Effects of a Late-Phase Exercise Program After Total Hip

Arthroplasty: A Randomized Controlled Trial

Elaine Trudelle-Jackson, PT, PhD, Susan S. Smith, PT, PhD
ABSTRACT. Trudelle-Jackson E, Smith SS. Effects of a
late-phase exercise program after total hip arthroplasty: a
randomized controlled trial. Arch Phys Med Rehabil 2004;85:
1056-62.
Objective: To investigate the effects of a late-phase exercise
program for patients who underwent total hip arthroplasty
(THA) 4 to 12 months earlier.
Design: A single-blind, randomized controlled trial.
Setting: Exercises were performed in subjects homes.
Exercise instruction and measurements taken before and after
the trial were performed in an outpatient research and
treatment center.
Participants: Convenience sample of 34 adults 4 to 12
months post-THA randomly allocated to experimental or
control groups. Twenty-eight subjects completed the study.
Intervention: An 8-week, hip-exercise intervention, during
which the control group received basic isometric and active
range of motion exercises; the experimental group received
strength and postural stability exercises.
Main Outcome Measures: Score on the 12-Item Hip
Questionnaire; fear of falling; hip flexor, extensor, abductor,
and knee extensor muscle torque; and postural stability in
single stance.
Results: There was a statistically significant improvement in
all measures of self-perceived function, muscle strength ( hip
flexors, 24.4%; hip extensors, 47.8%; hip abductors, 41.2 %;
knee extensors, 23.4%), and postural stability (36.8%) in the
experimental group and no significant change in the control
group. Neither group had statistically significant changes in
fear of falling measures.
Conclusions: An exercise program emphasizing weight
bearing and postural stability significantly improved muscle
strength, postural stability, and self-perceived function in
patients 4 to 12 months after THA.
Key Words: Arthroplasty, replacement, Hip; Exercise;
Physical therapy; Rehabilitation.
2004 by the American Congress of Rehabilitation
Medicine and the American Academy of Physical Medicine
and
Rehabilitation

OTAL HIP ARTHROPLASTY (THA) is among the most

widely performed procedures in orthopedic practice in the

From the School of Physical Therapy and Health Promotion & Research Center,
Texas Womans University, Dallas, TX.
Presented in part at the Combined Sections Meeting of the American Physical
Therapy Association, February 2003, Tampa, FL.
Supported in part by the Texas Physical Therapy Education and Research
Foundation, Austin, TX.
No commercial party having a direct financial interest in the results of the research
supporting this article has or will confer a benefit upon the author(s) or upon any
organization with which the author(s) is/are associated.
Correspondence to Elaine Trudelle-Jackson, PT, PhD, TWU Sch of Physical
Therapy, 8194 Walnut Hill Ln, Dallas, TX 75231, e-mail: Ejackson@twu.edu.
Reprints are not available from the author. 0003-9993/04/8507-8571$30.00/0
doi:10.1016/j.apmr.2003.11.022

