Archives of Gerontology and Geriatrics 59 (2014) 506–514

Contents lists available at ScienceDirect

Archives of Gerontology and Geriatrics

journal homepage: www.elsevier.com/locate/archger

Effect of different types of exercise on postural balance in elderly

women: A randomized controlled trial
Marcio R. de Oliveira a,b, Rubens A. da Silva a,b,*, Juliana B. Dascal c, Denilson C. Teixeira a,c
a
Health Science Research Center, Laboratory of Functional Evaluation and Human Motor Performance (LAFUP), University of Northern Parana (UNOPAR),
675 Paris Ave., Londrina 86041-120, PR, Brazil
b
Master and Doctoral Program in Rehabilitation Sciences UEL/UNOPAR, 675 Paris Ave., Londrina 86041-120, PR, Brazil
c
Physical Education Department, Universidade Estadual de Londrina (UEL), Rodovia Celso Garcia Cid, Pr 445 km 380, Campus Universitário,
Londrina 86051-980, PR, Brazil

A R T I C L E I N F O A B S T R A C T

Article history: Different types of exercise are indicated for the elderly to prevent functional capacity limitations due to
Received 3 April 2014 aging and reduce the risk of falls. This study aimed to evaluate the effect of three different exercises
Received in revised form 13 August 2014 (mini-trampoline, MT; aquatic gymnastics, AG and general floor gymnastics, GG) on postural balance in
Accepted 18 August 2014
elderly women. Seventy-four physically independent elderly women, mean age 69  4 years, were
Available online 24 August 2014
randomly assigned to three intervention groups: (1) MT (n = 23), (2) AG (n = 28), and (3) GG (n = 23). Each
group performed physical training, including cardiorespiratory, muscular strength and endurance, flexibility
Keywords:
and sensory-motor exercises for 12 weeks. To determine the effects on each intervention group, five postural
Aging
Rehabilitation
balance tasks were performed on a force platform (BIOMEC 400): the two-legged stand with eyes open
Exercise (TLEO) and two-legged stand with eyes closed (TLEC); the semi-tandem stand with eyes open (STEO) and
Postural balance semi-tandem stand with eyes closed (STEC) and the one-legged stand. Three trials were performed for each
task (with 30 s of rest between them) and the mean was used to compute balance parameters such as center
of pressure (COP) sway movements. All modalities investigated such as the MT, AG and GG were significantly
(P < 0.05) efficient in improving the postural balance of elderly women after 12 weeks of training. These
results provide further evidence concerning exercise and balance for promoting health in elderly women.
ß 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction (physical inactivity) that can directly affect the neuromuscular

Aging is a natural phenomenon that touches all people during
their life. This phenomenon is characterized by biological changes
such as muscle mass and strength reduction, loss of mobility and
balance and motor coordination impairment (Granacher, Muehl-
bauer, Gollhofer, Kressig, & Zahner, 2011). All these changes
contribute to an increased risk of falls among the elderly (Horak,
2006). In fact, poor postural balance can be a predictor of fall
occurrences in elderly subjects (Lord, Sherrington, & Menz, 2001).
With increased age, the mechanisms of postural control become
impaired from these changes and also due to disuse syndrome
* Corresponding author at: Centre for Research in Health Sciences (LAFUP),
Universidade Norte do Paraná (UNOPAR), Av. Paris, 675, Jd. Piza, CEP 86041-140, Cx.
P. 401, Londrina, PR, Brazil. Tel.: +55 43 3371 7700x7990; fax: +55 43 3371 7721;
mobile: +55 43 9993 0477.
E-mail addresses: marxroge@hotmail.com (M.R. de Oliveira),
rubens@kroton.com.br, rubensalex@hotmail.com (R.A. da Silva),
jbdascal@yahoo.com.br (J.B. Dascal), denict.9@gmail.com (D.C. Teixeira).
system in maintaining balance.
An efficient strategy to improve postural balance and reduce
the biological changes of aging and/or physical inactivity is to
practice regular physical exercise (ACSM, 2009). Evidence supports
the beneficial effects of physical exercise for improving postural
balance and reducing the risk of falls in elderly people (Alfieri et al.,
2012; Avelar, Bastone, Alcântara, & Gomes, 2010; Mian et. al.,
2007; Resende et al., 2008).
Seco et al. (2013), who evaluated 227 elderly subjects, reported
a significant improvement of balance in force platform measure-
ments after 36 weeks (3 per week) of an exercise program related
just to general gymnastics. Other studies (Aragão, Karamanidis,
Vaz, & Arampatzis, 2011; Avelar et al., 2010), which used
sensorimotor exercises and AG, also found similar positive results
in balance on force platform measurements, even with a lower
duration and frequency of training (2 per week for 14 weeks and
2 per week for 6 weeks, respectively).
On the other hand, some results in the literature were obtained
using only indirect measures of balance, such as balance scales and

