Full text of DOI:10.1007/s004210100437

Eur J Appl Physiol 42001) 85: 501±506
DOI 10.1007/s004210100437
ORIGINAL ARTICLE
S. Okada á K. Hirakawa á Y. Takada á H. Kinoshita
Relationship between fear of falling and balancing ability
during abrupt deceleration in aged women having similar habitual
physical activities
Accepted: 14 March 2001 / Published online: 6 July 2001
Ó Springer-Verlag 2001
Keywords Fear of falling á Balancing ability á
Abrupt deceleration á Aged women á Co-contraction
index
Abstract The purpose of this study was to evaluate the
in¯uence of fear of falling on balancing ability during
abrupt deceleration in aged women. The subjects were
20 women aged between 67 and 75 years. They were
classi®ed into two groups, one having a fear of falling
4group FF, n=10) and another without this fear 4group
NFF, n=10). The twogroups had similar daily physical
activities. Changes in the centre of foot pressure 4CFP)
were measured during postural sway following hori-
zontal deceleration of the force platform on which they
were standing, and the response time and CFP dis-
placement were evaluated. In addition, the electromyo-
gram 4EMG) onset in the tibialis anterior muscle and
medial gastrocnemius muscle during abrupt decelera-
tion, and its di€erence between the two muscles were
measured, and the relative level of co-contraction of
antagonistic muscles 4the co-contraction index, CCI) in
the lower extremity muscle group was calculated. The
response time and CFP displacement immediately after
abrupt deceleration were signi®cantly higher in group
FF than in NFF 4P<0.05). The EMG onset in the two
muscles did not signi®cantly di€er between the
twogroups. The di€erence in EMG onset between the
two muscles was signi®cantly lower in group FF than in
NFF 4P<0.05). The CCI was signi®cantly higher in
group FF than in NFF 4P<0.05). These results suggest
that there were negative e€ects of a fear of falling on the
balancing ability immediately after abrupt deceleration.
This may be because a fear of falling increases the co-
contraction of antagonist muscles in the lower extremity
muscle group.
Introduction
Falling is one of the most serious problems associated
with aging among the human population 4Maki et al.
1991; Perry 1985). Falling has been found to be a com-
plex, multifactorial problem, and deteriorated postural
control has often been cited as a major contributing
factor 4Barlet et al. 1986; Berg 1989; Lord et al. 1992;
Nevitt et al. 1991). Postural control has been reported to
require a complex interaction of musculoskeletal and
neural systems 4Shumway-Cook and Woollacott 1995).
In addition to this physiological factor contributing
to postural control, Spirduso 41995) suggested a fear of
falling as a psychological factor contributing to the lack
of maintenance of balance. McAuley et al. 41997),
Tinetti et al. 41990), and Walker and Howland 41990)
reported that about 50% of aged people with a fear of
falling have experienced falling. Maki et al. 41994), who
carried out a 1 year follow-up survey, reported that aged
people who had a fear of falling tend to fall more often
than those not having this fear. These studies suggest an
in¯uence of a fear of falling on postural stability in the
standing position.
On the other hand, there has been only one study, by
Maki et al. 41991), on the relationship between a fear of
falling and the ability to maintain balance following a
perturbation of an upright stance. They analysed the
centre of foot pressure 4CFP) during spontaneous-sway
and induced-sway tests and found a signi®cantly poorer
balancing ability in elderly people having a fear of falling
in spontaneous-sway tests but no signi®cant association
between a fear of falling and CFP displacement in in-
duced-sway tests using a movable platform. Based on
these results, they stated that they could not ascertain
whether a fear of falling a€ected balance-test perfor-
mance in an artefactual manner, or whether the fear and
S. Okada 4&) á K. Hirakawa á Y. Takada
Faculty of Human Development, Kobe University,
3±11 Tsurukabuto, Nada-Ku, Kobe 657±8501, Japan
E-mail: shuokada@kobe-u.ac.jp
Tel.: +81-78-8037814
Fax: +81-78-8037929
H. Kinoshita
School of Health and Sports Sciences,
University of Osaka, 1±17 Machikaneyama-cho,
Toyonaka, Osaka 560±0043, Japan

502
activity. Physical activity was assessed using the questionnaire for
investigating the activities of daily living described by Takada et al.