United States. In 1999, 168,000 THA procedures were

performed in the United States, and this figure is expected to
increase as our population ages and more people live longer,
healthier lives.1 The most common preoperative complaints by
patients who elect to have THA are pain and loss of mobility. It
follows that the most commonly reported outcomes of THA in
the literature relate to pain relief and restoration of mobility.
Outcome studies of pain reduction and range of motion ( ROM
) restoration, usually conducted 3 to 6 months after THA,
indicate an overall satisfaction by patients and physicians. 2,3
However, outcome studies performed at least 1 year post-THA
reveal that impairments and functional limitations persist in the
absence of pain.4-7 Impairments that persist at least 1 year after
THA include decreased muscle strength 5-7 and postural stability
on the side of the replaced hip. 7,8 Functional limitations that
persist after THA include reduced walking speed 4,9 and ability
to climb stairs,10,11 and overall lower ratings on various
assessment tools used to measure function after THA.7 , 11
Investigators6,12,13 have consistently reported significant
improvements in muscle strength at the 6-month follow-up
when compared with presurgical values. However, this
comparison is not very useful because presurgical values are
often reduced as a result of pain and inactivity. A more valid
assessment is to compare muscle strength in the operated hip
with normative values or with muscle strength values on the
uninvolved side. Trudelle-Jackson et al7 recently reported
strength deficits of 10% to 18% in muscles surrounding the
replaced hip when compared with the uninvolved side 1 year
after THA. These findings are consistent with those reported
by Shih et al,6 who found an 11% to 21% deficit in muscle
strength in the involved hip when compared with the
uninvolved hip at 1 year postsurgery. Long et al 5 also found
muscle strength deficits in the muscles surrounding the hip
joint 2 years after THA. Further, Long reported that
deterioration of the Harris hip score and weakness of muscles
during stance were the most consistent findings in subjects
who developed loosening of the hip components. Whether poor
muscle function is a result or cause of component loosening
was not clear. However, because muscle weakness potentially
results in reduced protection of the prosthetic implant fixation,
Long recommended that a supervised exercise program be
continued for a prolonged period of time to improve
postsurgical muscle strength.
In addition to muscle strength deficits, impairment of postural stability after THA has recently been shown in a group of
15 patients who had THA 1 year earlier.7 Postural stability was
measured by using single-leg stance on a force platform.
Stability values on the operated side were compared with
stability values on the uninvolved side. Trudelle-Jackson 7
reported mean differences of 25.9%, 27.2%, and 31.5%
between the involved and uninvolved sides for mediolateral
(ML) stability, anteroposterior (AP) stability, and total stability,
respectively. In an earlier study, Ellison et al8 compared Berg
Balance Scale (BBS) ratings in a group of patients who were 6
months post-THA with a group of age-matched healthy
subjects who had no hip replacement. Ellison reported
1

significantly lower scores on a subset of 4 more difficult tasks

from the BBS in the group of subjects with THA.
Current rehabilitative practices for THA are effective for
reducing pain and improving mobility,3,10 but recent literature
reveals that impairments in muscle strength and postural
stability and reduced function remain at least 1 to 2 years after
surgery.5-8 Most patients receive physical therapy (PT) at the
hospital immediately after surgery for THA, and many receive
home health PT for 2 to 5 weeks after hospital discharge.
Because patients are still recovering from surgery during this
time, their activities are limited. The PT usually consists of
self-care instructions, gait training, and isometric and active
ROM (AROM) exercises.14 Investigators have advocated
continuing the post-THA exercise program 5 or instituting a
new, more challenging exercise program when THA
restrictions have been removed.7 The effectiveness of either of
these exercise programs, however, has not been shown.
The purpose of this study was to investigate the efficacy of a
late-phase rehabilitative program initiated 4 to 12 months after
THA. Efficacy was assessed by comparing muscle strength of
the hip, postural stability, fear of falling, and self-perceived
function before and after an 8-week experimental program of
weight-bearing exercises and a control program of basic
isometric and AROM exercises. We hypothesized that the
group of subjects who performed the experimental exercise
program would significantly improve on all outcome measures
and that the control exercise group would not improve.
METHODS
Participants
Thirty-four adults of any age, race, or sex, who had
undergone THA 4 to 12 months earlier, were recruited from 4
orthopedic practices in the Dallas, TX, area. The orthopedic
practices provided us with lists of patients who had had THA
between 4 and 12 months earlier. Telephone calls were placed
to patients on the list who had not had hip revision surgery.
Patients who were interested in participating after hearing a
brief description of the study were then verbally screened for
the remaining exclusion criteria: (1) pain with weight bearing
on either extremity, (2) low back pain at the time of the study,
(3) diagnosed vestibular problems, (4) central or peripheral
nervous system involvement, and (5) dementia or decreased
cognitive status that would affect ability to follow simple
instructions. Although 34 subjects were initially entered into
the study, only 28 actually completed the study (fig 1). All
participants signed an informed consent approved by the
institutional review board at Texas Womans University.
Design
A single-blind, randomized controlled design was used. A
table of random numbers was used to allocate subjects to either
the experimental or the control exercise group, and subjects
were blind as to their intervention allocation. Both groups
received an exercise intervention, but the control exercise
program was a program similar to that used in the early
rehabilitation of patients with THA, and the experimental
program was a specially designed program to emphasize
controlled weight bearing. The efficacy of neither of the 2
exercise programs had ever been determined when instituted 4
to 12 months postTHA.