http://dx.doi.org/10.1016/j.archger.2014.08.009
0167-4943/ß 2014 Elsevier Ireland Ltd. All rights reserved.
 M.R. de Oliveira et al. / Archives of Gerontology and Geriatrics 59 (2014) 506–514
functional tests (Jehu, 2012). With regard to balance conditions,
Aragão et al. (2011) employed a two-legged stand task in their study,
while Alfieri et al., (2012) used a one-legged stand, which would
better challenge the balance. Jehu (2012) performed functional tests
and did not report any improvement in balance after the
intervention (at 12 weeks). Although the results from these studies
are relevant, it would also be interesting to apply the tests with a
main balance parameter related to the COP, which is a reliable and
valid measurement, and further in other balance tasks including: the
two-legged stand; the semi-tandem stand with eyes both open and
closed, and finally in a one-legged stand, which is a last predictive
task for falls. Also, it would be relevant to determine whether a
specific type of intervention, such as a MT or aquatic exercise, which
apparently promotes more sensory-motor stimulus from the
specific properties of their action, can promote better results in
balance than exercises such as GG. Some studies (Alfieri et al., 2012;
Avelar et al., 2010) have not compared more than two types of
exercises within the same experimental design and also never
compared the MT with other modes of exercise, which in turn limits
the conclusions and consequently warrant a new investigation on
health promotion for the elderly.
Finally, it would be interesting to investigate the effect of these
exercises in women, since the female sex is more susceptible to
neurophysiological changes with aging and to chronic degenera-
tive diseases such as osteoporosis, which in turn may impair the
postural balance and increase the risks of falls (Burke et al., 2010).
Women represent also a greater percentage of the population in
the world and show greater adherence to exercise programs
compared to men (Burke et al., 2010; Rydwik et al., 2013).
The main purpose of this study was to determine the effect of
three different types of exercise (MT, AG and GG) on five postural
balance tasks from force platform measurements in elderly women.
2. Materials and methods
2.1. Design and subjects
This study was a controlled and randomized clinical trial
registered with the ReBEC: Brazilian Registration of Clinical Trials –
http://www.ensaiosclinicos.gov.br: Rio de Janeiro (RJ): Institute of
Communication and Scientific and Technological Information in
Health (Brazil); 2010 – Identifier RBR-7ptrb5; and conducted from
April to September 2012.
Fig. 1. Design and flow of participants in the study. Three groups were allocated randomly: mini trampoline (MT), aquatic gymnastic (GH), and general floor gymnastic (GG).
Balance task condition related to STEO: semi-tandem with eyes open; STEC: semi-tandem with eyes closed; ONE: one-legged stand.
507
The inclusion criteria were: women over 60 years old, physically
independent in accordance with Spirduso (1995), a good mental
state using the mini-mental questionnaire (score > 18, Hughes,
Duncan, Rose, Chandler, & Studenski, 1996), not performing physical
activity for 2 months, absence of any mental or physical illness that
could interfere in assessment tests; surgeries in the locomotor
system, severe orthopedic, neurological or respiratory dysfunction,
visual and/or vestibular changes. The exclusion criterion was the
inability to perform the tests proposed.
From all criteria, 96 elderly women residents of the community
were included in the present study. All participants were recruited
by advertisement, such as in newspapers, flyers and personal
connections. However, the final sample of the study included only
74 elderly, as better detailed in Fig. 1. All participants were
informed about the study and agreed to participate voluntarily and
signed an informed consent. The research project was approved by
the local Ethics Committee (CEP/UEL:036/2012).
The sample size of the present study was defined by calculation
of the mean and standard deviation of the results reported by
Pluchino, Lee, Asfour, Roos, and Signorile (2012). The significance
level was 0.05 with the power of the sample estimated at 80%. The
estimated sample size, using the mean baseline value of the COP
balance parameter from the trained group (0.39  0.23 cm)
compared to control (0.61  0.34 cm), was 19 individuals in each
intervention group. However, 23 participants were recruited to
compensate for losses. The principle of intention to treat was used
with possible dropouts, and the missing data were included with the
mean of the remaining data. This strategy does not change the mean
of the sample and prevent statistical power.
First, all participants answered a structured interview on
personal information and health condition. The questionnaires on
mental state (mini-mental) and fear of falling (falls) were also
applied, as well as collecting anthropometric measurements.
Afterward, the participants were allocated randomly to three
groups by a numerical sequence with a sealed paper placed in a box
and carried out by an investigator who was not involved with the
intervention. On the first day of the assessment, the paper
allocated to each participant was opened by the investigator
(professionally trained in physical education) who provided the
intervention. All physical education professionals engaged in the
study (with the assistance of physical education students) and
responsible for each intervention were not involved in balance data
508 M.R. de Oliveira et al. / Archives of Gerontology and Geriatrics 59 (2014) 506–514
Fig. 2. Design of intervention from three exercise modalities. The exercise program for three groups (MT, AG and GG) was performed at 2 week, in alternate days, with
approximately 60 min of exercise in each session, totaling 24 sessions. This exercise program was progressive by the volume training. The 24 sessions were portioned in four
cycles of six sessions, which for each cycle the volume of training related to number of repetitions was increased in 10% of previous cycle.
collection. Also, a blinded assessor (physiotherapist) assessed the
balance outcomes on the first day and the last day of intervention (12
weeks after randomization) in a specific laboratory for postural
control assessment (LAFUP-UNOPAR, Londrina). Due to the nature of
the interventions it was not possible to blind the subjects with
regard to the three types of exercise in the study. Fig. 1 shows the
design and flow of the participants through the study: (1) MT, (2) AG
and (3) GG.
2.2. Intervention
The exercise program for the three groups (MT, AG and GG) was
performed twice per week, on alternate days, with approximately
60 min of exercise in each session, totaling 24 sessions (Fig. 2). This
exercise program was progressive by volume training and not
intensity by load because only inexpensive and alternate materials
such as elastic bands, dumbbells and ankle weights of 1 or 2 kg
maximum, balls and floats were used. The 24 sessions were
portioned into four cycles of six sessions (Fig. 2), and for each cycle
the volume of training related to the number of repetitions was
increased by 10% from the previous cycle.
For the three interventions the physical and neuromotor
abilities were performed as follows: (1) neuromotor ability from
balance exercises, body agility, reaction time, motor coordination
and rhythm (approximately 15 min); (2) muscular strength and
endurance from the lower limb (quadriceps, ischium, gluteus,
adductor, abductor and triceps surae), upper limb (pectoral,
latissimus dorsi, trapezius, biceps and triceps brachii) and trunk
(lumbar and abdominal) exercises for 25–30 min; (3) flexibility for
lower and upper limb segments with a static 20 s in a position of
stretching (15–20 min); and (4) aerobic exercises that were
performed more in aquatic and floor gymnastics. Exercises such
as squats, unilateral knee flexion, extension, adduction, unilateral
hip abduction, plantar flexion while standing, adduction and
extension of the arms, flexion and extension of the elbow were all
performed in accord with each modality.
2.2.1. Mini-trampoline group
For this group, the exercises were basically to improve
neuromotor abilities in static and dynamic balance conditions
with visual stimuli, the reaction time from visual, auditory and
tactile stimuli, body agility, motor coordination and rhythm with
choreographed songs. The exercises were varied to challenge the
participants on the MT.
As cited below, exercises for strength, endurance and
flexibility of the segments targeted were also performed on
the MT. All exercises were adapted and developed with support
bars to be safer on the MT when the elderly were performing. The
unstable surface of the MT provided proprioceptive stimuli
during exercise.
2.2.2. Aquatic gymnastic group
As pointed out for the MT group, the same sequence of exercises
was performed in AG. In this exercise modality, the aerobic exercises
were more stimulating. Neuromotor exercises were adapted to the
aquatic environment, where water properties have the advantage of
potentially stimulating neuromuscular actions in the elderly. The AG
were performed indoors with water temperature varying from 30 to
34 8C. The dimensions of the pool were 15 m  6 m, with 1.20 m of
depth. In order to increase the intensity of exercise in the water,
specific equipment for aquatic aerobic exercises were used, such as
dumbbells, ankle weights, pool noodles and floats, altering the speed
of action in each progressive cycle.
2.2.3. General floor gymnastics group
The sequence exercises in the general gymnastics were
performed similar to the MT and AG groups. Again, neuromotor
exercises were performed followed by muscular strength and
endurance, and flexibility. The main difference between the groups
was that in this case the exercises were all performed on the floor
using inexpensive materials such as mats, sticks, balls, elastic
bands, dumbbells and ankle weights.
2.3. Postural balance assessment
Postural balance was assessed before and after intervention by
a trained physiotherapist. The postural balance assessment was
performed on a force platform (BIOMEC 400, EMG system do Brasil,
SP, Ltda). Five task conditions were performed presented
randomly: (1) two-legged stand with eyes open (TLEO); (2) two-
legged stand with eyes closed (TLEC); (3) semi-tandem (heel of one
foot alongside the hallux of the other foot) with eyes open (STEO);
(4) semi-tandem with eyes closed (STEC) and (5) one-legged stand
on the preferred leg indicated by each participant (ONE).
All participants were familiarized with the equipment and
protocol until they were comfortable with the testing. Balance
assessment was performed with a standardized protocol: barefoot
with their arms at their sides or parallel to their trunk. During
testing with eyes open, the participant would look at a target (a
cross) placed on a wall at eye level 2 m away. To prevent falls
during testing, an investigator stood close to the volunteers during
all tasks. For each balance condition, three trials of 30 s with 30 s
rest intervals were performed and the mean was retained for
analysis (da Silva, Martin, Parreira, Teixeira, & Amorim, 2013). A
landmark on the force platform was used to standardize the
position of the feet during all balance conditions (Fig. 3).
The vertical ground reaction force data from the force platform
were sampled at 100 Hz. All force signals were filtered with a 35-
Hz low-pass second-order Butterworth filter. The signals from four
sensors were converted into COP data using computerized
stabilography, which was compiled with MATLAB routines (The
 M.R. de Oliveira et al. / Archives of Gerontology and Geriatrics 59 (2014) 506–514
Fig. 3. Balance task conditions were controlled by a landmark on force platform that it was used to standardize the position of feet during all conditions: (A) two-legged stand;
(B) semi-tandem stand and (C) one-legged stand.
Mathworks, Natick, MA). Stabilographic analysis of COP data led to
the computation of the main balance parameters: the 95%
confidence ellipse area of COP (A-COP in cm2), and mean velocity
(VEL in cm/s) in both anteroposterior (A/P) and mediolateral (M/L)
directions of movement. The validity and reliability (based on the
infraclass coefficient correlation: ICC > 0.80 and standardized
error mean: SEM < 1.30) for these parameters computed with this
typical force platform have been accepted for both young adults
and elderly subjects, as reported by a recent study from our
research team (da Silva et al., 2013).
2.4. Statistical analysis
Homogeneity data was checked using Levene’s test for each
variable. One-way ANOVA was performed to compare the differ-
ences between the three groups in anthropometric characteristics
(age, weight, height and BMI), mental state (mini-mental scores) and
509
fear of falling (falls) variables. To determine the effects of
intervention (pre- and post-intervention) and exercise modalities
(MT, AG and GG), two-way ANOVA with repeated measures were
computed for each balance parameter (A-COP, VEL A/P and VEL M/
L) and under each task condition (TLEO, TLEC, STEO, STEC, ONE).
When necessary, post hoc Tukey’s test was used to identify
differences between the three groups (MT, AG and GG). Finally, to
determine the magnitude of changes, the effect size was
calculated based on the values of d (Glass & Hopkins, 1996)
following the equation below:
mpost  mpre
d¼
SD
where mpost is the mean value after the intervention and mpre is the
mean value before the intervention and basal SD (standard
deviation), that is, the SD of the pre-intervention measurement
510 M.R. de Oliveira et al. / Archives of Gerontology and Geriatrics 59 (2014) 506–514