41996), and habitual physical activity 4a physical activity score) was
estimated. Falling was de®ned according to the de®nition of falling
produced by Tinetti et al. 41988), as follows: a stoppage of the body
on the ground or at a lower site, but excluding falling due to en-
dogenous causes or unavoidable accidents. Using the report by
Shumway-Cook et al. 41997) as a reference, the number of falls
during the past 6 months was investigated. As a result, it was found
in group FF that 8 subjects had experienced falling once; in group
the poorer performance were related to a true deterio-
ration in postural control.
Thus, since no consistent conclusion has been
obtained on the in¯uence of a fear of falling on the
balancing ability during perturbed stance, the necessity
for more experimental studies on the relationship
between a fear of falling and postural control has been
recognised 4Shumway-Cook and Woollacott 1995).
Vellas et al. 41987), Shumway-Cook and Woollacott
41995), and Spirduso41995) have suggested that self-
imposed restrictions on habitual physical activity due
to a fear of falling induce a deterioration in balance.
Inactivity has been reported to reduce muscle strength,
joint ¯exibility, and other neuromotor functions, re-
sulting in a deterioration in balance 4Vellas et al. 1987;
Spirduso1995).
This indicates the in¯uence of a fear of falling and
habitual physical activity as psycho-physical factors in-
¯uencing balancing ability. Therefore, the in¯uence of
habitual physical activity should be taken into account
when the possible in¯uences of a fear of falling on bal-
ancing ability during a perturbation of stance are being
evaluated. However, to our knowledge, there have been
nostudies that have evaluated the relationship between
a fear of falling and the balancing ability during a per-
turbed stance after eliminating the in¯uence of habitual
physical activity.
NFF, 3 subjects had alsoexperienced falling once. A
analysis showed that the two groups did not have signi®cantly
v-square
di€erent distributions of falling.
Table 1 shows the mean values and standard deviations of age,
body height and mass, foot length and physical activity score in the
two groups which showed no signi®cant di€erences. Interviews
with the subjects concerning their medical histories indicated that
none of them had disorders of the neuro-muscular system or were
receiving therapy with hypotensive drugs. After explanation of the
purpose and safety of this study, consent was obtained from all
subjects.
Apparatus
The experiment setup consisted of a platform for standing on, an
electromyogram 4EMG) recorder, and a personal computer. The
platform for standing on was built using a force plate 4Patela Inc.,
model K40, Tokyo, Japan) which was supported in two sliding rails
®xed to an iron plate. The force plate could be moved horizontally
using a solenoid 4Meikosha Inc., model S12, Tokyo, Japan) to
introduce an abrupt accelerated postural perturbation in a poste-
rior direction. The plate was stopped after moving a predetermined
distance by a bu€er togenerate a sudden deceleration. In the
present study, the platform moved at 182 4‹ 8) mmás±1 in an an-
terior direction. An anterior acceleration of the plate 4mean, about
0.6 g; 70 ms) was initially applied, followed by a posterior decel-
eration 4about ±5.1 g, 10 ms).The distance moved was set to be
15 mm, which was the maximal distance that would not cause falls
in the present elderly subjects.
Surface EMG was obtained from the tibialis anterior muscle
and the medial head of the gastrocnemius of the right leg. Bipolar
electrodes 410 mm diameter silver discs) were placed on the skin
around the probable motor point of the muscles 4Zipp 1982) at a
longitudinal separation of 20 mm. The EMG signals were ampli®ed
using bio-ampli®ers 4Nippon Denki Sanei Inc, model 511, Tokyo,
Japan) and stored in a personal computer via an A-D converter
sampling at 1,000 Hz. The time constant of the ampli®er was set at
0.03 s. Care was taken that the inter-electrode resistance was 10 kW
or less 4Fig. 1).
To clarify the in¯uence of a fear of falling on bal-
ancing ability during abrupt deceleration, we have
compared CFP displacement after abrupt deceleration
in aged women having similar physical activities, be-
tween those having a fear of falling and those not having
this fear. In addition, muscle activity in the lower ex-
tremity muscles during abrupt deceleration, and motor
ability, have been compared between the two groups,
and the relationship between a fear of falling and bal-
ancing ability has been discussed.
