Gait & Posture 33 (2011) 651–655
Contents lists available at ScienceDirect
Gait & Posture
journal homepage: www.elsevier.com/locate/gaitpost
Reliability of center of pressure measures of postural stability in healthy older
adults: Effects of postural task difficulty and cognitive load
Mojgan Moghadam a, Hassan Ashayeri b, Mahyar Salavati c, Javad Sarafzadeh a,*,
Keyvan Davatgaran Taghipoor d, Ahmad Saeedi e, Reza Salehi f
a
Department of Physical Therapy, Faculty of Rehabilitation Sciences, Tehran University of Medical Sciences, Tehran, Iran
Department of Basic Sciences, Faculty of Rehabilitation Sciences, Tehran University of Medical Sciences, Tehran, Iran
Department of Physical Therapy, University of Social Welfare and Rehabilitation Sciences, Tehran, Iran
d
Department of Aging Research, University of Social Welfare and Rehabilitation Sciences, Tehran, Iran
e
Department of Education Management, Ahvaz Shahid Chamran University, Ahvaz, Iran
f
Department of Physical Therapy, School of Rehabilitation Sciences, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran
b
c
A R T I C L E I N F O
A B S T R A C T
Article history:
Received 25 December 2009
Received in revised form 17 January 2011
Accepted 22 February 2011
Postural instability is a major risk factor of falling in the elderly. It is well documented that postural
control may decline while performing a concurrent cognitive task and this effect increases with age.
Despite the extensive use of dual tasking in balance assessment protocols, a lack of sufficient reliability
information is evident. This study determines the reliability of the postural stability measures in older
adults, assessed under single and dual-task conditions and different levels of postural difficulty. Sixteen
older adults completed quiet stance postural measurements at three levels of difficulty (rigid surfaceeyes open, rigid surface-eyes closed, and foam surface-eyes closed), with or without performing a
concurrent backward counting task, in two sessions 1 week apart. Force plate data was used to calculate
center of pressure (COP) parameters including mean velocity, phase plane portrait, area (95% confidence
ellipse), standard deviation (SD) of amplitude, and SD of velocity. Intraclass correlation coefficient (ICC),
standard error of measurement (SEM), coefficient of variation (CV), and minimal metrically detectable
change (MMDC) were calculated for each COP measure in all test conditions. Mean velocity, total phase
plane, phase plane in ML direction, and SD of velocity in ML direction were the most reliable COP
measures across all test conditions. ICC values were consistently higher in ML direction compared with
AP direction. In general, velocity-related COP measures in ML direction showed to be highly reliable.
Further research may explore the predictive and evaluative value of these COP parameters.
ß 2011 Elsevier B.V. All rights reserved.
Keywords:
Reliability
Center of pressure
Postural control
Dual-task
Elderly
1. Introduction
Falls are a major health problem in the elderly, which
contribute to substantial morbidity and mortality. Approximately
one-third of community-residing people over 65 years of age fall
every year, with this rate rising steadily with age [1–3]. Falls
involve multiple factors categorized into intrinsic (patient-related), extrinsic (environment-related), and behavioral (activityrelated) risk factors [1,4]. Postural instability has been identified as
a major intrinsic risk factor of falling which can potentially be
influenced with intervention [1,2,5]. It is well documented that
even healthy older adults show a marked decline in the ability to
control upright posture compared with young adults [6–8]. Early
* Corresponding author at: Physical Therapy Department, Faculty of Rehabilitation Sciences, Tehran University of Medical Sciences, Shahnazari St., Mirdamad
Blvd, Tehran, Iran. Tel.: +98 21 222 28051; fax: +98 21 222 20946.
E-mail address: j.sarafzadeh88@gmail.com (J. Sarafzadeh).
0966-6362/$ – see front matter ß 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.gaitpost.2011.02.016
detection of balance limitations is required to identify older people
who are at risk of falls and adopt effective fall prevention
strategies.
Force platform posturography is a commonly used method for
quantifying balance performance. Various parameters derived
from the center of pressure (COP) signal provide different types of
information on postural control mechanisms [9]. Numerous
studies have used a variety of COP measures to detect between
group differences [8,10], predict falling risk [11–14], and evaluate
the efficacy of balance training programs [15,16]. Because of
methodologic differences and controversial results of such studies,
as well as the intrinsic variability of the COP signal [17], it seems
necessary to first establish reliability of COP measure before using
it for the above purposes.