Table 1 and GG groups were 0.41, 0.46 and 0.62, respectively. Finally, for
Characteristics of participants.
the one-legged stand, the MT, AG and GG groups were 0.31, 0.23
Groups and 0.25.
MT (n = 23) AG (n = 28) GG (n = 23) P
3.2. Comparison between groups (MT, AG and GG)
Age (years) 69  5 69  3 69  4 0.990
Height (cm) 154  7 153  5 152  4 0.627
Weight (kg) 66  12 67  10 66  9 0.913 There was no significant interaction (P > 0.05) between groups
BMI (kg/cm2) 27  3 28  5 28  4 0.709 and intervention for each condition of postural balance (Tables 2–6).
Cognitive statusa 25  3 24  3 23  6 0.399 Also, no difference was found in favor of any modality over another
Falls 23  8 25  7 21  8 0.277
in the post-intervention effect. Only a significance of the AG group
Values are presented as mean and standard deviation (). was observed at both times, pre- and post-intervention, in relation to
P > 0.05: differences between groups were not significant. the other groups for the two-legged stand (TLEC, Table 3) and semi-
BMI: body mass index.
tandem (STEO, Table 4) conditions. Although the three groups
Intervention exercise groups: mini-trampoline (MT), aquatic gymnastic (AG), and
general floor gymnastic (GG). improved in balance, the AG group showed lower values for both
a
Mini-mental state examination (normal range with cutoff > 18). pre- and post-intervention. However, the effects intra-group were
controlled by analysis of variance and thus this effect in the AG group
was relative to its own initial state (baseline measurement).
was used since it is an unbiased estimate of the true variance of the
elderly (Conboy, 2003). The effect size was characterized by Cohen 4. Discussion
(1988) as weak, moderate and strong effects, i.e., d = 0.2 is small,
d = 0.5 medium and d = 0.8 large, respectively. The statistical The present study included three modalities of exercise with
significance was set at P  0.05 and all analyses were run in the same experimental design during a training program for 12
Statistical Package for Social Sciences (SPSS) v. 20, United States of weeks, in order to determine the best effects on postural balance.
America. Interestingly, all types of exercise promoted positive effects on the
main COP parameters across the five balance conditions.
3. Results
4.1. Effect of the intervention
The groups were homogeneous for anthropometric character-
istics, mental state and fear of falling (falls), with no significant The results of the present study corroborate with previous
difference (P > 0.05) between groups (Table 1). studies that showed the benefit of exercise on postural balance
(Alfieri et al., 2012; Aragão et al., 2011; Avelar et al., 2010; Resende
3.1. Effect of the intervention (pre- and post-intervention) et al., 2008). From these four studies, only Resende and Avelar
quantified indirectly the risks of falls across a group of elderly
After 24 sessions of the exercise program, balance was subjects using the adapted Berg scale and functional test (timed-up
immediately assessed in the same week. Postural balance and go). Resende et al. (2008), who evaluated 25 older women
significantly (P < 0.05) improved after intervention with the three during aquatic training 2 per week (40-min each session), showed
modalities, with significantly lower COP values after compared to a mean difference in scores between pre- and post-intervention that
before intervention (Tables 2–6), except for the STEC condition improved from 1.24  0.97 to 7.60  3.77 of Berg score from the sixth
(Table 5). to the 12th week of the exercise program. Although there were positive
The effect from these interventions for the A-COP parameter results of a reduction in the risk of falls, the authors recognized the
varied from weak to moderate across the balance conditions. For limits of the study related to specificity and sensitivity of clinical
the TLEO condition, the MT, AG and GG groups were 0.54, 0.55, and outcomes, such as the Berg scale and/or balance functional tests like
0.25, respectively. For the TLEC condition, the MT, AG and GG timed-up and -go for balance and fall evaluation. There are ceiling
groups were 0.25, 0.14 and 0.20, respectively. For the STEO effects from these measurements and consequently poor sensitivity
condition, the MT, AG and GG groups were 0.63, 0.73 and 0.65, and responsiveness in balance measures for clinical practice (Mancini &
respectively (better results). For the STEC condition, the MT, AG Horak, 2010).