Methods
Subjects
Procedure
Of 132 healthy women aged 65 years or over, 46 with a marked or
some fear of falling 4fear of falling group, group FF) and 53 with
no such fear 4no fear of falling group, group NFF) were selected
4total, 99 women) using the method of Maki et al. 41991). Next,
10 women were selected from each of the two groups so that the
two were matched for age, body height and mass, and physical
The subjects with their feet bare placed the posterior surfaces of the
two heels 0.2 m in front of the posterior edge of the force plate.
They assumed a standing position at rest with the medial surfaces
of their feet apart and parallel 4distance, 0.1 m) and with both
hands loosely attached to the sides of their bodies. The CFP in their
maximal forward and backward leaning positions were then mea-
sured for each subject. During the experiment, the subjects were
instructed to keep their CFP at the mean location of these two
maximal CFP. The mean CFP in the range of motion of CFP in the
anteroposterior direction was 0.103 4SD 0.01) m forwards from the
heel in the subjects. The subjects then looked at a target placed 2 m
before them at eye level, and the postural disturbances were then
applied in this state. To minimize the subjects' anticipation of
platform movement time, the delay between an auditory cue or
ready call to the platform movement was altered within 1 to 3 s in a
random manner. In our pilot study, no training e€ects were ob-
served in this experimental task until the third measurement, which
agreed with the ®ndings of some previous studies on postural
control 4Alexander et al. 1992; Studenski et al. 1991). The mean
Table 1 Characteristics of the subjects
Fear of falling
No fear of falling
Mean
SD
Mean
SD
Age 4years)
Height 4m)
Body mass 4kg)
Foot length 4m)
Physical activity score
70.9
1.535
55.4
0.229
11.2
2.4
0.027
2.8
0.006
1.8
69.5
1.540
56.7
0.230
12.2
2.6
0.029
2.9
0.005
1.7

503
Fig. 1 A diagram of the experimental apparatus for providing
an unexpected perturbation by an abrupt deceleration. EMG
Electromyogram
value of three measurements was used as the representative value
for each condition in each subject. The ampli®ed signals from ac-
celerometers on the platform and the EMG were A-D converted at
sampling rate of 1,000 Hz. All of these data were stored in a per-
sonal computer for o€-line analysis.
Fig. 2 Analysis of response time variables 4T1, T2, T3) and CFP
displacement variables 4D1, D2, D3, D4) for centre of foot pressure
data, and electromyogram 4EMG) onset time in lower extremity
muscles. Tibia. tibialis anterior, Gastro. gastrocnemius, A anterior,
P posterior
Analysis
Data between 2 s before and 3 s after the onset of movement of the
plate were subjected to the analysis of postural control.
As shown in Fig. 2, response time variables 4T1, T2, T3) and
displacement variables 4D1, D2, D3, and D4) were measured based
on anteroposterior displacements of CFP. The time from the ini-
tiation of plate movement to the D1 point was similar among the
subjects [122.9 4SD 6.8) ms].
For EMG data from the tibialis anterior and gastrocnemius
muscles, the time from the initiation of plate movement to the ®rst
appearance of muscle activity 4muscle activity latent time: EMG
onset) was measured. The appearance of the EMG activity was
determined when the signals exceeded the 2 SD above or below the
mean of ®rst 50 ms data in the 1-s period before plate movement
4Stelmach et al. 1990). This EMG onset was determined ®rstly
using a computer algorithm and secondly checked using an inter-
active graphics procedure. In addition, the di€erence in EMG onset
between the tibialis anterior and gastrocnemius muscles was
obtained.
To evaluate the relative level of co-contraction of the tibialis
anterior and gastrocnemius muscles observed immediately after
abrupt deceleration, the co-contraction index 4CCI) was calculated
using the method of Falconer and Winter 41985) and Maki and
Ostrovski 41993). The CCI was calculated over the 200 ms interval
beginning at the onset of activity. How the CCI was calculated
from the scaled EMG signals is illustrated in Fig. 3.