Several studies have reported the reliability of COP parameters
during quiet standing in the elderly. However, most of them did
not consider different sensory environments in their postural
measurements [18–20]. Given the well-documented decline in
sensorimotor integration with aging [21], previous studies have
652
M. Moghadam et al. / Gait & Posture 33 (2011) 651–655
demonstrated that the more challenging sensory conditions (e.g.
standing on a compliant surface with eyes closed) in evaluating
postural control were necessary to facilitate the discrimination
between age groups and identification of older people at high risk
of falls [21,22]. Because the COP measures are dependent on
experimental setup and protocol [18], it seems necessary to assess
test–retest reliability of COP measures by incorporating different
levels of postural difficulty via manipulation of both visual and
somatosensory inputs.
Assessing balance while attention is fully directed toward the
postural task may limit the research validity as in real-life
situations attention is usually divided between postural control
and other tasks. Interestingly, some falls in the elderly occur when
they perform multiple tasks [23]. In recent years, dual-tasking
postural control has been a growing topic of research with older
people [24]. However, most of the studies that used dual-task force
platform assessments of postural control did not provide any
information on reliability of the measurement protocols. The study
of Swanenburg et al. [20] is the only available work reporting
reliability for COP parameters measured during quiet standing in
single and dual-task conditions. In their assessment protocol,
however, the level of postural task difficulty was not manipulated
using different support surfaces [20]. In addition, some COP
measures that could specifically explain different aspects of the
nature of COP signal [9] were not included in their work [20].
The purpose of the present study was, therefore, to determine
the test–retest reliability of quantitative postural control measures
(each representing a unique dimension of COP position, velocity, or
both) in healthy older adults, assessed under single and dual-task
conditions and under different levels of postural task difficulty.
2. Methods
2.1. Participants
Sixteen community dwellers aged 60–83 (10 females and 6 males,
age = 69.60 4.5
years,
height = 161.4 6.22 cm,
weight = 68.65 9.57 kg,
mean SD) participated in the present study. Criteria for inclusion in the study were
as follows: participants had to be 60 years or older, be able to walk 10 m with or
without a walking aid, be able to stand independently for 90 s, live independently in
the community, and be able to understand and follow verbal instructions (score more
than 24 on the Mini-Mental State Examination). Subjects were excluded if they had
serious neurological or musculoskeletal disorders, or significant visual and auditory
impairments. All subjects signed an informed consent form approved by the Ethics
Committee at Tehran University of Medical Sciences.
2.2. Procedure
Postural stability was assessed by the same rater, in the same laboratory
environment, and in two sessions spaced 1 week apart. Quiet standing postural
sway was recorded at three levels of postural difficulty: (1) standing on the force
platform with eyes open (rigid-open), (2) standing on the force platform with eyes
closed (rigid-closed), and (3) standing on a foam surface with eyes closed (foamclosed). The aim of manipulating vision and somatosensory inputs was to change
the level of postural task difficulty and not simply provide different sensory
conditions. Three levels of postural task difficulty were: easy (rigid-open), moderate
(rigid-closed), and difficult (foam-closed). Subjects stood quietly while barefoot,
looking straight ahead, feet at 50% hip-to-hip distance, and arms at their sides. In
eyes closed conditions, vision was eliminated by asking subjects to wear a blindfold.
For the foam-closed condition, the force platform was covered with a 10-cm thick
piece of foam. Center of pressure (COP) data were obtained using a strain gauge
Bertec 4060-10 force platform and Bertec AM-6701 amplifier (Bertec Corp.,
Columbus, OH). Data were sampled at 100 Hz and transformed to obtain COP
values.
In dual task conditions, subjects were required to perform the postural standing
task concurrently with a cognitive task. The cognitive task was backward counting
by 7’s or 3’s (based on each subject’s counting abilities) as fast and as accurately as
possible for 30 s, beginning with a randomly selected number from a range of 111–
129. All subjects succeeded in counting backwards in steps of sevens as concurrent
cognitive task. Evaluation of cognitive task performance included the number of
responses and the mistakes made by the participant during the test period.
Collectively, participants were exposed to six experimental conditions in each
session. For each condition, three trials were performed. Each trial lasted for 30 s
and was followed by a rest period of 1 min. The three postural conditions were
Table 1
The formulae for calculating the COP measures.