Table 2
Comparison of postural balance between groups for the two-legged stand condition with eyes open.

Condition Groups P ANOVA

MT AG GG Intervention Groups Interaction

TLEO
A-COP
Before 1.8  1.1 1.4  0.9 1.7  1.2 0.012z After < before 0.162 0.836
After 1.2  0.8 0.9  0.4 1.4  1.4

VEL A/P
Before 0.9  0.2 0.9  0.3 0.8  0.3 0.076 0.217 0.178
After 0.8  0.2 0.7  0.1 0.8  0.1

VEL M/L
Before 0.6  0.1 0.6  0.3 0.6  0.2 0.001z After < before 0.643 0.444
After 0.5  0.1 0.4  0.9 0.5  0.1

Values are presented as mean and standard deviation ().

Condition TLEO: two-legged stand with eyes open.
Balance parameters: A-COP (cm2): area of center of pressure; VEL A/P (cm/s): sway mean velocity of COP in anteroposterior direction; VEL M/L (cm/s): sway mean velocity of COP in
medio-lateral direction. Intervention exercise groups: mini-trampoline (MT), aquatic gymnastic (AG), and general floor gymnastic (GG).
z
Significant differences in COP values (P  0.05) between before and after intervention.
 M.R. de Oliveira et al. / Archives of Gerontology and Geriatrics 59 (2014) 506–514 511

Table 3
Comparison of postural balance between groups for the two-legged stand condition with eyes closed.