Fig. 3 De®nition of the co-contraction index 4CCI). Note, electr-
omyogram 4EMG) responses are scaled with respect to each other
using results from isometric co-contraction trials before calculating
the CCI. Plantar¯exor EMG thin line, dorsi¯exor EMG wavy line,
antagonist EMG thick line, CCI=42Aantag)/4Adf+Apf), where
Aantag, Adf and Apf are the activities of antagonist, dorsi¯exor and
plantar ¯exor muscles, respectively
The mean value in three measurements was calculated and used
as the representative.
Measurement of motor ability
The following items were measured: dorsal ¯exion and plantar The range of ankle and hip joint motions was measured using
¯exion strengths, knee extension and knee ¯exion strengths, range a Todai type joint angle measuring apparatus 4Yagami Co., model
of joint motion of dorsal ¯exion, plantar ¯exion, hip extension, and S-30, Tokyo, Japan). The CFP ¯uctuations were evaluated in terms
hip ¯exion, one leg balancing time with eyes open and closed, and of CFP displacement during a 20-s period with the eyes open and
CFP ¯uctuations in the standing position at rest. In addition, sit-to- then closed using a Gravicorder 4Anima Co., model GS-10, Tokyo,
stand and walking tests were performed. Dorsal ¯exion and plantar Japan). Sit-to-stand tests were performed by the method of Mac-
¯exion strengths were measured using a muscle force measurement Rae et al. 41992), and the times required for standing-up from a
device for the lower extremities 4Wami Inc., model LS-10, Tokyo, chair for each of ®ve times were measured. Walking tests were
Japan). As knee extension and ¯exion strengths, the torques at performed using the method of Himann et al. 41988) and Wolfson
angular velocities of 1.05 and 3.14 radás±1 were measured using a et al. 41990), and the walking speed and stride length during
Cybex dynamometer 4Lumex Inc., CYBEX II, New York, USA). walking as fast as possible were measured or calculated.

504
Statistics
Table 3 Electromyogram onset of tibialis anterior and gastroc-
nemius muscles
One-way analysis of variance 4ANOVA) was used to determine the
di€erence in the mean values of each parameter between the FF
and NFF. A P<0.05 was considered signi®cant.
Muscle
Fear of falling
No fear of falling
Mean
SD
Mean
SD
Tibialis anterior 4ms) 139.5
Gastrocnemius 4ms) 147.7
22.3
27.6
135.1
152.2
21.0
25.1
Results
Response time and CFP displacement
in groups FF and NFF
Motor ability in groups FF and NFF
The response time variables 4T1, T2, T3) and CFP dis-
placement variables 4D1, D2, D3, D4) in groups FF and
NFF are shown in Table 2. The response time variables
and CFP displacement variables were higher in group FF
than in group NFF. Signi®cant di€erences were observed
in T1 and D2 between the twogroups 4 P<0.05).
Table 4 shows dorsal ¯exion and plantar ¯exion
strengths, knee extension and knee ¯exion strengths,
range of joint motion of dorsal ¯exion, plantar ¯exion,
hip extension, and hip ¯exion, one leg balancing time
with eyes open and then closed, CFP displacement with
the eyes open and then closed, sit-to-stand time, and
walking speed and stride length in groups FF and NFF.
Group FF showed a signi®cantly shorter one leg bal-
ancing time with the eyes closed 4P<0.05) and a sig-
ni®cantly longer CFP displacement with the eyes closed
4P<0.05) than group NFF. Excluding these items, no
other signi®cant di€erences were observed between the
twogroups.
EMG onset in the tibialis anterior and gastrocnemius
muscles and the di€erence in EMG onset
between the twomuscles in groups FF and NFF
The EMG onsets in the tibialis anterior and gastroc-
nemius muscles in groups FF and NFF are shown in
Table 3. Nosigni®cant di€erence was observed in the
EMG onset of either muscle between the two groups.
The di€erence in EMG onset between the tibialis
anterior and gastrocnemius muscles was 8.2 4SD 5.6) ms
in group FF and 17.1 4SD 8.4) ms in group NFF, being
signi®cantly lower in group FF 4P<0.05). This suggests
that the ®rst appearance of EMG activity of the gas-
trocnemius muscle after the ®rst appearance of EMG
activity of the tibialis anterior muscle occurred signi®-
cantly earlier in group FF than in group NFF.