Instrumentation
The 12-Item Hip Questionnaire (HQ-12) was used to
selfassess physical function. This questionnaire pertains
specifically to the functional ability of patients who have
undergone THA. Construct validity and test-retest reliability
have been determined in reference to the target population of
our study.15 The HQ-12 is a self-administered paper and pencil
examination in which the patient rates pain or difficulty
associated with performing a series of activities of daily living
(ADLs) on a scale from 1 to 5. The score on the questionnaire
is a total score obtained by summing the ratings of each of the
12 items. The range of possible scores is from 12 to 60, with a
low score indicating a high level of function.
Lower-extremity muscle strength and postural stability were
measured by using a computer-automated system (Human
Performance Measurement [HPM]a) that integrates a battery of
tests to evaluate a broad range of sensorimotor functions. 16 The
BEP-IV and BEP-IIIa are components of the HPM system that
were used in our study to measure postural stability and muscle
strength, respectively. The BEP for Windows software was
used to operate the BEP modules and to record and store data.
The BEP-IV postural stability measurement system uses a
lightweight, portable force platform to measure ML stability,
AP stability, and total stability by tracking changes in the
center of pressure (COP) over time as the subject stands erect
over 1 or 2 feet. The face validity of using forceplate measures
of COP changes as an index of postural steadiness has been
shown.17 To measure postural stability in single stance, the
BEP-IV samples COP movement at a rate of 60 samples per
second over a 10-second time period. The resulting samples are
then averaged over time and normalized for base of support by
calculating the ratio of average movement of the COP to the
size and placement of the stance foot. The resulting normalized
score represents a percentage instability score. This score
was then subtracted from 100% to provide the score of
percentage stability used for data analysis. The total stability
measures were used for data analysis in this study.
The BEP-IIIa was used to measure maximal isometric force
produced by muscles surrounding the hip joint. The device was
factory calibrated with respect to gain and instrumentation and
has a gain accuracy of 1% of full scale. A calibration check
was automatically performed each time the system was
powered up. Muscle torque in Newton meters was calculated
by the BEP for Windows software by using estimated segment
lengths that are based on each subjects height. The standard
error associated with estimated segment lengths that are based
on stature has been shown to be approximately 1.0cm when
compared with measured segment lengths. 18 The intrarater
reliability (intraclass correlation coefficient model 3,1 [ICC3,1])
of the BEP-IIIa has been determined to range from .85 to .97
for the measurement of muscle strength in patients with THA. 7
In our study, a repeated measure of muscle strength for 1
randomly selected muscle (hip flexors, hip extensors, hip
abductors, knee extensors) obtained for each subject was used
to calculate intrarater reliability of the strength measures taken
with the BEP-IIIa. Criterion validity of hand-held
dynamometers for the measurement of muscle strength has
been determined.19 , 20
Measurement Procedures
After being randomly assigned to a treatment group, all
subjects were pretested on measures of physical function, fear
of falling, strength, and postural stability. We then instructed
2

the patients in the appropriate exercises for their group

assignment and posttested at the completion of the 8-week
intervention. One examiner performed all data collection and
exercise instruction. Therefore, the examiner was not blind to
group assignment.