Condition Groups P ANOVA

GT AG GG Intervention Groups Interaction

TLEC
A-COP
Before 1.6  1.2 1.1  0.7 1.6  1 0.050z After < before 0.109 0.949
After 1.3  1.2 1  0.7 1.4  1

VEL A/P
Before 1.3  0.1 0.9 0.3 1.2  0.9 0.132 0.010y 0.706
GH < GT.GG
After 0.9  0.2 0.8  0.2 0.9  0.3

VEL M/L
Before 0.6  0.2 0.5  0.1 0.6  0.1 0.057 0.035y 0.838
GH < GT.GG
After 0.5  0.9 0.4 0.9 0.5  1

Values are presented as mean and standard deviation ().

Condition TLEC: two-legged stand with eyes closed.
Balance parameters: A-COP (cm2): area of center of pressure; VEL A/P (cm/s): sway mean velocity of COP in anteroposterior direction; VEL M/L (cm/s): sway mean velocity of COP in
medio-lateral direction. Intervention exercise groups: mini-trampoline (MT), aquatic gymnastic (AG), and general floor gymnastic (GG).
z
Significant differences in COP values (P  0.05) between before and after intervention.
y
Significant differences between groups (P  0.05).

Table 4
Comparison of postural balance between groups for the semi-tandem condition with eyes open.

Condition Groups P ANOVA

GT AG GG Intervention Groups Interaction

STEO
A-COP
Before 6.9  2.2 6.4  2.3 7.6  4.4 <0.001z 0.531 0.483
After < Before
After 5.5  2.2 4.7 2.5 4.7  2.6

VEL A/P
Before 1.7  0.6 1.6  1 1.8  1.2 0.131 0.290 0.701
After 1.6  0.5 1.2  0.5 1.7  0.9

VEL M/L
Before 1.8  0.1 1.6  0.5 1.9  1.3 0.096 0.031y 0.697
GH < GT. GG
After 1.7  0.4 1.4  0.4 1.7  0.8

Values are presented as mean and standard deviation ().

Condition STEO: semi-tandem with eyes open.
Balance parameters: A-COP (cm2): area of center of pressure; VEL A/P (cm/s): sway mean velocity of COP in anteroposterior direction; VEL M/L (cm/s): sway mean velocity of COP in
medio-lateral direction. Intervention exercise groups: mini-trampoline (MT), aquatic gymnastic (AG), and general floor gymnastic (GG).
z
Significant differences in COP values (P  0.05) between before and after intervention.
y
Significant differences between groups (P  0.05).

Table 5
Comparison of postural balance between groups for the semi-tandem condition with eyes closed.

Condition Groups P ANOVA

GT AG GG Intervention Groups Interaction

STEC
A-COP
Before 9.1  3.4 8.3  5 8.6  3.2 0.012z 0.388 0.884
After < before
After 7.7  7.7 6  3.6 6.6  4

VEL A/P
Before 2.3  0.7 1.9  0.7 1.8  0.7 0.319 0.080 0.541
After 2.0  0.8 1.6  0.6 1.8  0.9

VEL M/L
Before 2.5  0.7 2.1  0.81 2.2  1.1 0.244 0.073 0.645
After 2.3  0.9 1.8  0.5 2.2  0.8

Values are presented as mean and standard deviation ().

