Journal of Aging and Physical Activity, 2011, 19, 291-305
© 2011 Human Kinetics, Inc.
Effect of a Home Exercise Program
on Dynamic Balance in Elderly
With a History of Falls
Sharon L. Olson, Shu-Shi Chen, and Ching-Yi Wang
Objective: To determine exercise efficacy in improving dynamic balance in
community-dwelling elderly with a fall history. Methods: Thirty-five participants were randomly assigned to a treatment (TG; n = 19, 77 ± 7 yr) or control
group (CG; n = 16, 75 ± 8 yr). The TG received an individualized home exercise program, and the CG received phone calls twice per week for 12 weeks.
Participants’ dynamic-balance abilities— directional control (DC), endpoint
excursion (EE), maximum excursion (ME), reaction time (RT), and movement
velocity (MV)—were measured using the Balance Master at 75% limits of
stability. Functional reach (FR) was also measured. Results: At 12 weeks the
TG demonstrated significant improvements in DC (p < .0025), EE (p < .0005),
and ME (p < .0005), but the CG did not. No significant group differences were
found for MV, RT, or FR. Conclusions: Excursion distances and directional
control improved but not reaction time, suggesting that exercises requiring quick
responses may be needed.
Keywords: limits of stability, physical therapy, geriatrics
Falls in older adults are a leading cause of physical disability, functional decline,
emotional stress, and death (Fletcher & Hirdes, 2004; Stel, Smit, Pluijm, & Lips,
2004; Stevens, Mack, Paulozzi, & Ballesteros, 2008). Falling is a multifactorial
situation that results from the accumulated effects of intrinsic and extrinsic factors
(Bucher, Szczerba, & Curtin, 2007). An expert panel on falls prevention pointed to
muscle weakness, history of falls, gait deficits, and balance deficits as major factors
for falls in the elderly (American Geriatrics Society, British Geriatrics Society, &
American Academy of Orthopaedic Surgeons, 2001). Approximately 10–25% of
all falls have been attributed to poor balance (Shumway-Cook, Gruber, Baldwin,
& Liao, 1997). Balance is a complex skill based on the interaction of dynamic
sensorimotor processes and is dependent on the goal of the movement task, as well
as the environmental context (Horak, 2006). The ability to control balance involves
Olson and Chen are with the School of Physical Therapy, Texas Woman’s University, Houston, TX.
Wang is with the School of Physical Therapy and Center for Education and Research on Geriatrics and
Gerontology, Chung Shan Medical University, Taichung, Taiwan.
291
292 Olson, Chen, and Wang
using strategies to stabilize the body’s center of gravity over the base of support
during quiet standing or active movement—static or dynamic balance (Arampatzis,
Peper, & Bierbaum, 2010). The limits-of-stability (LOS) measure is defined
as the area over which an individual can lean in specified directions and displace
the center of gravity without changing the base-of-support stance as defined by
the position of the feet (Liston & Brouwer, 1996). The functional base of support
declines about 16% per decade beyond age 60. Forward- and backward-leaning
abilities also decline significantly after age 60; that population on average retains
only 66% of the forward-leaning ability and 34% of the backward-leaning ability
of those under age 60 (King, Judge, & Wolfson, 1994).
Exercises to improve joint flexibility, muscle strength, and balance have been
the focus of many falls-prevention programs for older adults (Chang et al., 2004;
Faber, Bosscher, Chin A Paw, & van Wieringen, 2006; Hauer, Pfisterer, Schuler,
Bartsch, & Oster, 2003; Lin, Wolf, Hwang, Gong, & Chen, 2007). Ankle flexibility in particular has been shown to predict balance ability (Menz, Morris, &
Lord, 2005). High-intensity strength training of postural muscles in older adults
with balance impairment has been shown to result in improved functional-balance
control (Hess & Woollacott, 2005). A published review of falls-prevention interventions concluded that balance training is a key component of exercise programs
in the reduction of fall rates (Sherrington et al., 2008). In summary, these results
imply that exercise is effective in improving various components of balance
and that improved balance in turn reduces an individual’s risk of experiencing
another fall.
Previous studies have used group exercise protocols, home exercise protocols,
or individualized home exercise programs and found that all these exercise methods
can improve balance, if exercise adherence is good, and that the program duration
should be at least 12 weeks (Costello & Edelstein, 2008: Cyarto, Brown, Marshall,
& Trost, 2008; Davis et al., 2010). The advantage of home exercise programs is that
they can be less costly and do not require participant transportation, but program
adherence and timely exercise progression can be problematic.