Parameter
SD of amplitude (mm)
AP
ML
SD of velocity (mm/s)
AP
ML
Formula
rffiP
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ðx x̄Þ2
i
s x ¼ rffiffiffiffiffiffiN1
P ffiffiffiffiffiffiffiffiffiffi2ffiffi
ðyi ȳÞ
sy ¼
N1
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
rffiP
ðv v̄Þ2
x
x
xi
s vx ¼ rffiffiffiffiffiffiffiN1
where vxi ¼ tiþ1 t i
iþ1
i
P ffiffiffiffiffiffiffiffiffiffiffi2ffi
ðvy v̄Þ
y
y
i
s vy ¼
where vyi ¼ tiþ1 t i
N1
iþ1
i
Phase plane portrait (arbitrary unit)
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
s rx ¼ s 2x þ s 2vx
AP
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ML
s ry ¼ s 2y þ s 2vy
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
s rx ¼ s 2rx þ s 2ry
Total
T qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
X
ðxtþ1 xt Þ2 þ ðytþ1 yt Þ2
Mean velocity (mm/s)
v̄ ¼ 1T
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1
ðx x̄Þðy ȳÞ
2
Area (mm )
A ¼ 2pF 0:05½2;N2 s 2x s 2y þ s 2xy where s xy ¼ i N1i
COP: center of pressure; SD: standard deviation; AP: anteroposterior; ML:
mediolateral.
presented randomly. In each postural condition, the single or dual-task testing was
randomly performed, as well. The whole experiment time was about 35 min.
COP parameters calculated from force plate signal were SD of amplitude in AP
and ML directions, phase plane portrait in AP and ML directions, SD of velocity in AP
and ML directions, mean velocity and area (95% confidence ellipse). Phase plane
portrait, a less commonly used measure, provided information on static and
dynamic dimensions of postural control via considering both position and velocity
of COP [25]. The COP parameters are defined in Table 1.
2.3. Statistics
The mean of three trials of the COP parameters in each condition was used for
statistical analysis. The level of significance was set at p < 0.05 for all statistical
tests.
Paired t-tests on the differences of scores obtained at test and retest sessions
were used to ensure the absence of any systematic bias [26].
A two-way random model of the intraclass correlation coefficient (ICC2,3) was
used to estimate the relative reliability. For each ICC, a 95% confidence interval (CI)
was reported to indicate the precision of the estimates. Munro’s classification for
reliability coefficients was used to report the degree of reliability: 0.00–0.25 – little,
if any correlation; 0.26–0.49 – low correlation; 0.50–0.69 – moderate correlation;
0.70–0.89 – high correlation and 0.90–1.00 – very high correlation [27].
To assess absolute reliability, the standard error of measurement
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
(SEM ¼ mean square error) and 95% confidence intervals (CI) were calculated
to provide an estimate of the amount of error associated with the measurement in
the same units as the measurement [26]. Minimal metrically detectable change
(MMDC), or change that could be considered clinically significant between two
times of measurement, was defined as 95% CI of SEM of a COP measure (1.96 SEM)
[28]. In addition, the coefficient of variation (CV) was determined for the
comparison of absolute reliability between different COP measures (SD/
mean 100). This was achieved by calculating mean CV from individual CVs [26].
3. Results
Table 2 shows mean scores and standard deviations of COP
measures for all test conditions. There was no significant difference
between test and retest mean scores for any COP measures in all
condition, which shows absence of any systematic bias (p > 0.05)
(Table 2).
The calculated ICC, SEM, MMDC, and CV values are provided in
Tables 3 and 4 for single and dual-task conditions, respectively. The
COP measures across most conditions had moderate to very high
reliability (ICC range, 0.51–0.98). The results were different for
each COP measure across various test conditions. No clear pattern
was observed with respect to the level of postural difficulty and
task condition. However, several trends were evident. For both
single and dual-task conditions, mean velocity, total phase plane,
phase plane (ML), and SD of velocity (ML) were the most reliable
COP measures having high to very high reliability across all levels
of postural difficulty (ICC range, 0.70–0.98). Patterns of the CV
653
M. Moghadam et al. / Gait & Posture 33 (2011) 651–655
Table 2
Descriptive data for test–retest COP measures in different test conditions.