Condition STEC: semi-tandem with eyes closed.
Balance parameters: A-COP (cm2): area of center of pressure; VEL A/P (cm/s): sway mean velocity of COP in anteroposterior direction; VEL M/L (cm/s): sway mean velocity of COP in
medio-lateral direction. Intervention exercise groups: mini-trampoline (MT), aquatic gymnastic (AG), and general floor gymnastic (GG).
z
Significant differences in COP values (P  0.05) between before and after intervention.
512 M.R. de Oliveira et al. / Archives of Gerontology and Geriatrics 59 (2014) 506–514
Table 6
Comparison of postural balance between groups for the one-legged stand condition.
Condition Groups
GT (n = 11) AG (n = 15)
ONE
A-COP
Before 14  6.4 12 4.3
After 12  5.8 11  .37
VEL A/P
Before 3.7  1.9 3.9  2.6
After 3.2  1.3 2.8  1.6
VEL M/L
Before 3.8  1.1 3.7  1.2
After 3.4  0.9 3.0  0.5
Values are presented as mean and standard deviation ().
Condition ONE: one-legged stand.
Balance parameters: A-COP (cm2): area of center of pressure; VEL A/P (cm/s): sway mean velocity of COP in anteroposterior direction; VEL M/L (cm/s): sway mean velocity of COP in
medio-lateral direction. Intervention exercise groups: mini-trampoline (MT), aquatic gymnastic (AG), and general floor gymnastic (GG).
z
Significant differences in COP values (P  0.05) between before and after intervention.
The primary purpose of clinical balance assessments is to
identify whether a balance problem exists in order to predict the
risk of falls and to determine the effectiveness of intervention, as
the present study did. Quantitative postural control measures from
COP parameters with a force platform can overcome the main
drawbacks of the Berg scales and/or functional balance tests,
especially with regard to the variability during the test perfor-
mance, the subjective nature of scores in scales and mainly the
sensitivity to small changes (Mancini & Horak, 2010). The
strengths of the present study were the use of main COP
parameters, associated with a neurophysiologic approach based
on underlying sensory motor mechanisms of postural control in
different balance tasks (n = 5), which are sensitive to improve-
ments in balance during training as well as being able to predict
future falls and recurrent fallers across elderly subjects (Pirtola &
Era, 2006). Although the present study did not evaluate directly the
risk of falls before and after intervention, balance per se is one of
the major risk factors for falls among the elderly (Hsiao &
Simeonov, 2001; Lord et al., 2001; Pirtola & Era, 2006), and in turn,
infers some association with this phenomenon.
Aragão et al. (2011), who evaluated 21 elderly subjects using
body mass center parameters computed with a force platform to
determine the effectiveness of a MT exercise program at 14 weeks,
showed an improvement in dynamic balance stability and mainly
in the ability of elderly to regain balance during a sudden forward
fall (35% improvement after training, while the control group did
not show a statistical difference). These results support the
association between force platform measurements related to
postural control (e.g. low values after intervention) and falls
(reduction in follow-up or recurrences in the future), as stated
earlier by Pirtola and Era (2006). Alfieri et al., (2012), who enrolled
41 elderly in a comparison of multisensory and strength training
for postural control, reported a reduction of 20% on average in the
COP oscillation after 12 weeks (3 per week) of intervention,
although the risk of falls was not computed. It is therefore possible
that the magnitude of improvement found in the present study
(e.g.: 25% after MT exercise training during a two-legged stand
condition with the eyes closed for the A-COP variable), which is in
agreement with previous work (as Alfieri et al., 2012), can impact
positively on fall prevention for elderly subjects and, in turn,
contribute to the quality of life of these individuals.
The effect across five balance conditions was consistent for each
parameter of the force platform used in the study. For example,
with balance conditions for which there was contact with both feet
(TLEO, TLEC, STEO and STEC), there was a significant difference in
P ANOVA
GG (n = 13) Intervention Groups Interaction
10  3.9 0.198 0.079 0.933
9  4.4
3.1  1.1 0.048z 0.572 0.498
After < before
31
3.1  0.8 0.241 0.515 0.324
3.3  0.9
COP area, which represents the entire area used to adjust the
center of mass (COM) around the limits of support (Winter, 1995).
On the other hand, the condition using only one leg (ONE), showed
a difference in the variable VEL COP A/P, which in turn represents
the ratio between the magnitude of COP displacement and the time
interval required to perform this displacement in a small area
(Winter, 1995). All the COP parameters of different balance
conditions may have predictive value for subsequent falls,
especially the various indicators of lateral control of posture, such
as sway velocity of COP, which is also supported by Pirtola and Era
(2006).
Regardless of the specificity of the modality, regular exercise is
known as being beneficial to the health of adults and elderly
subjects (ACSM, 2011). In fact, regular exercise reduces the
deleterious effects of aging and can promote the maintenance of
functional capacities such as balance (Orr, 2010). The exercises
used in the present study apparently affected postural control,
stimulating sensory-motor response through reflex responses,
mechanisms of neural and motor control and mostly neuromus-
cular and biomechanical strategies of feedback and feedforward
for postural adjustments (Enoka, 1997; Howe et al., 2009; Orr,
2010). Enoka (1997) suggested that eight weeks of training can
achieve neural gains that facilitate peripheral motor actions in
physical performance of muscular strength, endurance and
balance. After eight weeks of training, histomorphological changes
can then promote the gains related to individual physical
performance (Duchateau & Enoka, 2002).
All the modalities used here stimulated muscular strength and
endurance of the lower and upper limbs, which can contribute to
balance gains (Orr, 2010). Also, it is possible that neuromotor
exercises from the three modalities promoted neurophysiologic
adaptations to increase reflex responses and proprioceptition,
which in turn impact the balance (Gandevia, 1998; Lopopolo,
Greco, Sullivan, Craik, & Mangione, 2006). It is further possible that
individual psychological motivation could also have influenced the
positive results on balance. However, this outcome was not
assessed nor controlled in the present study since all groups were
similarly motivated to participate in the exercise programs.
With regard to volume of training (2 week with approximately
60 min of training per session), this was performed similar to other
studies that also evidenced positive results in the balance (Elbar
et al., 2013; Melzer, Elbar, Tsedek, & Oddsson, 2008; Resende et al.,