In a systematic review, Howe, Rochester, Jackson, Banks, and Blair (2008)
concluded that balance in older adults can be improved with exercise programs
and that muscle strengthening and dynamic-balance activities had a large impact
on this improvement. They also reported the specific balance measures used in
the reviewed studies and demonstrated that—for multiple interventions, as well
as just balance and strengthening exercises—functional reach, the Berg Balance
Scale, and measures of static stability were used most frequently. Fewer published
intervention studies that have tested LOS have demonstrated that this measure
of dynamic balance can capture specific balance components not addressed
with other measures (Bulat et al., 2007; Islam et al., 2004; Tsang & Hui-Chan,
2004).
The purpose of this study was to assess the efficacy of home exercise programs developed to address balance deficits in a population of older adults with a
recent history of falls by testing their dynamic-balance capability before and after
a 12-week intervention. We hypothesized that improvements would be seen in
specific LOS test parameters and the clinical functional-reach test in the treatment
group compared with the control group.
Effect of Home Exercise on Falls 293
Method
Study Population
Community-dwelling participants were recruited from local senior citizen organizations by an individual who was blinded to group assignment. Inclusion criteria
for all participants were being over 60 years of age, ability to follow commands,
awareness of time and place, and having experienced a fall in the past year. A fall
was defined as an event that resulted in the person landing on the floor or a surface
below knee level that was not caused by a severe blow, unconsciousness, paralysis,
or seizure (Lach et al., 1991). Participants were provided this definition of a fall and
then asked to report the number of falls they had experienced during that time frame.
They were interviewed and excluded if they reported acute or unstable medical
conditions such as uncontrolled hypertension or diabetes, had been diagnosed with
osteoporosis, or reported difficulty standing or walking for 10 min because of other
health problems. A priori power analysis was conducted based on an effect size
for functional reach developed from a pilot study of 17 participants with a history
of falls who received a similar individualized 12-week intervention of flexibility,
strengthening, and balance exercises, with baseline and postintervention functionalreach assessment. The calculated effect size of 1.42 was used to determine that a
minimum group size of 7 participants was needed to achieve a power of .95. We
therefore concluded that a minimum of 15 participants in each group at the end
of the study would have sufficient power to avoid a Type II error at a .025 alpha
level. Considering a possible attrition rate of 25% for this two-group study, we
recruited 43 participants who consented to participate. Two recruited volunteers
did not meet the study criteria, but 41 participants received a concealed assignment
to either a treatment or a control group through use of a random-number table. As
illustrated in Figure 1, 4 of the 41 original people who met the inclusion criteria
and volunteered to participate, 1 in the treatment group and 3 in the control group,
were excused from the study during the first week (Schulz, Altman, & Moher, 2010).
One participant had an exacerbation of her rheumatoid arthritis, 1 developed severe
lower back pain, 1 moved out of town, and 1 did not return from vacation for the
first scheduled appointment. Therefore, 37 participants began the study. Informed
consent was obtained from all participants before their participation, in accordance
with the policies of the facility’s institutional review board.
Measures
All outcome assessments were conducted in a university balance laboratory. Dynamic
balance was evaluated by the 75% LOS test on a computerized posturography platform (Portable Balance Master, Neurocom, International, Inc., Clackamas, OR), with
software version 5.0. This test was selected based on the results of a previous project
that demonstrated target achievement at 75% LOS to be very challenging for a different set of participants with fall histories. Jbabdi, Boissy, and Hamel (2008) also
found that their population of community-living older adults reached less than 75%
of their theoretical LOS in the forward direction and less than 55% in lateral directions. The posturography platform houses ­force-plate transducers used to compute
vertical pressure from the standing ­participant’s feet. The software calculates the
Figure 1 — Flow diagram of study participants.
294
Effect of Home Exercise on Falls 295
center of gravity and the theoretical LOS from the participant’s height and weight. A
monitor positioned at eye level displays a central starting position and eight targets
evenly distributed in cardinal and diagonal directions around the perimeter of a circle
representing the 75% LOS.
The measured parameters of the 75% LOS test include directional control
(DC), endpoint excursion (EE), maximum excursion (ME), maximum velocity
(MV), and reaction time (RT). The DC value is the amount of toward-target centerof-gravity movement relative to away-from-target center-of-gravity movement.
DC is expressed as a percentage of the total toward-target movement, and a higher
percentage reflects better movement coordination. EE is the distance (%LOS) of
the initial attempt to reach a target. A higher EE value indicates a more accurately
anticipated amount of movement needed to reach the target. ME is the farthest distance (%LOS) toward the target achieved during the trial. MV (°/s) is the average
speed of center-of-gravity displacement, and RT is the time in seconds between
the command to move and the subject’s movement response.
Participants were instructed to stand barefoot on the force plate with their arms
at their sides and to keep their feet in the same correct position on a permanent grid
throughout the testing procedure. They wore a gait belt with loops that were used by
an assisting researcher to ensure safety but not guide or interrupt their movement.