Function was assessed by having participants complete the

HQ-12. After completing the questionnaire, we assessed fear of
falling by asking each subject the following 2 yes or no
questions. Are you afraid of falling? and Since your
surgery, are there any activities that you avoid doing because
you are fearful of falling? The number of yes responses
before the intervention were then compared with the number of
yes responses after the intervention.
Next, we assessed postural stability in single stance with
eyes open by using the BEP-IV force platform. Stability was
measured as subjects attempted to stand steadily on the
involved lower extremity while holding the opposite leg in full
hip extension and 90 of knee flexion. Subjects did not wear
shoes during the testing. Audible beeps from the HPM system
signaled the beginning and end of the 10-second test. Trials
that were terminated before the 10-second maximum, when
subjects lost their balance and had to touch down with a foot or
hand, were calculated over the actual trial time. After a practice
trial, 3 trials were measured with a 5- to 6-second rest between
trials. The BEP software calculated a mean, based on the 2
trials with the most similar percentage stability values.
Muscle strength of the hip flexors, hip extensors, hip
abductors, and knee extensors was measured last by using the
BEP-IIIa force transducer. Muscle force was measured in a
gravity-lessened position by using a make test. To perform a
make test, the examiner held the dynamometer steady while
manually stabilizing the subject with the other hand. Subjects
were instructed to gradually start pushing against the
dynamometer and then to increase the force to their maximum.
Muscle tests were performed in the following order: hip
flexors, hip extensors, hip abductors, and knee extensors. The
specific muscle testing position and force transducer placement
used for testing each muscle group are published elsewhere. 7
As with the tests for postural stability, subjects practiced once
before the test trials. Patients then performed 2 trials, separated
by a 2- to 3-second rest and the mean of the 2 trials was used
for data analysis.
Exercise Interventions
Subjects were instructed in a different set of exercises
depending on group assignment. The exercise protocol for the
control group consisted of 7 basic isometric and AROM
exercises commonly performed during the acute phase of
recovery from THA. These exercises consisted of gluteal
muscle sets, quadriceps sets, hamstring sets, ankle pumps, heel
slides, hip abduction in supine, and internal and external
rotation in supine. Subjects in the experimental group received
a set of 7 weight-bearing exercises, consisting of the following:
sit to stand, unilateral heel raises, partial knee bends, 1-legged
standing balance, knee raises with alternating arm raises
(marching), side and back leg raises in standing, and unilateral
pelvic raising and lowering in standing. No resistance was
added to any of the exercises, and abdominal contraction was

emphasized during all weight-bearing exercises to promote

trunk stability. Good form and slow, controlled movement
were emphasized for both exercise groups.
Subjects in both groups were instructed to perform 15
repetitions of each exercise in their program at home 3 to 4
times a week for 8 weeks. The first follow-up visit was
scheduled within the next 2 weeks to ensure that subjects were
performing the exercises correctly and to progress the
exercises as needed. If tolerated, all exercises were progressed
to 20 repetitions at this time. A second follow-up visit was
scheduled within the next 2 weeks. During the second followup visit, exercises were again checked for proper form and
progressed to 2 sets of 15 repetitions. Subjects were instructed
to maintain these exercise parameters for 2 weeks and then to
gradually increase the number of repetitions in each set to 20
repetitions as tolerated, so that they would be performing 2 sets
of 20 repetitions by the end of their 8-week program. After the
second follow-up visit, subjects continued to exercise on their
own at home for the remainder of the 8-week period. Study
participants recorded their exercise compliance in an exercise
log and turned the log in at the end of the 8 weeks. Subjects
who completed the study received $50 in remuneration.
Data Analysis
Data were analyzed by using the 9.0 version of SPSS b for
Windows. Analysis of variance (ANOVA) was conducted by
using repeated measures of muscle strength in randomly
selected muscles to calculate an ICC (ICC3,2) that provided a
combined estimate of intrarater reliability for the strength
measurements of the 4 muscle groups tested. An ANOVA was
also conducted between measures of postural stability obtained
during each of the 2 tests within the same session, and an
ICC3,2 was calculated to estimate reliability.
The Mann-Whitney U was used to test for differences
between the groups on hip questionnaire scores. Within-group
differences between the pre- and postintervention measures on
the HQ-12 scores for the control group and the experimental
group were analyzed by using the Wilcoxon signed-rank test.
A chi-square test was used to analyze differences between
the groups on the number of yes responses to the 2 fearoffalling questions. Significance of within-group differences in
the number of yes responses before and after the intervention
was analyzed using the McNemar change test.
A multivariate analysis of variance (MANOVA) was used to
test differences between and within groups on the 5 scaled
dependent variables (4 strength variables, 1 postural stability
variable). The MANOVA was a 22 mixed design with
repetition over 1 factor; namely, the pre- and postexercise
scores of strength and postural stability. An level of .05 was
used for all statistical analyses.
Table 1: Description of Subjects by Group and Combined
( all )
Characteristic

Mean SD

Range

Age ( y )
All*
Experimental group
Control group
Body weight ( kg )
All
Experimental group
Control group

59.511.2
59.410.8
59.612.1

36.077.0
36.076.0
37.077.0

81.717.8
83.017.2
80.418.9

46.8113.2
46.8113.2
50.0111.4

Height ( cm )

All
Experimental group
Control group Time
postsurgery ( mo )
All
Experimental group
Control group

169.88.9
169.17.6
170.510.2

153.5188.0
157.6186.7
153.5188.0

7.42.0
7.62.1
7.21.9

4.512.0
4.512.0
4.511.0

*n28.

n14.