2008). Several studies have found positive results in the postural
balance with a frequency of training, of at least two to three times/
week of 1 h or less. Along this line, Pluchino et al. (2012) evaluated 27
 M.R. de Oliveira et al. / Archives of Gerontology and Geriatrics 59 (2014) 506–514
elderly subjects in three types of intervention (Tai Chi, balance
exercise program and standardized Wii Fit balance program), twice a
week, totaling 16 sessions (eight weeks), and found that the three
programs were effective and improved postural control in variable
displacement and velocity of COP. Similarly, a study by Lelard,
Doutrellot, David, and Ahmaidi (2010) indicated improvement in
postural control of elderly people who attended training in Tai Chi
and balance for 12 weeks, twice a week in sessions of 30 min. In the
study of Seidler and Martin (1997), who assessed three weekly
sessions of balance training in a shorter period than the studies
presented above (five weeks), intervention also resulted in moderate
improvement in the postural control variables evaluated.
Finally, these studies support that both the frequency, session
duration and the period of the exercise program used in this study
are sufficient to produce significant improvements in the variables
of postural control in elderly women, although we found only one
study in the literature (Nagy et al., 2007) from six, in which the
interventions similar to our study did not lead to significant
improvements in these variables.
4.2. Effects between groups
This is the first study to determine whether a specific type of
intervention, such as the MT and aquatics, that apparently promote
more sensory-motor stimulus from their specific properties of
action, can produce better results on balance than exercises like
floor gymnastics. Unexpectedly, the MT and aquatic exercise
showed equally satisfactory results as gymnastics. Overall, the MT
promotes better balance strategies in motor coordination and
proprioceptive stimulus on an apparatus system (Heitkamp,
Horstmann, Mayer, Weller, & Dickhuth, 2001). However, aquatics
also promotes balance stimulus from water properties and
exercises that encourage the mechanisms of postural control
during constant displacement in water (Lord, Matters, & George,
2006). Likewise, exercises performed on the floor also improve
balance. Thus, it is expected that all types of exercises that promote
any neuromuscular capacities and reflect activities of daily living
can be used successfully for balance improvement and conse-
quently for promoting health in the elderly.
Furthermore, the effect size d for each modality across the five
balance conditions showed weak to moderate changes in the
magnitude of balance. Interestingly, the aquatic group presented
the nine best scores (0.14–0.73) compared to MT (0.25–0.63) and
GG (0.20–0.65), even though all modalities were similar in
intervention effects for improving balance. Also, the STEO
condition showed better effect size scores than the other balance
modalities (>0.63). An explanation for the best effect size score in
the aquatic group could be due to water properties (Salzman,
1998). However, there is no consensus in the literature, since
Avelar et al. (2010) reported no significant difference in balance
between groups trained on the floor versus in water. With regard to
balance, all these modalities of exercise are apparently efficient, as
pointed out in the present study, which in turn is relevant for
health professionals for the best clinical decision in prescribing
exercises for the elderly for reaching individual purposes.
4.3. Future perspectives
The findings of this study support the use of three exercise
modalities for improving balance during physical training pro-
grams for elderly subjects. The present work has uniquely
compared three types of exercise in the same experimental design.
Others in the literature have proposed the use of virtual reality during
functional physical training in the elderly through the use of video
games. A recent study reported positive results with this new tool of
intervention for balance gains in the elderly (Duque et al., 2013). In
513
the future, more research is needed to compare all these modalities
for the best clinical outcome, taking into account the time,
convenience, cost and access for elderly people.
4.4. Limitations of the study
Muscular strength, endurance, and flexibility measures as well
as psychological outcomes were not evaluated. The results can also
not be generalized to other physically independent elderly (with
different age and characteristics), nor to men. Another limitation is
that the protective effect of exercise was not evaluated (6–12
months after the intervention) as well as the prevalence and
incidence of falls after 12 weeks. However, positive results on
postural balance in elderly women were observed even with a
reduced intervention time of 12 weeks (24 sessions), and longer
periods would probably achieve better results and have more
impact on these variables. Balance is a predictor of falls and
evaluating this functional capacity makes it possible to infer
factors associated with both outcomes (balance versus falls).
5. Conclusion
All modalities of exercise investigated (MT, AG and GG) are
efficient for improving postural balance in elderly women after 12
weeks of training. These results provide further evidence
concerning exercise and balance and also contribute to the use
of new forms of exercise for health promotion in elderly women.
Conflict of interest
The authors declare no conflicts of interest in this work.
Acknowledgment
The authors would like to thank the students of the physical
education (UEL) and physical therapy (UNOPAR) courses who
assisted in data collection for the study. We also thank all
participants who contributed to the development of the study. A
CAPES scholarship supported the Master’s student from the
Rehabilitation program at the UNOPAR.
References
Alfieri, F. M., Riberto, M., Gatz, L. S., Ribeiro, C. P. C., Lopes, J. A. F., & Battistella, L. R.
(2012). Comparison of multisensory and strength training for postural control in
the elderly. Clinical Interventions in Aging, 7, 119–125.
American College of Sports Medicine, Chodzko-Zajko, W. J., Proctor, D. N., Fiatarone, M.
A., Minson, C. T., Nigg, C. R., et al. (2009). American College of Sports Medicine
position stand. Exercise and physical activity for older adults. Medicine & Science in
Sports & Exercise, 41, 1510–1530.
American College of Sports Medicine, Garber, C. E., Bryan, B., Deschenes, M. R., Franklin,
B. A., Lamonte, M. J., et al. (2011). Quantity and quality of exercise for developing
and maintaining cardiorespiratory, músculoskeletal, and neuromotor fitness in
apparently healthy adults: Guidance for prescribing exercise. Medicine & Sciences
in Sports & Exercise, 43, 1334–1359.
Aragão, F. A., Karamanidis, K., Vaz, M. A., & Arampatzis, A. (2011). Mini-trampoline
