Validity and reliability of the 2-minute walk test in individuals with spinal cord injury

Study population

For this multicentre study, individuals with SCI were recruited at the Spinal Cord Injury Center Balgrist and the Swiss Paraplegic Center Nottwil. The inclusion criteria were: acute or chronic SCI, ≥18 years and a minimum walking speed of 0.17 m/s (~corresponding to the ability to perform a 10-meter walk test (10MWT) in <60 s), and ability to walk without physical assistance. We excluded individuals who walked <60 meters in 6 min due to a potential high variability of walking performance as this may occur in individuals with poor walking function [16]. The exclusion criteria were: current orthopedic problems, major psychosis or depression, and history of severe heart condition.

Procedure

This study was approved by the Ethical Committee of the Canton of Zurich and the Ethics Committee for Northwest/Central Switzerland (BASEC 2020-01473) and was conducted in accordance with Good Clinical Practice guidelines and the Declaration of Helsinki. Prior to enrollment, written informed consent was obtained from all participants. Individuals were invited to participate on two test days, separated by 1–7 days. A maximum of 7 days has been chosen because it can be assumed that the walking function remains stable during this time period. On the first day, a familiarization run of the 2mWT was performed. After a break of at least 30 min, the 2mWT and the 10MWT were performed in randomized order, again separated by at least 30 min of rest. The WISCI II was applied and scored based on the performance in the 10MWT. The second testing day consisted of 2mWT and 6mWT, performed in randomized order and separated by a 30 min break.

Measures

Examiners from both centers were trained all together in performing the assessments to optimize the standardized collection of all outcome measures.

2mWT and 6mWT

The walk tests were performed according to the Guidelines of the American Thoracic Society [19], except that the hallway length was 35 meters (instead of 30 meters) in both study centers. An examiner accompanied each subject for safety reasons, walking behind the subject to allow them to set the pace.

Any braces and/or habitual assistive devices were permitted but needed to be kept consistent on both testing days. Subjects were instructed to walk as far as possible, but safely, during the respective test time of 2 or 6 min. Standardized instructions were read out to the participants. Every minute during the test, they were informed about the remaining time and were encouraged to keep up the good work. No other communication occurred during the test.

Subjects were allowed to take rest breaks if needed, but time continued to run during the break.

The 6mWT has shown very good construct validity (10MWT: r = 0.86) [8] and excellent test-retest reliability (ICC = 0.99) [12] in individuals with chronic SCI.

10MWT

The 10MWT was performed with a flying start, i.e., subjects were instructed to walk a total distance of 14 meters including two meters to accelerate and two meters to decelerate. Time was measured for the middle ten meters and rounded to the next tenth of a second.

Participants performed a total of four 10MWT runs: two at their preferred walking speed (10MWT self), and two at their maximal walking speed (10MWT max), i.e., as fast as possible, but still safe. The average of the two runs each was taken as the final result for each variant.

The 10MWT has shown very good construct validity (TUG: r = 0.89; 6mWT: r = 0.95) [20] and excellent test-retest reliability (ICC = 0.983 [20]) in individuals with SCI.

WISCI II

The WISCI II is an ordinal scale to assess walking capacity [21]. It captures the extent and nature of assistance a person with SCI requires to walk ten meters. This assessment index includes a rank ordering along a dimension of impairment, from the level of most severe impairment (level 0) to least severe impairment (level 20). The level is based on the use of devices, braces, and physical assistance of one or more persons. Each successive level is a less impaired level than the former. The ranking is based on the severity of the impairment and not on functional independence in the environment.

For the present study, the WISCI II was scored based on the performance of the 10MWT.

The WISCI II has shown good construct validity (TUG: 0.76; 10MWT: 0.68; 6mWT: 0.60) [8] and excellent test-retest reliability (self-selected WISCI: ICC = 0.994; maximal WISCI: ICC = 0.995) [22].

Statistics

The sample size of 50 was chosen based on the COSMIN guidelines [17].

There was no missing data as, in accordance with the study protocol, the three drop-outs have been replaced. No further cleaning was necessary since all datasets were complete and plausible. Data were analyzed using RStudio for Mac, version 1.3.1093. Descriptive data are reported using mean ± standard deviation or median (range).

Construct validity

Data were tested for normality using the Shapiro–Wilk test. If data were normally distributed, construct validity was tested using the Pearson correlation. Otherwise, a Spearman rank correlation was applied.

According to the guide from Evans et al. [23], the following ranges for the correlation coefficient (R) were used for the interpretation of the strength of the association: 0.00–0.19: “very weak”, 0.20–0.39: “weak”, 0.40–0.59: “moderate”, 0.60–0.79: “strong” and 0.80–1.0: “very strong”.

Based on studies in other neurological populations [14,15,16], a “very strong” Pearson correlation between the 2mWT and the 6mWT was predefined as requirement to recommend the use of the 2mWT for routine clinical use in individuals with SCI.

Test-retest reliability

Test-retest reliability was determined by calculating the interclass correlation coefficient (ICC) (two-way mixed effects, total agreement) between the 2mWT from the first and the second test day [24]. To interpret the ICC values, we have referred to the recommendation from Koo et al. [24], <0.5: “poor”, 0.5–0.76: “moderate”, 0.75–0.9 “good” and >0.9 “excellent”. ICCs > 0.8 are acceptable for clinical work [25].

In addition, a Bland–Altman plot was created to estimate the absolute agreement of the two measurements of the 2mWT.

The SEM \(\left( }}}}} = }}}}} \times \sqrt }}}}}} \right)} } \right)\) [26] and the MDC \(\left( }}}}} = }}}}} \times 1.96 \times \sqrt 2 } \right)\) [27] were calculated based on the respective formulas.

Minute intervals

A paired t-test was used to compare the distances walked in the 2mWT in the first and the 2nd min. Moreover, a Friedman test was applied to compare the minute intervals of the 6mWT with paired post-hoc Wilcoxon Tests (Bonferroni corrected), if applicable. A significance level of p < 0.05 was selected.

Subgroup analysis

Two subgroup analysis were performed: (1) the influence of time since injury on construct validity and test-retest reliability and (2) the influence of walking performance on SEM and MDC was analyzed, each by creating two subgroups.

(1)

Although the maximal time window for the two test days was narrow with 7 days it cannot be excluded that individuals with a sub-acute SCI could show a spontaneous improvement in walking performance within this time window. Therefore, the study sample was divided into an acute/sub-acute (0–6 months after injury) and a chronic (>6 months after injury) SCI group.

(2)

To investigate whether SEM and MDC are influenced by the walking performance, the study sample was separated into slow and fast walkers. Community ambulation is often measured at crosswalks, where speed is the primary concern [28]. In the US the recommended walking speed to safely cross an intersection range from 0.9 to 1.2 m/s [29]. Since the median of the speed in the 2mWT of our sample was 0.9 m/s we decided to use this as cut off to divide our sample into two groups.

留言 (0)

沒有登入
gif