Task
Rigid-open
Test
SD amplitude (AP)
SD velocity (AP)
Phase plane (AP)
SD amplitude (ML)
SD velocity (ML)
Phase plane (ML)
Mean velocity
Area (95% ellipse)
Total phase plane
Single
Dual
Single
Dual
Single
Dual
Single
Dual
Single
Dual
Single
Dual
Single
Dual
Single
Dual
Single
Dual
0.33
0.39
1.33
1.35
1.37
1.41
0.19
0.22
0.90
0.91
0.92
0.94
1.37
1.39
1.17
1.52
1.66
1.69
(0.07)
(0.08)
(0.14)
(0.12)
(0.15)
(0.11)
(0.06)
(0.08)
(0.12)
(0.13)
(0.13)
(0.14)
(0.16)
(0.16)
(0.68)
(0.80)
(0.19)
(0.16)
Rigid-closed
Retest
p
Test
0.35
0.39
1.36
1.40
1.41
1.46
0.22
0.26
0.89
0.91
0.92
0.95
1.39
1.44
1.54
1.93
1.69
1.75
0.25
0.97
0.51
0.17
0.23
0.38
0.25
0.11
0.24
0.41
0.23
0.15
0.58
0.05
0.06
0.17
0.37
0.66
0.34
0.45
1.51
1.63
1.57
1.69
0.22
0.25
0.91
0.97
0.95
0.94
1.49
1.61
1.46
2.19
1.86
1.98
(0.09)
(0.14)
(0.12)
(0.13)
(0.12)
(0.11)
(0.11)
(0.10)
(0.09)
(0.11)
(0.10)
(0.12)
(0.13)
(0.14)
(1.26)
(1.79)
(0.14)
(0.13)
(0.06)
(0.12)
(0.17)
(0.21)
(0.16)
(0.23)
(0.11)
(0.09)
(0.13)
(0.18)
(0.15)
(0.10)
(0.15)
(0.20)
(1.15)
(1.51)
(0.23)
(0.25)
Foam-closed
Retest
p
Test
0.37
0.42
1.52
1.53
1.57
1.60
0.20
0.26
0.91
0.96
0.94
1.01
1.51
1.55
1.35
2.06
1.82
1.89
0.08
0.40
0.90
0.06
0.66
0.05
0.37
0.54
0.92
0.51
0.77
0.17
0.84
0.07
0.86
0.86
0.45
0.69
0.74
0.85
2.71
2.69
2.81
2.82
0.56
0.56
1.60
1.52
1.70
1.62
2.64
2.58
7.84
9.13
3.30
3.27
(0.06)
(0.14)
(0.16)
(0.18)
(0.16)
(0.20)
(0.09)
(0.09)
(0.10)
(0.14)
(0.10)
(0.19)
(0.13)
(0.19)
(0.55)
(1.50)
(0.17)
(0.24)
(0.09)
(0.16)
(0.61)
(0.78)
(0.61)
(0.78)
(0.16)
(0.16)
(0.33)
(0.52)
(0.35)
(0.53)
(0.53)
(0.75)
(2.95)
(4.36)
(0.66)
(0.93)
Retest
p
0.69
0.80
2.76
2.91
2.85
3.02
0.51
0.65
1.54
1.71
1.63
1.85
2.66
2.85
6.57
9.89
3.30
3.56
0.06
0.34
0.56
0.51
0.86
0.28
0.24
0.72
0.56
0.18
0.46
0.14
0.90
0.32
0.09
0.67
0.92
0.65
(0.01)
(0.16)
(0.73)
(0.85)
(0.72)
(0.85)
(0.08)
(0.17)
(0.45)
(0.53)
(0.44)
(0.52)
(0.69)
(0.81)
(1.73)
(3.61)
(0.82)
(0.98)
Values are mean standard deviation (SD). p refers to p-values of paired t-test on test–retest differences. COP: center of pressure; AP: anteroposterior; ML: mediolateral. Units of
COP measures are as follows: cm (SD of amplitude); cm/s (SD of velocity/mean velocity); cm2 (area). Phase plane is in an arbitrary unit.
Table 3
Test–retest reliability analysis of COP measures for single-task condition.
