007;88:762–767.
    OpenUrlCrossRefPubMedWeb of Science
  15. ↵
    1. Cohen J
    . Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Associates; 1988.
  16. ↵
    1. Liang MH,
    2. Fossel AH,
    3. Larson MG
    . Comparison of five health status instruments for orthopedic evaluation. Med Care. 1990;28:632–642.
    OpenUrlCrossRefPubMedWeb of Science
  17. ↵
    1. Stratford PW
    . Getting more from the literature: estimating the standard error of measurement from reliability studies. Physiother Can. 2004;56:27–30.
    OpenUrlCrossRef
  18. ↵
    1. Robinett CS,
    2. Vondran MA
    . Functional ambulation velocity and distance requirements in rural and urban communities: a clinical report. Phys Ther. 1988;68:1371–1373.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Andrews AW,
    2. Chinworth SA,
    3. Bourassa M,
    4. et al
    . Update on distance and velocity requirements for community ambulation. J Geriatr Phys Ther. 2010;33:128–134.
    OpenUrlPubMed
  20. ↵
    1. Perry J,
    2. Garrett M,
    3. Gronley JK,
    4. Mulroy SJ
    . Classification of walking handicap in the stroke population. Stroke. 1995;26:982–989.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    1. Schmid A,
    2. Duncan PW,
    3. Studenski S,
    4. et al
    . Improvements in speed-based gait classifications are meaningful. Stroke. 2007;38:2096–2100.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. van Hedel HJ
    , EMSCI Study Group. Gait speed in relation to categories of functional ambulation after spinal cord injury. Neurorehabil Neural Repair. 2009;23:343–350.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Forrest GF,
    2. Lorenz DJ,
    3. Hutchinson KJ,
    4. et al
    . Variability conundrum for walking measures after SCI. In: 2012 Neuroscience Meeting Planner. New Orleans, LA: Society for Neuroscience; 2012. Program No. 85.19. Available at: http://www.abstractsonline.com/Plan/ViewAbstract.aspx?sKey=382542fd-e438-4eff-a54b-b77858267d90&cKey=f5e06701-6740-4ef9-b7f4-c74be241c7d1&mKey={70007181-01C9-4DE9-A0A2-EEBFA14CD9F1}. Accessed February 15, 2013.
  24. ↵
    1. Maynard FM Jr,
    2. Braken MB,
    3. Creasey G,
    4. et al
    . International Standards for Neurological Functional Classification of Spinal Cord Injury. Spinal Cord. 1997;35:266–274.
    OpenUrlCrossRefPubMedWeb of Science
  25. ↵
    1. van Loo MA,
    2. Moseley AM,
    3. Bosman JM,
    4. et al
    . Test–re-test reliability of walking speed, step length and step width measurement after traumatic brain injury: a pilot study. Brain Inj. 2004;18:1041–1048.
    OpenUrlCrossRefPubMedWeb of Science
  26. ↵
    1. Flansbjer UB,
    2. Holmbäck AM,
    3. Downham D,
    4. et al
    . Reliability of gait performance tests in men and women with hemiparesis after stroke. J Rehabil Med. 2005;37:75–82.
    OpenUrlCrossRefPubMedWeb of Science
  27. ↵
    1. Steffen T,
    2. Seney M
    . Test-retest reliability and minimal detectable change on balance and ambulation tests, the 36-Item Short-Form Health Survey, and the Unified Parkinson Disease Rating Scale in people with parkinsonism. Phys Ther. 2008;88:733–746.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    1. Dobkin B,
    2. Apple D,
    3. Barbeau H,
    4. et al
    . Weight-supported treadmill vs over-ground training for walking after acute incomplete SCI. Neurology. 2006;66:484–493.
    OpenUrlCrossRef
  29. ↵
    1. Jackson AB,
    2. Carnel CT,
    3. Ditunno JF,
    4. et al
    . Outcome measures for gait and ambulation in the spinal cord injury population. J Spinal Cord Med. 2008;31:487–499.
    OpenUrlPubMedWeb of Science
  30. ↵
    1. Labruyère R,
    2. van Hedel HJ
    . Curved walking is not better than straight walking in estimating ambulation-related domains after incomplete spinal cord injury. Arch Phys Med Rehabil. 2012;93:796–801.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Perera S,
    2. Mody SH,
    3. Woodmann RC,
    4. Studenski SA
    . Meaningful change and responsiveness in common physical performance measures in older adults. J Am Geriatr Soc. 2006;54:743–749.
    OpenUrlCrossRefPubMedWeb of Science
  32. ↵
    1. Musselman KE
    . Clinical significance testing in rehabilitation research: what, why and how? Phys Ther Rev. 2007;12:287–296.
    OpenUrlCrossRef
  33. ↵
    1. Terwee CB,
    2. Bot SD,
    3. de Boer MR,
    4. et al
    . Quality criteria were proposed for measurement properties of health status questionnaires. J Clin Epidemiol. 2007;60:34–42.
    OpenUrlCrossRefPubMedWeb of Science
  34. ↵
    1. Mokkink LB,
    2. Terwee CB,
    3. Patrick DL,
    4. et al
    . COSMIN checklist manual. Available at: http://www.cosmin.nl/images/upload/File/COSMIN%20checklist%20manual%20v9.pdf. January 2012. Accessed September 19, 2013.
  35. ↵
    1. Yang JF,
    2. Musselman KE
    . Training to achieve over ground walking after spinal cord injury: a review of who, what, when and how. J Spinal Cord Med. 2012;35:293–304.
    OpenUrlCrossRefPubMedWeb of Science
  36. ↵
    1. Musselman KE,
    2. Fouad K,
    3. Misiaszek JE,
    4. Yang JF
    . Training of walking skills overground and on the treadmill: case series on individuals with incomplete spinal cord injury. Phys Ther. 2009;89:601–611.
    OpenUrlAbstract/FREE Full Text
View Abstract
Back to top
Email
Print
Spinal Cord Injury Functional Ambulation Profile: A Preliminary Look at Responsiveness
Kristin E. Musselman, Jaynie F. Yang
Physical Therapy Feb 2014, 94 (2) 240-250; DOI: 10.2522/ptj.20130071

Citation Manager Formats

Download Powerpoint
Save to my folders

Spinal Cord Injury Functional Ambulation Profile: A Preliminary Look at Responsiveness
Kristin E. Musselman, Jaynie F. Yang
Physical Therapy Feb 2014, 94 (2) 240-250; DOI: 10.2522/ptj.20130071
del.icio.us logo Digg logo Reddit logo Technorati logo Twitter logo CiteULike logo Connotea logo Facebook logo Google logo Mendeley logo

Subjects