Of the 34 subjects, 28 completed the study. The 28 subjects

who completed the study consisted of 15 women and 13 men
with a mean age standard deviation (SD) of 59.511.2 years. A
description of the subjects who completed the study is found in
table 1. Fourteen subjects were in the experimental exercise
group, which consisted of 6 women and 8 men with a mean
age of 59.410.8 years. The 14 subjects in the control group
consisted of 9 women and 5 men with a mean age of 59.612.1
years. Six of the 34 subjects initially recruited dropped from
the study for a variety of reasons (fig 1).
Of the 28 study participants, 24 had a primary diagnosis of
osteoarthritis (OA). The remaining 4 subjects had primary
diagnoses of avascular necrosis (n1), rheumatoid arthritis (n1),
congenital dysplasia (n1), and status post an old hip fracture
that was originally fixed with pins (n1). Although a specific
surgical approach was not an inclusion criterion for this study,
all participants had an anterolateral approach for their THA.
Twelve of the participants had a right THA, 15 had a left THA,
and 1 participant had a bilateral procedure.
All patients received inpatient PT during their hospital stay,
followed by continuation of an exercise program with varying
degrees of supervision. Nineteen participants had home health
PT after their hospital discharge that ranged from 1 week to 2
months. Six subjects attended unsupervised exercise sessions,
and 3 started a water aerobics program. None of the
participants was receiving PT at the time of our study.
Reliability
The intrarater reliability coefficient obtained for the method
of measuring muscle strength was ICC 3,2 equal to .94, and the
intrarater reliability coefficient for measures of postural
stability was ICC3,2 equal to .95. Both are considered excellent.
Self-Perceived Function
Self-perceived function was assessed by using the HQ-12.
The median and maximum-minimum values for pre- and
postexercise scores are shown in table 2. A low score on the
HQ-12 signifies high function. The Wilcoxon signed-rank test
used to analyze differences in scores on the hip questionnaire
before and after the intervention revealed significant
differences for the experimental group (z2.55, P.01), but not
for the
Table 2: Median and Range of Scores on the
HQ-12 Pre- and Postexercise

All (n28)
Experimental (n14)

19.0

1332

17.5

38
12
33

control group (z21.10, P.26). That is, subjects in the

experimental group perceived that their function was improved
after the 8-week intervention, but the control subjects did not
feel that their function had improved.
Fear of Falling

RESULTS

Preexercise

Control (n14)

Postexercise

Median

Range

Median

19.5

1333

16.5

21.0

1533

16.0

Range

12
38
12

Responses to the 2 questions posed on fear of falling are

summarized in table 3. The McNemar change test used to
analyze differences in pre- and postintervention responses to
the questions about fear of falling and avoidance of activities
showed no differences on either question in either the
experimental (P.50, P.62, respectively) or control (P.50, P.12,
respectively) groups. That is, neither of the exercise programs
appeared to affect the subjects fear of falling or avoidance of
activities.
Postural Stability and Muscle Strength
The 22 repeated-measures MANOVA revealed a significant
group by pre-post value interaction (F5,225.72, P.002). A
multivariate related samples t test was performed to analyze
differences between pre- and postintervention measures for
each group (simple main effects of group). The analysis
showed that pre- and postintervention differences were
significant for the experimental group (F5,910.83, P.001) but
not for the control group (F 5,9.14, P.98). That is, the
experimental group performed better on collective measures of
strength and postural stability after the intervention, but the
control group did not. Univariate paired samples t tests were
then performed for each strength and postural stability variable
in the experimental group, to determine which variables
showed significant differences between pre- and
postintervention measures. Significant differences between
pre- and postexercise values were found for each of the muscle
strength and postural stability variables. Postural stability
improved 36.8 %, whereas muscle strength improved 24.4%
for hip flexors, 47.8% for hip extensors, 41.2% for hip
abductors, and 23.4 % for knee extensors. Table 4 shows
percentage change after the exercise intervention for each
group. Figures 2 and 3 show the comparison of pre- and
postexercise measures of postural staTable 3: Responses to Fear-of-Falling (fear) and
Avoidance-ofActivities (avoidance) Questions Pre- and Postexercise
Preexercise