Discussion
There were four response phases of CFP displacement
during balance recovery after abrupt deceleration in this
study 4Fig. 2). The ®rst response was a posterior CFP
displacement due to the deceleration resulting from the
moving plate hitting the bu€er 4passive phase: D1).
Next, a response to this displacement occurred. In trying
to recover CFP in orientation, but due to overshooting,
the balance was destroyed, resulting in an anterior dis-
placement 4®rst active phase: T1, D2). A further recov-
CCI for the tibialis anterior and gastrocnemius
muscles in groups FF and NFF
Table 4 Motor ability assessments of the fear of falling and no fear
of falling groups. OLBT One leg balancing time, CFP centre of foot
pressure, CFPdisp CFP displacement
The CCI for the antagonistic muscles was
0.56 4SD 0.114) in group FF and 0.45 4SD 0.098) in
group NFF, being signi®cantly higher in group FF
4P<0.05).
Variable
Fear of falling
No fear of falling
Mean
SD
Mean
SD
Dorsal ¯exion 4N)
Plantar ¯exion 4N)
Knee extention 4Nám)
Knee ¯exion 4Nám)
Dorsal ¯exion 4°)
110.9
200.1
26.7
18.8
14.2
52.4
72.8
25.1
30.7
3.4
23.2
61.7
7.4
5.6
4.0
7.5
8.9
3.6
l8.8
7.7
8.1
10.3
2.8
120.5
230.0
30.1
21.1
15.7
55.5
75.2
27.0
38.2
11.3*
24.3
33.8*
7.0
213
62.2
Table 2 Response time variables 4T1, T2, T3) and centre of foot
pressure 4CFP) displacement variables 4D1, D2, D3, D4) of the fear
of falling and no fear of falling groups
7.1
5.9
3.9
7.4
8.8
3.4
19.1
8.2
7.8
9.1
2.2
Fear of falling
No fear of falling
Plantar ¯exion 4°)
Hip ¯exion 4°)
Mean
SD
Mean
SD
Hip extension 4°)
OLBT: eyes open 4s)
OLBT: eyes closed 4s)
CFPdisp: eyes open 4cm)
CFPdisp: eyes closed 4cm) 43.7
Sit-to-stand: 5 times 4s)
Walking speeda 4más±1
Stride lengtha 4m)
T1 4ms)
T2 4ms)
T3 4ms)
D1 4%)a
D2 4%)a
D3 4%)a
D4 4%)a
541.4
641.1
538.8
22.1
34.2
27.7
92.2
105.3
111.6
4.8
7.5
7.7
439.9*
607.2
520.5
20.6
26.9*
23.2
8.1
915
99.9
25.0
96.1
4.3
6.0
5.9
2.8
7.7
)
1.82
0. 63
0.52
0.11
1.99
0.69
0.33
0.09
10.7
4.2
aWalking as fast as possible
*P<0.05
aCPF displacements were normalized by dividing by foot length
*P<0.05

505
ery response to this displacement was observed, but with a fear of falling adopt a sti€ened muscle strategy
overshooting occurred, and CFP was displaced posteri- compared with those without this fear.
orly once again 4second active phase: T2, D3). Subse-
From the results of motor ability, no signi®cant dif-
quently, a response to recover CFP in orientation ferences were observed in the muscle strength and ¯ex-
occurred 4third active phase: T3, D4), gradually recov- ibility of the legs and hips or the static balancing ability
and mobility with the eyes open between groups FF and
ering the orientation of CFP.