Rigid-open
SD amplitude (AP)
SD velocity (AP)
Phase plane (AP)
SD amplitude (ML)
SD of velocity (ML)
Phase plane (ML)
Mean velocity
Area (95% ellipse)
Total phase plane
Rigid-closed
Foam-closed
ICC (95% CI)
SEM
MMDC
CV (%)
ICC (95% CI)
SEM
MMDC
CV (%)
ICC (95% CI)
SEM
MMDC
CV (%)
0.76
0.64
0.70
0.69
0.96
0.95
0.89
0.86
0.88
0.15
0.30
0.28
0.17
0.08
0.11
0.20
1.42
0.23
0.30
0.59
0.55
0.35
0.16
0.22
0.39
2.78
0.46
9.61
5.77
4.67
20.24
2.60
3.30
4.07
20.17
3.83
0.75
0.38
0.33
0.81
0.92
0.88
0.70
0.80
0.70
0.11
0.44
0.44
0.16
0.14
0.18
0.30
1.60
0.42
0.22
0.86
0.86
0.31
0.28
0.36
0.59
3.13
0.83
9.05
7.17
7.01
13.96
3.96
4.47
5.40
16.62
5.11
0.78
0.65
0.67
0.68
0.86
0.84
0.78
0.67
0.78
0.16
1.48
1.45
0.27
0.61
0.64
1.16
4.85
1.38
0.33
2.90
2.84
0.54
1.19
1.25
2.27
9.50
2.69
7.21
11.64
12.14
12.16
10.62
11.06
8.98
18.43
9.25
(0.13,
(0.00,
(0.00,
(0.00,
(0.86,
(0.79,
(0.58,
(0.44,
(0.53,
0.94)
0.91)
0.93)
0.92)
0.99)
0.99)
0.97)
0.96)
0.97)
(0.10,
(0.00,
(0.00,
(0.27,
(0.65,
(0.50,
(0.00,
(0.18,
(0.00,
0.94)
0.85)
0.84)
0.95)
0.98)
0.97)
0.92)
0.95)
0.93)
(0.19,
(0.00,
(0.00,
(0.00,
(0.42,
(0.39,
(0.03,
(0.00,
(0.52,
0.94)
0.92)
0.92)
0.92)
0.96)
0.96)
0.95)
0.91)
0.95)
COP: center of pressure; ICC: intraclass correlation coefficient; SEM: standard error of measurement; MMDC: minimal metrically detectable change; CV: coefficient of
variation;SD: standard deviation; AP: anteroposterior; ML: mediolateral. Units of COP measures are as follows: cm (SD of amplitude); cm/s (SD of velocity/mean velocity);
cm2 (area). Phase plane is in an arbitrary unit.
values were greatly consistent with the ICCs so that for most
conditions mean velocity, total phase plane, phase plane (ML), and
SD of velocity (ML) had lower values relative to other measures (CV
range, 1.85–15.54%) (Tables 3 and 4).
In addition, for phase plane and SD of velocity, the ICC values
were consistently higher in the ML than the AP direction. A similar
pattern was evident for CV and SEM values, which were lower in
the ML than the AP direction except for the foam-closed dual-task
condition (Tables 3 and 4).
4. Discussion
The purpose of the present study was to evaluate the test-retest
reliability of a postural control assessment protocol using different
levels of postural difficulty, under single and dual-task conditions
in healthy older adults.
Although several studies have reported ICCs for COP measures
in older population, we found only one study that specifically
estimated reliability for selected COP variables under dual task
Table 4
Test–retest reliability analysis of COP measures for dual-task condition.
Rigid-open
SD amplitude (AP)
SD velocity (AP)
Phase plane (AP)
SD amplitude (ML)
SD of velocity (ML)
Phase plane (ML)
Mean velocity
Area (95% ellipse)
Total phase plane
Rigid-closed
Foam-closed
ICC (95% CI)
SEM
MMDC
CV (%)
ICC (95% CI)
SEM
MMDC
CV (%)
ICC (95% CI)
SEM
MMDC
CV (%)
0.77
0.73
0.77
0.78
0.98
0.98
0.93
0.81
0.90
0.22
0.25
0.20
0.15
0.06
0.08
0.14
2.33
0.16
0.42
0.48
0.38
0.30
0.12
0.16
0.28
4.58
0.31
12.01
5.80
4.67
18.54
1.85
2.29
3.39
19.24
3.54
0.88
0.83
0.86
0.87
0.96
0.95
0.93
0.87
0.91
0.18
0.28
0.27
0.14
0.14
0.17
0.19
2.27
0.25
0.36
0.57
0.54
0.28
0.28
0.33
0.38
4.44
0.49
12.32
4.83
4.69
15.50
3.78
4.23
3.88
25.30
4.14
0.51
0.81
0.82
0.60
0.84
0.81
0.82
0.62
0.82
0.40
1.37
1.36
0.37
0.78
0.83
1.27
9.08
1.54
0.78
2.68
2.67
0.72
1.52
1.64
2.48
17.80
3.02
11.22
13.03
12.14
18.97
14.63
15.54
12.62
24.54
12.19
(0.00,
(0.04,
(0.15,
(0.21,
(0.94,
(0.91,
(0.67,
(0.29,
(0.52,
0.94)
0.93)
0.94)
0.95)
0.99)
0.99)
0.98)
0.95)
0.98)
(0.54,
(0.31,
(0.38,
(0.49,
(0.83,
(0.78,
(0.00,
(0.45,
(0.54,
0.97)
0.95)
0.97)
0.97)
0.99)
0.99)
0.98)
0.97)
0.98)
(0.00,
(0.29,
(0.31,
(0.00,
(0.40,
(0.29,
(0.33,
(0.00,
(0.34,
0.88)
0.95)
0.95)
0.89)
0.96)
0.95)
0.95)
0.91)
0.95)
COP: center of pressure; ICC: intraclass correlation coefficient; SEM: standard error of measurement; MMDC: minimal metrically detectable change; CV: coefficient of
variation; SD: standard deviation; AP: anteroposterior; ML: mediolateral. Units of COP measures are as follows: cm (SD of amplitude); cm/s (SD of velocity/mean velocity);
cm2 (area). Phase plane is in an arbitrary unit.