When prompted by auditory and visual cues, participants were expected to move
their bodies to shift the center-of-gravity cursor as accurately as possible toward
each of eight targets within 8 s and to remain there until instructed to return to the
starting position. One practice trial was undertaken to familiarize participants with
the procedure before testing. Two test trials were then performed. The reported
generalizability coefficients over 2 consecutive days for the DC, EE, ME, and MV
measures of this test ranged from .58 to .87, suggesting a wide range of reliability
values (Clark & Rose, 2001). We examined the procedural reliability of the LOS
Balance Master test on our population by calculating the test–retest reliability using
intraclass correlation coefficient (ICC) Model 2 analysis. Our results demonstrated
a range of acceptable reliability values for two-trial averages, from .64 for DC and
RT to >.90 for EE, ME, and MV.
Functional reach was first posited as a clinical measure of dynamic balance
in 1990 (Duncan, Weiner, Chandler, & Studenski, 1990) and has been validated
as both a predictor of fall risk (Muir, Berg, Chesworth, Klar, & Speechley, 2010;
Murphy, Olson, & Protas, 2003) and an intervention outcome measure (Howe et
al., 2008) since that time. For the clinical functional-reach measure, participants
were asked to lift their dominant arm to shoulder level, with their elbows fully
extended, and then reach as far forward as possible without moving their feet.
The reach distance was measured. Each participant performed one practice trial
and two test trials. The functional reach was repeated if the participant stepped or
demonstrated excessive trunk rotation during any trial. Within-session test–retest
reliability—ICC(2, 2)—for this measurement in our population was .97.
Exercise Protocol
Most treatment-group participants exercised at home, but one third of them preferred to exercise in a local clinical facility or come to the laboratory twice a week
for the intervention. Regardless of exercise location, the selected exercises could
be carried out in a home environment. Each home exercise program addressed
296 Jehn et al.
the individual participant’s specific impairments in lower extremity joint range
of motion, muscle strength, and balance, as determined at the treatment group’s
baseline evaluation and the control group’s 12-week evaluation. Exercise difficulty
was increased as the participants demonstrated progress, generally on a weekly
basis. Flexibility exercises started with 10 repetitions throughout available range
and progressed to 30 repetitions with a 5-s hold at end of range. Strengthening
exercises started with a 10-repetition maximum using elastic-band resistance
and progressed to three sets of 10 repetitions before the resistance level of the
elastic band was increased. Balance exercises began with shifting the center of
gravity over a stable base of support for as few as 5 min and progressed to 30
min of wide-ranging tasks including dynamic-balance activities such a grapevine
stepping and tandem walking. All treatment-group participants were able to complete 24 sessions that included progressive flexibility, strengthening, and balance
exercises as indicated by their baseline testing. See Table 1 for exercise program
examples.
The principal investigator and two additional physical therapists, who
received three training sessions in the exercise program by the principal investigator, provided the onsite monitoring and progression of exercises. They met
weekly to discuss participant progress and the exercise progression to control
for procedural differences. The therapists were to observe the exercise program
and increase the level of exercise difficulty once milestones were met, similar to
clinical reevaluation and progression of home exercise programs.
Table 1 Sample Home Exercise Program
Exercise
program
Duration
(min)
Focus
Type and frequency of exercise
1
5–10
Range
of motion
To increase joint mobility (hip, knee, or ankle),
participants moved actively through the available
range and held at end range for 10 s, then progressed
from 10 to 30 repetitions for each restricted joint.
2
15
Strength
To strengthen weak lower extremity muscles,
participants used graded-resistance, elastic exercise
bands beginning with 10-repetition-maximum
resistance and gradually increasing to 3 sets of 10
repetitions. Progressed with new 10-repetitionmaximum resistance from one to three sets.
3
5–30
Balance
Balance tasks were individualized, according to
participant need and ability. Time spent on weight-­
bearing balance activities progressed from 5 to
30 min as exercise tolerance increased. Examples
included moving center of gravity to extremes
of base of support in sitting and standing, static
standing balance in tandem and single-leg stance
with eyes open and closed, and dynamic balance
activities such a grapevine steps, ball sitting and
throwing, and fast gait activities with obstacles.
Effect of Home Exercise on Falls 297
General Procedure
All assessments were conducted by one of two possible evaluators working together,
with one evaluator ensuring patient safety while the other gave instructions and
performed the tests. Each outcome measure was assessed by the same evaluator
throughout the study. The evaluators were not blinded to group assignment, as they
also monitored some of the exercise programs. At the baseline and 12-week testing
sessions, LOS and functional reach were assessed in random order with rest periods
as needed between tests. During the baseline session, each participant also received
an individual evaluation that included a personal medical history, measurement
of height and weight, and a brief musculoskeletal examination to assess lower
extremity joint range of motion and muscle strength. Active range of motion of
the hips, knees, and ankles was measured with a universal goniometer according
to standard protocol (Norkin & White, 2009). A handheld dynamometer was used
to test leg-muscle strength in standard postures for manual muscle testing (Wang,
Olson, & Protas, 2002). Immediately after this baseline testing to determine deficits
for each participant, individually prescribed exercises were started by the treatment
group. They performed their progressive home exercise program independently but
always under a physical therapist’s onsite monitoring and direction twice a week
for 12 consecutive weeks. The duration of the exercise sessions ultimately averaged
55 min but varied slightly according to the initial endurance, strength, and flexibility, as well as the progress, of each participant. The control-group participants
received friendly phone calls from two graduate research students twice per week
between the baseline and 12-week testing, after which they were given a similar
home exercise program but without onsite monitoring or postexercise testing.