Compared with group NFF, group FF showed a NFF. However, the response time and displacement
delayed response 4T1) and marked CFP displacement after abrupt deceleration di€ered between the two
4D2) in the ®rst active phase. These results suggest that a groups, suggesting the in¯uence of intrinsic factors other
fear of falling perhaps caused by the experience of a than motor ability on the balancing ability during
previous fall a€ects the recovery response immediately abrupt deceleration. Maki et al. 41991) evaluated the
after abrupt deceleration but not the recovery responses possible relationship between a fear of falling and CFP
in the second active and subsequent phases.
displacement in the standing position at rest and
Concerning the mechanism of the in¯uence of a fear observed a signi®cant relationship between the two
of, or anxiety about, falling on balancing ability, Alex- parameters when blindfolded but not with the eyes open.
ander 41994) and Maki et al. 41991) have proposed a They suggested that this di€erence between the blind-
hypothesis that aged people with these fears tend to folded and eyes-open states was due to a deterioration in
sti€en their bodies when stance is perturbed. It has been proprioceptive or vestibular function in the fearful
suggested that this sti€ening can be induced by co-con- subjects. Based on this speculation, the signi®cantly
traction of antagonist muscles 4Kearney and Hunter poorer results in terms of the displacement of CFP with
1990). In this study, therefore, the CCI of the lower the eyes closed in group FF than in group NFF in this
extremity muscles was calculated. Aged women with a study may also indicate deterioration in proprioceptive
fear of falling indeed had a higher CCI value, indicating or vestibular function in group FF. In people with a fear
a higher level of co-contraction in their lower extremity of falling, deprivation of vision has been shown to en-
muscles than those lacking this fear.
hance the in¯uence of anxiety 4Maki et al. 1991). Anx-
Maki et al. 41991) suggested the following mechanism iety has been found to induce changes in c-motoneuron
for co-contraction. Aged people with fear of a pertur- activity 4Ribot et al. 1986) and reduce attention 4Stel-
bation of their upright stance tend to adopt a sti€ened mach et al. 1990), which may have caused the signi®cant
strategy to minimize the risk of falling, and such com- di€erences in the one leg balancing time and CFP dis-
pensation is achieved by sti€ening of the postural con- placement with the eyes closed between the two groups.
trol system by co-contraction of antagonist muscles.
In conclusion, a fear of falling has negative e€ects on
During the horizontal anterior-posterior movement of the balancing ability immediately after an abrupt de-
the moving platform, sti€ening of the ankle due to co- celeration. This may be associated with a more marked
contraction of the lower extremity muscles e€ectively co-contraction of antagonist muscles in the lower ex-
acts on postural stability. However, during the abrupt tremity muscle groups in subjects with a fear of falling
decelerations due to the rapid movements and sudden than those without this fear.
stops of the plate in this study, an increase in the sti€ness
of the ankle may have resulted in direct transmission of
the impact due to the sudden stop to the body centre of
mass, increasing displacement. Considering the EMG
References
onset time in the tibialis anterior muscle that was ®rst
activated, the in¯uence of the co-contraction appeared
in the ®rst active phase. This may have resulted in the
higher values in T1 and D2 in group FF with a higher
co-contraction level than in group NFF.
Alexander NB 41994) Postural control in older adults. J Am Ge-
riatr Soc 42:93±108
Alexander NB, Shepard N, Gu MJ, Schltz A 41992) Postural
control in young and elderly adults when stance is perturbed:
kinematics. J Gerontol Med Sci 47:M79±M87
Barlet SA, Maki BE, Fernie GR, Holliday PJ 41986) On the clas-
si®cation of a geriatric subject as a faller or non-faller. Med Biol
Eng Comput 24:219±222
Berg K 41989) Balance and its measure in the elderly: a review.
Physiother Can 41:240±246
Brooks VB 41986) The neural basis of motor control. Oxford
University Press, Oxford
Falconer K, Winter DA 41985) Quantitative assessment of co-
contraction at the ankle joint in walking. Electromyogr Clin
Neurophysiol 25:135±149
Himann JE, Cunningham DA, Rechnitzer PA, Paterson DH 41988)
Age-related changes in speed of walking. Med Sci Sports Exerc
20:161±166
Kearney RE, Hunter IW 41990) System identi®cation of human
joint dynamics. Crit Rev Biomed Eng 18:55±87
Lord SR, McLean D, Stathers G 41992) Physiological factors as-
sociated with injurious falls in older people living in the com-
munity. Gerontology 38:338±346
Satoand Fujita 41998) have evaluated ankle joint
viscoelasticity during changes in posture and vision and
suggested a control of ankle joint viscoelasticity by a
higher postural control system. Brooks 41986) described
how muscle sti€ness is controlled by both re¯exes and
upper levels of the central nervous system. These studies
indicate that co-contraction of the lower extremity
muscles is programmed from higher level neural systems.