Postexercise

Yes

No

Yes

No

16
7
9

12
7
5

12
5
7

16
9
7

10
7
3

18
7
11

4
3
1

24
11
13

Fear Question
All (n28)
Experimental (n14)
Control (n14)
Avoidance Question
All (n28)
Experimental (n14)
Control (n14)

Table 4: Percentage Change in Muscle Strength and

Postural Stability After an 8-Week, Late-Phase Exercise
Intervention
Preexercise

Postexercise

Experimental group (n14)

Postural stability
(%)
66.1

%
Change

90.4

36.8*

40.5
53.0
53.9
76.6

50.4
78.4
76.1
94.5

24.4*
47.8*
41.2*
23.4*

76.3

77.0

0.9

41.7
52.1
51.6
68.8

44.7
54.0
53.3
69.5

7.2
3.6
3.3
1.0

Muscle strength
( Nm )
Hip flexors
Hip extensors
Hip abductors
Knee extensors
Control group (n14)
Postural stability
(%)
Muscle strength
( Nm )
Hip flexors
Hip extensors
Hip abductors
Knee extensors

Fig 2. Comparison of pre- and postexercise measures of

postural stability in the experimental and control
groups. *Significant difference (P<.05) between before
and after values.

*P.05.

bility and strength for the experimental group and the control
group.
The multivariate independent Hotelling T used to test for
differences between groups on preintervention and
postintervention measures revealed that the groups were
equivalent before the exercise intervention (F 5,22.607, P.12) but
differed significantly after the intervention (F5,224.14, P.01).
DISCUSSION
The primary findings of this study support our hypothesis
that significant improvements in postural stability, muscle
strength, and self-assessed function would be found in a group
of patients with THA after a home program of weight-bearing
exercises. Also, as hypothesized, a similar group of patients,
who performed exercises commonly used in the rehabilitation
of patients with THA, did not show significant improvements
in the same outcome measures. Our study showed that patients
who are a minimum of 4 months post-THA do not benefit from

Fig 3. Comparison of preand

postexercise
measures
of
musclestrength(inNmofto
rque)in the experimental
and
control
groups.
*Significant
difference
(P<.05) between before
and after values.

a group of exercises consisting of basic isometric and AROM,

such as the exercise program commonly used for the
rehabilitation of patients with THA. Although these exercises
have been shown to be effective for the reduction of pain and
loss of mobility during the earlier phase of rehabilitation, 2,3 they
were not effective for improving function, postural stability,
and muscle strength. Therefore, these results do not support the
practice of counseling patients to continue isometric
strengthening and AROM for a year, or indefinitely, as is
sometimes advised. The results do, however, support the
initiation of a brief late-phase exercise program emphasizing
weight bearing and postural stability once the patient is at least
4 months post-THA or as soon as the patient is cleared, by the
physician of all weight-bearing restrictions.
The experimental group exercise protocol consisted of
exercises performed in the standing position, emphasizing
weight bearing on the affected extremity and slow, controlled
movement of extremities. No external resistance was used, but
form was monitored closely during rechecks and corrected as
necessary. Breaks in form were commonly observed, for
example, as subjects attempted slow, controlled hip abduction
of the uninvolved leg in the standing position. Any attempt to
drop the pelvis or to allow the pelvis to rotate back to gain an
advantage by the weak hip abductors on the surgical side was
corrected and pointed out to the subject as substitution or
incorrect form. The performance of all exercises with precise
form is critical in the late phase of rehabilitation, because
studies show that patients who elect to have THA have muscle
weakness around the affected hip before surgery.6,13 Because the
most common diagnosis that leads patients to elect to have
THA is OA,21 it is presumed that the preoperative muscle
weakness is fairly long-standing. Long-term weakness leads to
substitutions and the development of abnormal movement
patterns that must be corrected to effectively gain strength in
the targeted muscle groups.
Our results revealed no changes in fear of falling after
participation in the experimental or control group exercise
programs. Because the experimental group showed significant
improvements in function, muscle strength, and postural
stability after the 8-week intervention, the nonsignificant
improvements in fear of falling were somewhat unexpected
and not as hypothesized. One explanation could be that the 2
questions used to assess fear of falling did not adequately
measure the characteristic of interest, or the questions may not
have been sensitive enough to measure actual change. The 2
questions used in our study about fear of falling and avoidance
of activities because of fear were chosen both for their
simplicity and their documented use in the measurement of
fear in other studies.7,22 However, in a study by Tinetti et al, 23 the
authors argued that use of the 2 questions about fear of falling
may be oversimplified and could lead to inaccuracies in
reporting because people use different standards to make
judgments about fear. Tinetti has since developed a Falls
Efficacy Scale 23 (FES) that is used to rate a persons selfconfidence in avoiding a fall during each of 10 specific ADLs.