The students were directed to ask about recent health in general and whether the
participant had fallen since the last phone call. If a fall had occurred, the students
asked what the participant was doing at the time of the fall and whether he or she
sustained injuries requiring medical assistance. This information was then recorded
and placed in the participant’s record. These participants were advised not to begin
additional physical activities or exercise programs during the 12 weeks.
Statistical Analyses
Demographic data were tabulated, with comorbidities expressed as an average
number reported per person. The differences in baseline characteristics and fall
incidence were compared between the treatment and control groups. The assumption of homogeneity of variance for the five parameters and functional reach was
also tested. One-way analyses of variance (ANOVA) were conducted to determine
differences in the five LOS parameters and the functional reach between the treatment and control groups at baseline.
For the 75% LOS test, a composite score for each parameter was calculated
by averaging the scores from all eight targets for each trial, followed by averaging
the trial scores. For the functional-reach test, the two trial scores were averaged.
A two-factor multivariate ANOVA (MANOVA) with repeated measures over time
for the LOS measures and a two-factor univariate ANOVA with repeated measures
over time for the functional reach were conducted, with a Bonferroni adjustment of
the alpha level to .025 to control for a Type I error. Follow-up univariate ANOVAs
for the LOS measures were also conducted as warranted. When group-by-time
interactions occurred, simple effects were tested, at a conservative alpha level of
298 Olson, Chen, and Wang
.01, using one-way repeated-measures ANOVAs for the time factor. All data were
analyzed using SPSS, version 16.
Results
Two control-group participants did not take the 75% LOS test at the 12th week.
One was unable to perform the test in the time allowed, and a temporary electrical
problem occurred at the time of the other participant’s 12-week test. Therefore, the
final study sample analyzed was 35, with 19 participants in the treatment group and
16 in the control group (Figure 1). A number of participants reported comorbidities, namely, arthritis (n = 13), diabetes (n = 7), heart disease (n = 6), stroke (n =
5), Parkinson’s disease (n = 1), traumatic brain injury (n = 1), visual problems (n
= 9), hearing problems (n = 12), and dizziness (n = 18). One participant in each
group reported falling more than 30 times in the past year, and we calculated two
equivalent modes for fall frequency that were one (n = 9) or three (n = 9) falls.
There were no significant baseline differences in the demographics, the number of
comorbidities per participant, or the fall incidence between the two groups (Table 2).
The assumption of homogeneity of variance was tenable for the five LOS
parameters (p ≥ .2) and the functional reach (p ≥ .5). For the 75% LOS-test data,
there were no differences in any of the five parameters between the treatment and
control groups at baseline. The results of the split-plot MANOVA revealed the
effect of group to be nonsignificant, F(5, 29) = 0.20, p < .1; the effect of time to
be significant, F(5, 29) = 7.03, p < .0005; and the group-by-time interaction to be
significant, F(5, 29) = 6.27, p < .0005, according to Pillai’s criterion. A posteriori
calculation showed power to be .97 for the MANOVA at .025 alpha.
Follow-up univariate analyses, as presented in Table 3, showed that there were
statistically significant interactions for DC, F(1, 33) = 13.86, p < .0015, 1 – β = .89;
Table 2 Demographic Characteristics and Fall Incidence by Group
(N = 35)
Characteristic
Age, years, M ± SD (range)
Weight, kg, M ± SD
Height, m, M ± SD
Race, n
White
African American
Hispanic
Gender, n
male
female
Fall incidence
once only
occasional (2 or more times)
Average number of comorbidities
per subject
aStudent’s
t test. bChi-square test.
Treatment group
(n = 19)
77 ± 7 (62–90)
73.4 ± 16.1
1.66 ± 0.13
Control group
(n = 16)
75 ± 8 (61–87)
80.5 ± 15.3
1.67 ± 0.08
14
3
2
10
5
1
9
10
8
8
p
.49a
.19a
.67a
.54b
.88b
.26b
7 (36.8%)
12 (63.2%)
2 (12.5%)
14 (87.5%)
2.2
1.9
.37a
299
Parameter
*Significant difference, p < .025.