654
M. Moghadam et al. / Gait & Posture 33 (2011) 651–655
conditions [18–20]. To date, no methodological study could be
found that used a standardized dual-task balance assessment
protocol incorporating distinct levels of postural difficulty. This
limitation in reliability information is surprising given the
increasing number of studies using force plate to assess postural
control under dual-task conditions in older people.
Many studies investigated the association between single and
dual-task force plate measurements and falls (see Refs. [11,23] for a
review). There is some evidence that COP data, especially indicators
of lateral stability may have predictive value for future falls [11–13].
The use of dual-task paradigms also appears promising as a predictor
of injurious falls [12,23]. However, there are studies showing no
association between force platform measurements and falls [11]. In
addition, some previous methodological studies have shown that
reliability of quiet stance COP measurements cannot always be
assumed (see Ref. [28] for a review). We attempted to assess the
reliability of COP measures because one possible explanation for
these contradictory results could be poor reliability of dual-task
testing protocols and parameters used.
Despite the diverse levels of reliability obtained, the results of
the present study indicated high to very high reliability of mean
velocity, total phase plane, phase plane (ML), and SD of velocity
(ML), consistently in all levels of postural difficulty for both single
and dual-task conditions.
With regard to the single task condition, our results are in
accordance with the results of Lin et al. [19] who reported
between-day ICC values of 0.92 and 0.91 for mean velocity (AP and
ML, respectively) and 0.90 for sway area in 16 healthy older adults.
However, their study was only conducted in an eyes-closed
condition [19].
Considering postural balance measurements with and without
cognitive loading, our results are consistent with the findings of
Swanenburg et al. [20]. In their study, COP measurements were
completed in single and dual-task conditions, with and without
vision, and in both faller and non-faller older adults. According to the
results for the non-faller group, test–retest ICC levels of 0.78–0.87 for
mean velocity and 0.62–0.81 for area of 95th percentile ellipse were
found. They did not considered higher levels of postural difficulty or
other COP parameters presented in our study [20].
Even though there are other studies available addressing
reliability of COP measures, it is difficult to compare our results
with them due to the major differences in terms of measurement
protocol, COP measures, study sample, and device used (see Ref.
[28] for a review).
Among the numerous COP measures proposed, parameters with
frequent use in the literature and representing unique dimensions of
COP position, velocity, or both were considered in this study [9,25].
However, some measures like minimum, maximum, and peak-topeak amplitude were not included as they used only one or two data
points among the entire time series in a trial and may result in great
variances between subjects and trials [9]. Interestingly, one rarely
used COP measure, phase plane portrait, showed a high to very high
reliability in the current study. Riley et al. demonstrated that
velocity information alone and in combination with displacement
information (phase plane portrait) discriminated between balanceimpaired and control subjects more effectively than displacement
information [25]. Other studies also revealed that COP mean velocity
was the most discriminating variable for assessment of the agerelated changes of the postural steadiness and risk of falling [10,29].
These findings could be partly explained by the high reliability of the
COP mean velocity and phase plane portrait found in the present
study. In addition, these results suggest further work on exploring
the predictive and evaluative values of the less common but highly
reliable COP parameters.
