460
M. van Bloemendaal et al.
Table III. Number of footstep repetitions for reliable spatiotemporal gait analysis system (SGAS) data for the barefoot comfortable gait
speed condition by the stationary placed camera (n = 25)
Footsteps, n Step length
ICC 3.1 (95% CI) [%CoV] Step time
ICC 3.1 (95% CI) [%CoV] Stance time
ICC 3.1 (95% CI) [%CoV] Double support time
ICC 3.1 (95% CI) [%CoV]
1
2
3
4
5
6
7
8
9 0.87
0.97
0.98
0.99
0.99
1.00
1.00
1.00
1.00 0.86
0.96
0.98
0.99
0.99
1.00
1.00
1.00
1.00 0.87
0.96
0.98
0.99
0.99
1.00
1.00
1.00
1.00 0.85
0.97
0.97
0.99
0.99
0.99
1.00
1.00
1.00
(0.73–0.94)
(0.93–0.99)
(0.95–0.99)
(0.97–1.00)
(0.98–1.00)
(0.99–1.00)
(1.00–1.00)
(1.00–1.00)
(1.00–1.00)
[18.7]
[12.8]
[10.4]
[7.4]
[6.2]
[5.1]
[3.5]
[2.6]
[2.3]
(0.70–0.93)
(0.91–0.98)
(0.95–0.99)
(0.98–1.00)
(0.98–1.00)
(0.99–1.00)
(0.99–1.00)
(1.00–1.00)
(1.00–1.00)
[21.4]
[14.4]
[10.3]
[7.5]
[5.9]
[4.8]
[4.0]
[2.6]
[1.3]
(0.73–0.94)
(0.92–0.98)
(0.94–0.99)
(0.98–1.00)
(0.98–1.00)
(0.99–1.00)
(0.99–1.00)
(1.00–1.00)
(1.00–1.00)
[24.0]
[17.0]
[13.0]
[8.7]
[7.2]
[5.7]
[4.6]
[3.0]
[1.7]
(0.70–0.93)
(0.92–0.98)
(0.94–0.99)
(0.97–1.00)
(0.98–1.00)
(0.99–1.00)
(0.99–1.00)
(1.00–1.00)
(1.00–1.00)
[49.3]
[33.5]
[30.8]
[19.6]
[17.6]
[13.9]
[11.2]
[7.1]
[5.0]
ICC: intraclass correlation coefficient; CI: confidence interval; %CoV: coefficient of variation (the 95% limits of agreement interval for n vs 10 footsteps divided
by the mean value for 10 footsteps) as percentage.
GAITRite® data (Table II). In addition, data were col-
lected for between 150 and 256 valid trials of strides
across conditions for the moving setup and for the
conditions slow gait speed and toe walking from the sta-
tionary setup of the SGAS and the GAITRite® system.
Concurrent validity and measurement error of the
stationary SGAS camera
Excellent agreement for step length, step time, stance
time, swing time, stride length, and stride time was
found between the SGAS and the GAITRite® system
in all walking conditions (ICC ≥ 0.95 and lower limit of
the 95% CIs ≥ 0.78; Table II). Double support time sho-
wed poor agreement for the condition comfortable gait
speed (ICC = 0.21 and lower limit of the 95% CI = 0.02),
moderate agreement for the condition toe walking
(ICC = 0.50 and lower limit of the 95% CI = –0.06), and
good agreement for the conditions of shod walking and
slow gait speed (ICC ≤ 0.83 and lower limit of the 95%
CIs ≤ 0.75, Table II). Moreover, systematic differences
for double support time were found between systems for
the conditions of comfortable gait speed (mean –0.05
s and SD 0.07 s), slow gait speed (mean 0.28 s and SD
0.49 s), and toe walking (mean –0.17 s and SD 0.06 s).
SEM and RCs were below the respective thresholds of
5 and 8% for all spatiotemporal gait parameters (≤ 2.4%
and ≤ 6.3%, respectively), except for swing time (SEM
range 3.9–9.4% and RC range 10.6–19.2%) and double
support time (SEM range 9.3–40.4% and RC range
24.8–95.5%; Table II).
Concurrent validity and measurement error of the
moving SGAS camera
Excellent agreement for step length, step time, stance
time, stride length, and stride time was found between
the moving SGAS camera and the GAITRite® system
with barefoot walking at comfortable speed (ICC
≥ 0.97 and lower limit of the 95% CIs ≥ 0.95; Table
II). Paired t-tests revealed no differences exceeding the
cut-off points of measurement error. Moderate agre-
ement between systems was found for double support
time and good agreement for swing time. SEM and
RCs were below the respective thresholds of 5% and
8% for all spatiotemporal gait parameters (≤ 2.5% and
≤ 6.1%, respectively) except for double support time
(SEM 10.2% and RC 25.6%).
Minimum number of footsteps needed for adequate
reliability
Ten valid trials of footsteps were available for 25 sub-
jects. Two footsteps were required to obtain excellent
reliability scores (ICC > 0.90) for all 4 assessable spa-
Fig. 4. Bland-Altman plots of step length, step time, and stance time for 4 footsteps averaged with the stationary spatiotemporal gait analysis
system (SGAS).
www.medicaljournals.se/jrm