Directional control, %
Endpoint excursion, %
Maximum excursion, %
Movement velocity, °/s
Reaction time, s
61.61 ± 14.81
44.91 ± 10.70
56.78 ± 12.21
1.89 ± 0.68
1.10 ± 0.26
Treatment group
68.59 ± 10.75
51.37 ± 11.83
62.22 ± 12.80
2.16 ± 0.95
1.01 ± 0.44
Control group
Pretreatment (Week 1)
73.75 ± 11.17
57.74 ± 7.90
70.37 ± 6.75
2.40 ± 0.75
0.96 ± 0.22
Treatment group
64.01 ± 15.30
50.78 ± 11.47
65.46 ± 13.00
2.37 ± 0.96
0.91 ± 0.28
Control group
Posttreatment (Week 12)
13.86
24.37
11.52
3.18
0.15
F(1, 33)
.001*
.0005*
.002*
.084
.71
p
Group × Time
Table 3 Means, Standard Deviations, and ANOVA Results for the 75% Limits-of-Stability Test Parameters
300 Olson, Chen, and Wang
EE, F(1, 33) = 24.37, p < .0005, 1 – β = 1.0; and ME, F(1, 33) = 11.52, p < .0025,
1 – β =. 91, but no interaction effects were found for MV, F(1, 33) = 3.18, p < .1,
1 – β = .41, or RT, F(1, 33) = 0.15, p > .7, 1 – β = .07. The main effect of time was
significant for MV, F(1, 33) = 17.56, p < .0005, but there was no interaction. The
one-way repeated-measures ANOVAs showed that there was a significant improvement in DC, F(1, 18) = 13.85, p < .0025, d = 1.06; EE, F(1, 18) = 46.39, p < .0005,
d = 2.33; and ME, F(1, 18) = 42.49, p < .0005, d = 2.54, in the treatment group
after 12 weeks of exercise, but not in the control group. The pre–post control-group
effect sizes were d = 0.6 for DC, d = 0.11 for EE, and d = 0.51 for ME.
The one-way ANOVA showed that there was no difference in functional reach
between the treatment and control groups at baseline. The split-plot ANOVA for
functional reach demonstrated a significant main effect for time, F(1, 33) = 15.04,
p < .0005, but neither main effect of group, F(1, 33) < 0.0005, p > .9, nor groupby-time interaction, F(1, 33) = 0.02, p > .85, 1 – β = 0.05, was found (Figure 2).
Figure 2 — Means and standard deviations of functional reach for the treatment and
control groups.
Effect of Home Exercise on Falls 301
Discussion
The results supported the study hypothesis that a 12-week home exercise program
can improve specific dynamic-balance skills of older community-dwelling adults
with a history of falls. This outcome is similar to the findings of Donat and Ozcan
(2007), who compared group- and home-based exercises undertaken by older
adults with fall risk three times per week for 8 weeks and found significant balance improvement in both groups. Kamide, Shiba, and Shibata (2009) assessed
balance in elderly women after a 6-month unsupervised home exercise program
and demonstrated significant improvements in their balance measures as compared
with a control group, suggesting that program supervision, as provided in the current study, is not needed to achieve improvement in a healthy population of older
adults. However, Donat and Ozcan contacted participants monthly to encourage
continued participation.
The treatment group demonstrated significant improvement in DC, EE, and ME,
supporting the results of other studies that assessed these LOS parameters before
and after exercise interventions focused on strength or balance control (Bulat et al.,
2007; Islam et al., 2004; Tsang & Hui-Chan, 2004). The percent improvement in
postintervention mean values of the treatment group was clinically meaningful for
DC (20%), EE (29%), and ME (24%), but control-group DC and EE mean values
decreased over time, and the mean value of ME for the control group increased
only 0.05%. Not all current study measures improved in treatment versus control
group; MV and RT did not have statistically significant interactions. Bulat et al.
also did not demonstrate significant improvement in MV and RT at a .01 alpha
level in their pretest–posttest assessment of an 8-week group exercise program
for older veterans, but they did report a trend of improvement in those measures.
There was a statistically significant increase in mean MV across groups, with
a mean score improvement of 27% in the treatment group and 9.7% in the control
group. Within-group variance was therefore a likely factor in this nonsignificant
interaction result for MV. The lack of any significant improvement in RT in our
study contrasts with other work (Rooks, Kiel, Parsons, & Hayes, 1997) that demonstrated quicker foot-press reaction time after a 10-month intervention program
for community-dwelling older adults consisting of either resistance training or
walking activity as compared with a control group. However, their subjects were
seated for the test and the reaction-time measurement did not entail an upright
postural response. Vogler, Sherrington, Ogle, and Lord (2009) also demonstrated
improved reaction time after 12 weeks of weight-bearing exercise in a population
of older adults just dismissed from a hospital. They measured reaction time by a
finger-press response, which also did not challenge upright postural stability. In
addition, their population was not limited to older adults with a fall history, nor were
groups stratified according to fall risk. Lajoie and Gallagher (2004) suggested that
even verbal-response times during conditions of postural demand can differentiate
participants according to fall history.