It has been shown that attentional demands for postural control
increase as sensory information decreases. Therefore, for both
clinical and research purposes, it is of great significance to assess
the ability of older people to maintain postural stability in varied
sensory environments while performing multiple demanding tasks
[30]. We estimated the reliability of a postural control assessment
protocol addressing the interplay between the sensory context and
dual-task balance performance. This point was not taken into
account in most previous reliability studies in which limited test
conditions were included. Our results showed that both relative
and absolute reliability for a given COP parameter varied across
different test conditions.
With regard to the directional differences of reliability indices,
SD of velocity and phase plane portrait showed generally higher
ICC values in ML direction compared with AP direction. Swanenburg et al. reported a similar pattern for the maximal and the RMS
amplitude [20]. However, these results are not in accordance with
the results of Corriveau et al., who found better ICC values in AP
direction than in ML direction for center of pressure-center of mass
(COP-COM) variable [28]. These controversial results could be
explained by the different stance width used. Pelvic width distance
used by Corriveau et al. could result in less variability in ML
direction compared with the 50% hip-to-hip distance used in our
study [20,28]. The higher reliability for the COP measures in ML
direction seems clinically relevant, as ML sway measures have
been previously demonstrated to discriminate well between fallers
and non-fallers, especially under dual-task conditions [12,13].
The pattern found for absolute reliability indices was nearly
similar to what mentioned for the relative reliability. Absolute
reliability was higher for mean velocity, total phase plane, phase
plane (ML), and SD of velocity (ML) measures compared with other
variables in most test conditions, indicating small measurement
errors across repeated measurements. In fact, in comparison with
the other measures, these four parameters have the potential to not
only discriminate better between groups (due to the higher relative
reliability) but also to detect smaller changes in individual’s balance
performance over time (due to the higher absolute reliability)
[26,27]. The estimated MMDC of each COP measure provides
information about the amount of measurement error that should be
taken into consideration when setting the least significant changes
expected following an intervention in clinical trials [27].
The current reliability results could not be generalized to older
adults with a history of falls or disorders affecting postural control.
Given the healthy community-dwelling older adults included in
our study, the power of the presented protocol for early detection
of subtle balance impairments in a similar population can be
further investigated. With regard to our findings, the next step
would be to explore the predictive value of the reliable COP
measures for future falls in the elderly.
5. Conclusion
This study established the relative and absolute reliability of
some COP measures across different levels of postural difficulty,
under both single and dual-task conditions in a group of healthy
older adults. Mean velocity, total phase plane, phase plane (ML),
and SD of velocity (ML) showed high to very high reliability,
consistently in all levels of postural difficulty for both single and
dual-task conditions. In general, the velocity-related COP measures
in ML direction showed better relative and absolute reliability
compared with other COP measures. Further research may use
these results to determine the evaluative and predictive value of
the COP parameters.
Source of support
Partially supported by Tehran University of Medical Sciences,
Tehran, Iran.
M. Moghadam et al. / Gait & Posture 33 (2011) 651–655
Institutional review board
The Institutional Review Board of Tehran University of Medical
Sciences, Tehran, Iran
Acknowledgment
The experiment was conducted in Biomechanics Lab., Rehabilitation Research Center, Tehran University of Medical Sciences.
Conflicts of interest statement
The authors declare no known conflicts of interest.
References
[1] Tinetti ME, Speechley M, Ginter SF. Risk factors for falls among elderly persons
living in the community. N Engl J Med 1988;319:1701–7.
[2] Shumway-Cook A, Baldwin M, Polissar NL, Gruber W. Predicting the probability for falls in community-dwelling older adults. Phys Ther 1997;77:812–9.
[3] Rubenstein LZ. Falls in older people: epidemiology, risk factors and strategies
for prevention. Age Age 2006;35. ii37–41.
[4] Nevitt MC, Cummings SR, Kidd S, Black D. Risk factors for recurrent nonsyncopal falls: a prospective study. J Am Med Assoc 1989;2:2663–8.
[5] Tinetti ME, Leon C, Doucette JT, Baker DI. Fear of falling and fall related efficacy
in relationship to functioning among community-living elders. J Gerontol Med
Sci 1994;49:M140–7.
[6] Slobounov SM, Haibach PS, Newell KM. Aging-related temporal constraints
to stability and instability in postural control. Eur Rev Aging Phys Act
2006;3:55–62.
[7] Blaszczyk JW, Michalski A. Aging and postural stability. Stud Phys Culture Tour
2006;13:11–4.