In this study there was a signi®cantly earlier appearance
of the ®rst EMG activity of the gastrocnemius in group
FF than in group NFF. Shumway-Cook and Woollacott
41995) reported modi®cation of the postural control
strategy based on cognition of abrupt acceleration in the
aged. These ®ndings seem toindicate that aged people

506
MacRae PG, Lacours M, Moldavon R 41992) Physical perfor- Spirduso WW 41995) Physical dimensions of aging. Human Ki-
mance measures that predict faller status in community-dwell-
ing older adults. J Orthop Sports Phys Ther 16:123±128
netics, Champaign
Stelmach GE, Zelaznik HN, Lowe D 41990) The in¯uence of aging
and attentional demands on recovery from postural instability.
Aging 2:155±161
Maki BE, Ostrovski G 41993) Scaling of postural responses to
transient and continuous perturbations. Gait & Posture 1:93±104
Maki BE, Holliday PJ, Topper AK 41991) Fear of falling and Studenski S, Duncan MA, Chandler J 41991) Postural responses
postural performance in the elderly. J Gerontol Med Sci and e€ectors in persons with unexplained falls: results and
46:M123±131 methodological issues. J Am Geriatr Soc 39:229±234
Maki BE, Holliday PJ, Topper AK 41994) A prospective study of Takada Y, Okada S, Oda Y, Hirakawa K 41996) The relationship
postural balance and risk of falling in an ambulatory and in-
dependent elderly population. J Gerontol Med Sci 49:M72±84
McAuley E, MihaikoSL, Rosengren K 41997) Self-ecacy and
balance correlates of fear of falling in the elderly. J Aging Phys
Activity 5:329±340
between the actual physical ®tness level of elderly people and
their self-reported habitual physical activity scores as investi-
gated by questionnaire method 4in Japanese with English ab-
stract). J Phys Educ Sports Sci 5:41±47
Tinetti ME, Speechley M, Ginter S 41988) Risk factors for falls
among elderly persons living in the community. N Eng J Med
319:1701±1707
Nevitt MC, Cummings SR, Hudes ES 41991) Risk factors for inju-
rious falls: a prospective study. J Gerontol Med Sci 46:M164±170
Perry B 41985) Falls among the elderly: a review of the methods Tinetti ME, Richman D, Powell L 41990) Falls ecacy as a
and conclusions of epidemiologic studies. J Am Geriatr Soc measure of fear of falling. J Gerontol Psychol Sci 45:P239±
30:367±371 243
Ribot E, Roll JP, Vedel JP 41986) E€erent discharges recorded Vellas B, Cayla F, Bocquet H, Pemille F de, Albarede JL 41987)
from single skeletalmotor and fusimotor ®bres in man. J Physiol Prospective study of restriction of activity in old people after
4Lond) 375:251±268 falls. Age Ageing 16:189±193
Sato H, Fujita K 41998) Analysis of ankle joint viscoelasticity in Walker JE, Howland J 41990) Falls and fear of falling among el-
human upright posture 4in Japanese with English abstract). derly persons living in the community: occupational therapy
Biomechanisms 14:49±58 interventions. Am J Occup Ther 45:119±122
Shumway-Cook AS, Woollacott MH 41995) Motor control ± theory Wolfson L, Whipple R, Amerman P, Tobin JN 41990) Gait as-
and practical applications-. Williams and Wilkins, Baltimore sessment in the elderly: a gait abnormality rating scale and its
Shumway-Cook AS, Woollacott MH, Kerns KA, Baldwin M relation to falls. J Gerontol Med Sci 45:M12±19
41997) The e€ects of two types of cognitive tasks on postural Zipp P 41982) Recommendations for the standardization of lead
stability in older adults with and without a history of falls.
J Gerontol Med Sci 52:M232±242
positions in surface electromyography. Eur J Appl Physiol
50:41±54