The exercise programs, which included flexibility, strength, and balance training as indicated, were specifically designed so that treatment-group participants
could perform them safely on their own at home, even though they were monitored
twice per week during this study to ensure program adherence, which was 100%.
302 Olson, Chen, and Wang
No aggressive postural-perturbation exercises that would require rapid reactive
responses were included in the home exercise programs because of safety concerns
for this population at risk for falling. Therefore, the home exercise program simply
might have been ineffective in improving anticipatory RT in the treatment group
as compared with the control group.
Specificity of exercise is a factor to consider when designing interventions to
improve specific balance components, as well as adding activities requiring quick
responses, to improve RT and MV. Improvements in the other LOS parameters
suggest greater ability to manage the larger moment arms created by the center of
gravity shifting farther away from the base of support. Activities of daily living
requiring large movements of the trunk during standing should be easier with this
improved dynamic control.
Functional reach did not improve significantly more in the treatment group
than the control group, a finding that differed from our pilot study’s results. Ballard,
McFarland, Wallace, Holiday, and Roberson (2004) found similar results after a
15-week exercise program for elderly women consisting of resistance and balance
exercises comparable to those used in the current study. Wernick-Robinson, Krebs,
and Giorgetti (1999) suggested that functional-reach distance can be increased by
a variety of movement strategies that do not always increase the moment arm produced when the center of pressure is behind the center of gravity. Without significant
moment-arm increases during the task, the functional-reach test then becomes a
more static and less dynamic measure of balance. Wernick-Robinson et al. cited
hip flexion and trunk rotation as two strategies that produced a low moment arm.
The current study evaluators encouraged participants to perform the functional
reach correctly, but it is possible that subtle changes in movement strategy were
not observed. Although functional reach is an accepted clinical measurement for
dynamic balance, Wallmann (2001) found no significant relationship between the
anterior displacement on the LOS test and the functional reach. Our study findings
also suggest that instrumented LOS tests can capture aspects of dynamic balance
not detected with the functional-reach test.
Study limitations include our choice to use the 75% LOS test rather than the
100% LOS test, so we might not have measured participants’ full capability (Clark
& Rose, 2001; Clark, Rose, & Fujimoto, 1997), thereby creating a ceiling effect.
However, examination of the composite scores suggested that this was not the
case. Reaching for at least seven of the eight targets at 75% LOS was very challenging for our participants. In addition, evaluators were not purposefully blinded
to participants’ group during the assessments, and tester bias might have affected
the functional-reach measurement, which is a weakness of the study. However, the
LOS assessment was automated and the results were retrieved and calculated at
a later date by a third examiner who did not participate in either the testing or the
intervention. Standard instructions were given to all participants, and they received
no verbal guidance during the tests themselves. The wide variation in reliability
of the 75% LOS measures, with ICC values less than .70 for DC and RT, is also
a possible study weakness—it could account for a larger true-score error range.
Other limitations of this research were the relatively small sample size and the
variation in dynamic-balance skills of the sampled population. The participants of
the current study exhibited a wide age range and reported multiple comorbidities.
However, this heterogeneity did not appear to affect the statistical power of our
Effect of Home Exercise on Falls 303
overall analysis, which was high. This study also reflected the realities of clinical
falls-prevention programs with heterogeneous populations.
The results of this study suggest that a home exercise program is safe and effective in improving selected dynamic-balance control parameters in older adults who
have fallen. These positively affected parameters reflected the range and control
of movement rather than the speed of movement, which is important for reacting
to perturbations, a frequent cause of falling. More challenging balance exercises
that require quick responses and higher speeds should be included, if the client can
perform them safely in the home, to determine whether RT in weight bearing can
actually be improved with exercise intervention. Regular exercise progression was
integral to this study, and therefore frequent, periodic reevaluation is recommended
for those exercising without supervision. Periodic group exercise sessions might be
considered as a useful addition to a home exercise program to include perturbation
exercises, as well as to increase the program difficulty and encourage adherence.
Acknowledgments
This study was supported by a grant from the Foundation for Physical Therapy. We wish
to acknowledge the contributions of Elizabeth J. Protas, Clay Smith, Michael Pile, Elaina
Iazzetti, and Jayme Scott.
References
American Geriatrics Society, British Geriatrics Society, & American Academy of Orthopedic
Surgeons. (2001). Guideline for the prevention of falls in older persons. Journal of the
American Geriatrics Society, 49(5), 664–672.
Arampatzis, A., Peper, A., & Bierbaum, S. (2010). Exercise of mechanisms for dynamic
stability control increases stability performance in the elderly. Journal of Biomechanics.