The FES uses a continuous scale rather than a dichotomous

scale (being fearful or not) and has the advantage of assessing
the extent to which fear of falling affects willingness to
perform specific activities. Use of the FES may be more
effective in assessing fear of falling as well as in assessing
change in fear of falling as a result of an intervention such as
the exercise program used in our study. Another explanation
for the lack of significant changes in fear of falling could be
that not many of the subjects indicated that they were afraid of
falling before the exercise intervention. In the experimental
group, for example, only 7 of 14 indicated that they were afraid
of falling or avoided activities because they were afraid of
falling. Of those 7 subjects, 2 were no longer fearful after
exercise and 4 no longer avoided activities. Because of the
small numbers to begin with, however, the changes were not
statistically significant. The fact that only 7 of 14 indicated that
they were afraid of falling or avoided activities could be
attributed to the inability of the 2 questions to accurately assess
fear. It is also possible that the subjects in our study truly did
not fear falling before or after THA.
Compliance with the prescribed exercise program was
calculated from the subjects completed exercise logs that were
kept during the 8-week intervention. Compliance rate in the
control group (95.2%) was similar to the calculated compliance
rate for the experimental group (99.3%). The difference (4.1%)
between the groups was considered small and not confounding.
Although the nature of the 2 exercise programs did not
significantly affect compliance in this study, it is likely to
affect compliance over a longer period of time and when
patients are accountable only to themselves rather than to a
researcher.
Patients are likely to be less compliant with isometric exercises
because they are viewed as boring.
CONCLUSIONS
Our study supports the use of an exercise intervention that is
initiated a minimum of 4 months postsurgery when tissues are
healed and weight bearing is no longer restricted or painful. It
is possible that the exercise intervention could be initiated
earlier, but further research is needed to determine this. The
current practice of providing all PT intervention during the first
weeks after surgery, however, has been shown to leave residual
impairments in postural stability and muscle strength and,
more importantly, in overall decreased function. We
determined that continuing the exercises used during the initial
phase of rehabilitation when a patient is 4 to 12 months
postsurgery does not improve postural stability, muscle
strength, or function. Rather, a late-phase home exercise
program, consisting of weightbearing exercises performed
slowly and with good form, should be initiated. Two to 3
follow-up sessions with a physical therapist are recommended
to ensure that exercises are performed precisely and are
progressed as necessary.
References
Popovic JR, Hall MJ. 1999 National Hospital Discharge
Survey: advance data from Vital and Health Statistics. No. 319.
Hyattsville (MD): National Center for Health Statistics; 2001.
Publication No. (PHS) 2001-1250.
2.
Barber TC, Roger DJ, Goodman SB, Schurman DJ. Early
outcome of total hip arthroplasty using the direct lateral vs the
posterior surgical approach. Orthopedics 1996;19:873-5.
3.
Gogia PP, Christensen CM, Schmidt C. Total hip
replacement in patients with osteoarthritis of the hip:
1.