[8] Baloh RW, Fife TD, Zwerling L, Socotch T, Jacobson K, Bell T, et al. Comparison of
static and dynamic posturography in young and older normal people. J Am
Geriatr Soc 1994;42:405–12.
[9] Palmieri RM, Ingersoll CD, Stone MB, Krause BA. Center-of-pressure parameters used in the assessment of postural control. J Sport Rehabil 2002;11:
51–66.
[10] Era P, Heikkinen E. Postural sway during standing and unexpected disturbance
of balance in random samples of men of different ages. J Gerontol 1985;40:
287–95.
[11] Piirtola M, Era P. Force platform measurements as predictors of falls among
older people – a review. Gerontology 2006;52:1–16.
[12] Bergland A, Wyller TB. Risk factors for serious fall related injury in elderly
women living at home. Inj Prev 2004;10:308–13.
655
[13] Maki BE, Holliday PJ, Topper AK. A prospective study of postural balance and
risk of falling in an ambulatory and independent elderly population. J Gerontol
1994;49:M72–84.
[14] Pajala S, Era P, Koskenvuo M, Kaprio J, TÖrmäkangas TT, Rantanen T. Force
platform balance measures as predictors of indoor and outdoor falls
in community-dwelling women aged 63–76 years. J Gerontol Med Sci
2008;63A(2):171–8.
[15] Crilly RG, Willems DA, Trenholm KJ, Hayes KC, Delaquerrière-Richardson LF.
Effect of exercise on postural sway in the elderly. Gerontology 1989;35:
137–43.
[16] Judge JO, Lindsey C, Underwood M, Winsemius D. Balance improvements in
older women: effects of exercise training. Phys Ther 1993;73:254–62.
[17] Van Wegan EEH, Van Emmerik REA, Riccio GE. Postural orientation: agerelated changes in variability and time-to-boundary. Hum Mov Sci 2002;21:
61–8.
[18] Lafond D, Corriveau H, Hébert R, Prince F. Intrasession reliability of center of
pressure measures of postural steadiness in healthy elderly people. Arch Phys
Med Rehabil 2004;85(6):896–901.
[19] Lin D, Seol H, Nussbaum MA, Madigan ML. Reliability of COP-based postural
sway measures and age-related differences. Gait Posture 2008;28:337–42.
[20] Swanenburg J, Bruin ED, Kathrin Favero K, Uebelhart D, Mulder T. The
reliability of postural balance measures in single and dual tasking in elderly
fallers and non-fallers. BMC Musculoskel Disord 2008;9:162.
[21] Shumway-Cook A, Woollacott M. Motor control therapy and practical
applications, 2nd ed., Philadelphia: LWW; 2001. p. 228–40.
[22] Buatois S, Gueguen R, Gauchard GC, Benetos A, Perrin PP. Posturography and
risk of recurrent falls in healthy non-institutionalized persons aged over 65.
Gerontology 2006;52:345–52.
[23] Zijlstra A, Ufkes T, Skelton DA, Lundin-Olsson L, Zijlstra W. Do dual tasks have
an added value over single tasks for balance assessment in fall prevention
programs? A mini-review. Gerontology 2008;54:40–9.
[24] Woollacott M, Shumway-Cook A. Attention and the control of posture and
gait: a review of an emerging area of research. Gait Posture 2002;16:1–14.
[25] Riley PO, Benda BJ, Gill-Body KM, Krebs DE. Phase plane analysis of stability in
quiet standing. J Rehabil Res Dev 1995;32:227–35.
[26] Atkinson G, Nevill AM. Statistical methods for assessing measurement error
(reliability) in variables relevant to sports medicine. Sports Med 1998;26(4):
217–38.
[27] Domholdt E. Rehabilitation research: principles and applications, 3rd ed.,
Philadelphia: Elsevier Saunders; 2005.
[28] Corriveau H, Hebert R, Prince F, Raiche M. Postural control in the elderly: an
analysis of test–retest and interrater reliability of the COP-COM variable. Arch
Phys Med Rehabil 2001;82:80–5.
[29] Maki BE, Holliday PJ, Fernie GR. Aging and postural control: a comparison
of spontaneous- and induced-sway balance tests. J Am Geriatr Soc 1990;38:
1–9.
[30] Shumway-Cook A, Woollacott M. Attentional demands and postural control:
the effect of sensory context. J Gerontol A Biol Sci Med Sci 2000;55:M10–6.