exercise related to mechanisms of dynamic stability improves the ability to regain
balance in elderly. Journal of Electromyography and Kinesiology, 21, 512–518.
Avelar, N. C. P., Bastone, A. C., Alcântara, M. A., & Gomes, W. F. (2010). Efetividade do
treinamento de resistência à fadiga dos músculos dos membros inferiores dentro e
fora d’água no equilı́brio estático e dinâmico de idosos. Revista Brasileira Fisioter-
apia, 14, 229–236.
Burke, T. N., França, F. J. R., Meneses, S. R. F., Cardoso, V. I., Pereira, R. M. R., Danilevicius,
C. F., et al. (2010). Postural control among elderly women with and without
osteoporosis: Is there a difference? Sao Paulo Medical Journal, 128, 219–224.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences. New Jersey:
Lawrence Erlbaum Associates.
Conboy, J. E. (2003). Algumas medidas tı́picas univariadas da magnitude do efeito.
Análise Psicológica, 2, 145–158.
da Silva, R. A., Martin, B., Parreira, R. B., Teixeira, D. C., & Amorim, C. F. (2013). Age-
related differences in time-limit performance and force platform-based balance
measures during one-leg stance. Journal of Electromyography and Kinesiology, 23,
634–639.
514 M.R. de Oliveira et al. / Archives of Gerontology and Geriatrics 59 (2014) 506–514
Duchateau, J., & Enoka, R. M. (2002). Neural adaptations with chronic activity patterns in
able-bodied humans. American Journal of Physical Medicine & Rehabilitation, 81, 17–27.
Duque, G., Boersma, D., Loza-Diaz, G., Hassan, S., Hamlet, S., et al. (2013). Effects of
balance training using a virtual-reality system in older fallers. Clinical Interventions
in Aging, 8, 257–263.
Elbar, O., Tzedek, I., Vered, E., Shvarth, G., Friger, M., & Melzer, I. (2013). A water-based
training program that includes perturbation exercises improves speed of voluntary
stepping in older adults: A randomized controlled cross-over trial. Archives of
Gerontologic and Geriatrics, 56, 134–140.
Enoka, R. M. (1997). Neural adaptations with chronic physical activity. Journal of
Biomechanics, 30, 447–455.
Gandevia, S. C. (1998). Neural control in human muscle fatigue: Changes in muscle
afferents, moto neurones and moto cortical drive. Acta Physiologica Scandinavica,
162, 275–283.
Glass, G., & Hopkins, K. (1996). Statistical methods in education and psychology (3rd Ed.).
Needham Heights, MA: Allyn & Bacon.
Granacher, U., Muehlbauer, T., Gollhofer, A., Kressig, R. W., & Zahner, L. (2011). An
intergenerational approach in the promotion of balance and strength for fall
prevention – a mini-review. Gerontology, 57, 304–315.
Heitkamp, H. C., Horstmann, T., Mayer, F., Weller, J., & Dickhuth, H. H. (2001). Gain in
strength and muscular balance after balance training. International of Journal Sports
Medicine, 4, 285–290.
Horak, F. B. (2006). Postural orientation and equilibrium: What do we need to know
about neural control of balance to prevent falls? Age and Ageing, 35, ii7–ii11.
Howe, T. E., Rochester, L., Jackson, A., Banks, P. M. H., & Blair, V. A. (2009). Exercise for
improving balance in older people (Review). Cochrane Database of Systematic
Reviews, 1–150.
Hsiao, H., & Simeonov, P. (2001). Preventing falls from roofs: A critical review.
Ergonomics, 44, 537–561.
Hughes, M. A., Duncan, P. W., Rose, D. K., Chandler, J. M., & Studenski, S. A. (1996). The
relationship of postural sway to sensorimotor function, functional performance, and
disability in the elderly. Archives of Physical Medicine Rehabilitation, 77, 567–572.
Jehu, D. B. (2012). The effects of a 12-week multifactorial exercise and balance training
program on balance control in older adults. St. Catharines: [s.n].
Lelard, T., Doutrellot, P.-L., David, P., & Ahmaidi, S. (2010). Effects of a 12-week Tai Chi
Chuan program versus a balance training program on postural control and walking
ability in older people. Archives of Physical Medicine and Rehabilitation, 1, 9–14.
Lopopolo, R. B., Greco, M., Sullivan, D., Craik, R. L., & Mangione, K. K. (2006). Effect
of therapeutic exercise on gait speed in community-dwelling elderly people:
A meta-analysis. Physical Therapy, 86, 520–540.
Lord, S. R., Sherrington, C., & Menz, H. B. (2001). Falls in older people: risk factors and
strategies for prevention. New York: Cambridge University Press.
Lord, S. R., Matters, B., & George, R. S. (2006). The effects of water exercise on physical
Functioning on older people. Australisian Journal on Ageing, 25, 36–41.
Mancini, M., & Horak, F. B. (2010). The relevance of clinical balance assessment tools to
differentiate balance deficits. European Journal of Physical and Rehabilitation Medi-
cine, 46, 239–248.
Melzer, I., Elbar, O., Tsedek, I., & Oddsson, L. (2008). A water-based training program
that include perturbation exercises to improve stepping responses in older
adults: Study protocol for a randomized controlled cross-over trial. BMC
Geriatric, 17, 8–19.
Mian, O. S., Baltzopoulos, V., Minetti, A. E., & Narici, M. V. (2007). The impact of physical
training on locomotor function in older people. Sports Medicine, 8, 663–701.
Nagy, E., Feher-Kiss, A., Barnai, M., Domján-Preszner, A., Angyan, L., & Horvath, G.
(2007). Postural control in elderly subjects participating in balance training.
European Journal of Applied Physiology, 1, 97–104.
Orr, R. (2010). Contribution of muscle weakness to postural instability in the elderly.
European Journal of Physical Rehabilitation and Medicine, 46, 183–220.
Pirtola, M., & Era, P. (2006). Force platform measurements as predictors of falls among
older people – a review. Gerontology, 52, 1–16.
Pluchino, A., Lee, S. Y., Asfour, S., Roos, B. A., & Signorile, J. F. (2012). Pilot study
comparing changes in postural control after training using a video game balance
board program and 2 standard activity-based balance intervention programs.
Archives of Physical Medicine and Rehabilitation, 11, 38–46.
Resende, S. M., Rassi, C. M., & Viana, F. P. (2008). Effects of hidrotherapy in balance
and prevention of falls among elderly women. Revista Brasileira de Fisioterapia,
12, 57–63.
Rydwik, E., Welmer, A. K., Kåreholt, I., Angleman, S., Fratiglioni, L., & Wang, H. X. (2013).
Adherence to physical exercise recommendations in people over 65. The SNAC-
Kungsholmen study. European Journal of Public Health, 23, 799–804.
Salzman, A. P. (1998). Evidence-based aquatic therapy for proprioceptive-training. The
aquatic resources network. Atri’s Aquatic Symposium (pp. 95–99).
Seco, J., Abecia, L. C., Echevarrıa, E., Barbero, I., Torres-Unda, J., Rodriguez, V., et al.
(2013). A long-term physical activity training program increases strength and
flexibility, and improves balance in older adults. Rehabilitation Nursing, 38, 37–47.
Seidler, R. D., & Martin, P. E. (1997). The effects of short term balance training on the
postural control of older adults. Gait & Posture, 3, 224–236.
Spirduso, W. (1995). Physical dimension of aging. Illinois: Human Kinetics.
Winter, D. A. (1995). Human balance and posture control during standing and walking.
Gait Posture, 3, 193–214.