4.
5.
6.
7.
8.

9.
10.
11.
12.

13.

14.
15.
16.
17.
18.
19.
20.

21.
22.
23.

improvements in pain and functional status. Orthopedics

1994;17:145-50.
Brown M, Hislop HJ, Waters RL, Porell D. Walking
efficiency before and after total hip arthroplasty. Phys Ther
1980;60:125963.
Long WT, Dorr LD, Healy B, Perry J. Functional recovery
of noncemented total hip arthroplasty. Clin Orthop
1993;Mar(288): 73-7.
Shih CH, Du YK, Lin YH, Wu CC. Muscular recovery
around the hip joint after total hip arthoplasty. Clin Orthop
1994;May(302): 115-20.
Trudelle-Jackson E, Emerson RH, Smith SS. Outcomes of
total hip arthroplasty: a study of patients one year post-surgery. J
Orthop Sports Phys Ther 2002;32:260-7.
Ellison J, Miller J, Hocate MF, Levitan S, Nandhini M.
Comparison of Berg Balance Scale scores between rehabilitated
patients with total hip arthroplasty and matched healthy subjects.
J Rehabil Outcomes Meas 2000;4:49-54.
Sashika H, Matsuba Y, Watanabe Y. Home program of
physical therapy: effect on disabilities of patients with total hip
arthroplasty. Arch Phys Med Rehabil 1996;77:273-7.
Brander VA, Malhotra S, Jet J, Heinemann AW, Stulberg
SD. Outcome of hip and knee arthroplasty in persons aged 80
years and older. Clin Orthop 1997;Dec(345):67-78.
Wilcock GK. Benefits of total hip replacement to older
patients and the community. Br Med J 1978;2:37-9.
Murray MP, Brewer BJ, Zuege RC. Kinesiological
measurements of functional performance before and after McKeeFarrar total hip replacement. A study of thirty patients with
rheumatoid arthritis, osteoarthritis, or avascular necrosis of the
femoral head. J Bone Joint Surg Am 1972;54:237-55.
Vaz MD, Kramer JF, Rorabeck DH, Bourne RB. Isometric
hip abductor strength following total hip replacement and its
relationship to functional assessments. J Orthop Sports Phys Ther
1993 ; 18:526-31.
Enloe LJ, Shields RK, Smith K, Leo K, Miller B. Total hip
and knee replacement treatment programs: a report using
consensus. J Orthop Sports Phys Ther 1996;23:3-11.
Dawson J, Fitzpatrick R, Carr A, Murray D. Questionnaire
on the perception of patients about total hip replacement. J Bone
Joint Surg Br 1996;78:185-90.
Smith SS, Kondraske GV. Computerized system for
quantitative measurement of sensorimotor aspects of human
performance. Phys Ther 1987;67:1860-6.
Goldie PA, Bach TM, Evans OM. Force platform measures
for evaluating postural control: reliability and validity. Arch Phys
Med Rehabil 1989;70:510-7.
Webb Associates. Anthropometric source book. Vol 1.
Washington (DC): National Aeronautics and Space
Administration; 1978. NASA Rep No. IV6-IV14.
Surburg PR, Suomi R, Poppy WK. Validity and reliability
of a hand-held dynamometer with two populations. J Orthop
Sports Phys Ther 1992;16:229-34.
Trudelle-Jackson E, Jackson AW, Frankowski CM, Long
KM, Meske NB. Interdevice reliability and validity assessment of
the Nicholas hand-held dynamometer. J Orthop Sports Phys Ther
1994;20:302-6.
NIH consensus conference: total hip replacement. NIH
Consensus Development Panel on Total Hip Replacement. JAMA
1995;273: 1950-6.
Tinetti ME, Speechley M, Ginter SF. Risk factors for falls
among elderly persons living in the community. N Engl J Med
1988;319: 1701-7.
Tinetti ME, Richman D, Powell L. Falls efficacy as a
measure of fear of falling. J Gerontol 1990;45:P239-43.

Suppliers
Human Performance Measurement Inc, 2715 Ave E East,
Ste 614 , Arlington, TX 76011.
b.
SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.
a.