Ballard, J.E., McFarland, C., Wallace, L.S., Holiday, D.B., & Roberson, G. (2004). The
effect of 15 weeks of exercise on balance, leg strength, and reduction in falls in 40
women aged 65 to 89 years. Journal of the American Medical Women’s Association,
59(4), 255–261.
Bucher, G.M., Szczerba, P., & Curtin, P.M. (2007). A comprehensive fall prevention program
for assessment, interventions, and referral. Home Healthcare Nurse, 25(3), 174–183.
Bulat, T., Hart-Hughes, S., Ahmed, S., Quigley, P., Palacios, P., Werner, D.C., & Foulis,
P. (2007). Effect of a group-based exercise program on balance in elderly. Clinical
Interventions in Aging, 2(4), 655–660.
Chang, J.T., Morton, S.C., Rubenstein, L.Z., Mojica, W.A., Maglione, M., Suttorp, M.J., . . .
Shekelle, P.G (2004). Interventions for the prevention of falls in older adults: Systematic review and meta-analysis of randomised clinical trials. BMJ (Clinical Research
Ed.), 328(7441).
Clark, S., & Rose, D.J. (2001). Evaluation of dynamic balance among community-dwelling
older adult fallers: A generalizability study of the limits of stability test. Archives of
Physical Medicine and Rehabilitation, 82(4), 468–474.
Clark, S., Rose, D.J., & Fujimoto, K. (1997). Generalizability of the limits of stability test in
the evaluation of dynamic balance among older adults. Archives of Physical Medicine
and Rehabilitation, 78(10), 1078–1084.
Costello, E., & Edelstein, J.A. (2008). Update on falls prevention for community-dwelling
older adults: Review of single and multifactorial intervention programs. Journal of
Rehabilitation Research and Development, 45(8), 1135–1152.
304 Olson, Chen, and Wang
Cyarto, E.V., Brown, W.J., Marshall, A.L., & Trost, S.G. (2008). Comparative effects of
home- and group-based exercise on balance confidence and balance ability in older
adults: Clustered randomized trial. Journal of Gerontology, 54, 272–280.
Davis, J.C., Robertson, M.C., Ashe, M.C., Liu-Ambrose, T., Khan, K.M., & Marra, C.A.
(2010). Does a home-based strength and balance programme in people aged > or = 80
years provide the best value for money to prevent falls? A systematic review of economic analyses of falls prevention interventions. British Journal of Sports Medicine,
44(2), 80–89.
Donat, H., & Ozcan, A. (2007). Comparison of the effectiveness of two programmes on older
adults at risk of falling: Unsupervised home exercise and supervised group exercise.
Clinical Rehabilitation, 21, 273–283.
Duncan, P.W., Weiner, D.K., Chandler, J., & Studenski, S. (1990) Functional reach: A new
clinical measure of balance. Journal of Gerontology, 45(6), M 192–M196.
Faber, M.J., Bosscher, R.J., Chin A Paw, M.J., & van Wieringen, P.C. (2006). Effects of
exercise programs on falls and mobility in frail and pre-frail older adults: A multicenter
randomized controlled trial. Archives of Physical Medicine and Rehabilitation, 87(7),
885–896.
Fletcher, P.C., & Hirdes, J.P. (2004). Restriction in activity associated with fear of falling
among community-based seniors using home care services. Age and Ageing, 33(3),
273–279.
Hauer, K., Pfisterer, M., Schuler, M., Bartsch, P., & Oster, P. (2003). Two years later: A prospective long-term follow-up of a training intervention in geriatric patients with a history
of severe falls. Archives of Physical Medicine and Rehabilitation, 84(10), 1426–1432.
Hess, J.A., & Woollacott, M. (2005). Effects of high-intensity strength training on functional
measures of balance ability in balance impaired older adults. Journal of Manipulative
and Physiological Therapeutics, 28, 582–590.
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(Suppl. 2), ii7–ii11.
Howe, T.E., Rochester, L., Jackson, A., Banks, P.M.H., & Blair, V.A. (2008). Exercise
for improving balance in older people. Cochrane Database of Systematic Reviews,
2007(Iss. 4), CD004963.
Islam, M.M., Nasu, E., Rogers, M.E., Koizumi, D., Rogers, N.L., & Takeshima, N. (2004).
Effects of combined sensory and muscular training on balance in Japanese older adults.
Preventive Medicine, 39, 1148–1155.
Jbabdi, M., Boissy, P., & Hamel, M. (2008). Assessing control of postural stability in
community-living older adults using performance-based limits of stability. BMC
Geriatrics, 8, 8.
Kamide, N., Shiba, Y., & Shibata, H. (2009). Effects on balance, falls, and bone mineral
density of a home-based exercise program without home visits in community-dwelling
elderly women: A randomized controlled trial. Journal of Physiological Anthropology,
28(3), 115–122.
King, M.B., Judge, J.O., & Wolfson, L. (1994). Functional base of support decreases with
age. Journal of Gerontology, 49(6), M258–M263.
Lach, H.W., Reed, A.T., Arfken, C.L., Miller, J.P., Paige, G.D., Birge, S.J., & Peck, W.A.
(1991). Falls in the elderly: Reliability of a classification system. Journal of the American
Geriatrics Society, 39(2), 197–202.
Lajoie, Y., & Gallagher, S.P. (2004). Predicting falls within the elderly community: Comparison of postural sway, reaction time, the Berg Balance Scale and the Activities-Specific
Balance Confidence (ABC) scale for comparing fallers and non-fallers. Archives of
Gerontology and Geriatrics, 38(1), 11–26.
Lin, M.R., Wolf, S.L., Hwang, H.F., Gong, S.Y., & Chen, C.Y. (2007). A randomized, controlled trial of fall prevention programs and quality of life in older fallers. Journal of
the American Geriatrics Society, 55(4), 499–506.
Effect of Home Exercise on Falls 305
Liston, R.A., & Brouwer, B.J. (1996). Reliability and validity of measures obtained from
stroke patients using the Balance Master. Archives of Physical Medicine and Rehabilitation, 77(5), 425–430.
Menz, H.B., Morris, M.E., & Lord, S.R. (2005). Foot and ankle characteristic associated with
impaired balance and function ability in older people. The Journals of Gerontology.
Series A, Biological Sciences and Medical Sciences, 60, 1546–1552.
Muir, S.W., Berg, K., Chesworth, B., Klar, N., & Speechley, M. (2010). Balance impairment
as a risk factor for falls in community-dwelling older adults who are high functioning:
A prospective study. Physical Therapy, 90(3), 338–347.
Murphy, M.A., Olson, S.L., & Protas, E.J. (2003). Screening for falls in community-dwelling
elderly. Journal of Aging and Physical Activity, 11, 66–80.
Norkin, C.C., & White, D.J. (2009). Measurement of joint motion: A guide to goniometry
(4th ed.). Philadelphia, PA: FA Davis.
Rooks, D.S., Kiel, D.P., Parsons, C., & Hayes, W.C. (1997). Self-paced resistance training
and walking exercise in community-dwelling older adults: Effects on neuromotor
performance. The Journals of Gerontology. Series A, Biological Sciences and Medical
Sciences, 52(3), M161–M168.
Schulz, K.F., Altman, D.G., & Moher, D., for the CONSORT Group. (2010). CONSORT
2010 Statement: Updated guidelines for reporting parallel group randomised trials.
Annals of Internal Medicine, 152. Epub 24 March.
Sherrington, C., Whitney, J.C., Lord, S.R., Herbert, R.D., Cumming, R.G., & Close, J.C.
(2008). Effective exercise for the prevention of falls: A systematic review and metaanalysis. Journal of the American Geriatrics Society, 56(12), 2234–2243 .
Shumway-Cook, A., Gruber, W., Baldwin, M., & Liao, S. (1997). The effect of multidimensional exercises on balance, mobility, and fall risk in community-dwelling older adults.
Physical Therapy, 77(1), 46–57.
Stel, V.S., Smit, J.H., Pluijm, S.M.F., & Lips, P. (2004). Consequences of falling in older
men and women and risk factors for health service use and functional decline. Age
and Ageing, 33(1), 58–65.
Stevens, J.A., Mack, K.A., Paulozzi, L.J., & Ballesteros, M.F. (2008). Self-reported falls
and fall-related injuries among persons aged ≥ 65 years—United States, 2006. Journal
of Safety Research, 39(3), 345–349.
Tsang, W.W.N., & Hui-Chan, C.W.Y. (2004). Effect of 4- and 8-wk intensive Tai Chi training on balance control in the elderly. Medicine and Science in Sports and Exercise,
36(4), 648–657.
Vogler, C.M., Sherrington, C., Ogle, S.J., & Lord, S.R. (2009). Reducing risk of falling in
older people discharged from hospital: A randomized controlled trial comparing seated
exercises, weight-bearing exercises, and social visits. Archives of Physical Medicine
and Rehabilitation, 90(8), 1317–1324.
Wallmann, H.W. (2001). Comparison of elderly nonfallers and fallers on performance measures of functional reach, sensory organization, and limits of stability. The Journals of
Gerontology. Series A, Biological Sciences and Medical Sciences, 56(9), M580–M583.
Wang, C.Y., Olson, S.L., & Protas, E. (2002). Test–retest strength reliability: Hand-held
dynamometry in community-dwelling elderly fallers. Archives of Physical Medicine
and Rehabilitation, 83, 811–815.
Wernick-Robinson, M., Krebs, D.E., & Giorgetti, M.M. (1999). Functional reach: Does it
really measure dynamic balance? Archives of Physical Medicine and Rehabilitation,
80(3), 262–269.