You are on page 1of 11

Research Report

Long-Term Effect of Exercise Therapy


and Patient Education on Impairments
I. Svege, PT, PhD, Norwegian
Research Center for Active Reha- and Activity Limitations in People
bilitation, Department of Ortho-
paedics, Oslo University Hospital,
Kirkeveien 166, 0450 Oslo, Nor-
With Hip Osteoarthritis: Secondary
way. Address all correspondence
to Dr Svege at: ida.svege@ Outcome Analysis of a Randomized
ous-hf.no.

L. Fernandes, PT, PhD, Norwegian Clinical Trial


Research Center for Active Reha-
bilitation, Department of Ortho- Ida Svege, Linda Fernandes, Lars Nordsletten, Inger Holm, May Arna Risberg
paedics, Oslo University Hospital,
and Department of Orthopaedic
Surgery and Traumatology, Background. The effect of exercise on specific impairments and activity limitations in
Institute of Clinical Research, Uni- people with hip osteoarthritis (OA) is limited.
versity of Southern Denmark,
Odense, Denmark. Objective. The study objective was to evaluate the long-term effect of exercise therapy and
L. Nordsletten, MD, PhD, Depart- patient education on range of motion (ROM), muscle strength, physical fitness, walking
ment of Orthopaedics, Oslo Uni- capacity, and pain during walking in people with hip OA.
versity Hospital, and Institute of
Clinical Medicine, Faculty of Med- Design. This was a secondary outcome analysis of a randomized clinical trial.
icine, University of Oslo, Oslo,
Norway.
Setting. The setting was a university hospital.
I. Holm, PT, PhD, Department of
Orthopaedics, Oslo University Participants. One hundred nine people with clinically and radiographically evident hip
Hospital, and Institute of Health OA were randomly allocated to receive both exercise therapy and patient education (exercise
and Society, Faculty of Medicine, group) or patient education only (control group).
University of Oslo.

M.A. Risberg, PT, MA, PhD, Nor- Intervention. All participants attended a patient education program consisting of 3 group
wegian Research Center for Active meetings led by 2 physical therapists. Two other physical therapists were responsible for
Rehabilitation, Department of providing the exercise therapy program, consisting of 2 or 3 weekly sessions of strengthening,
Orthopaedics, Oslo University functional, and stretching exercises over 12 weeks. Both interventions were conducted at a
Hospital. sports medicine clinic.
[Svege I, Fernandes L, Nordsletten
L, et al. Long-term effect of exer- Measurements. Outcome measures included ROM, isokinetic muscle strength, predicted
cise therapy and patient educa- maximal oxygen consumption determined with the Astrand bicycle ergometer test, and
tion on impairments and activity distance and pain during the Six-Minute Walk Test (6MWT). Follow-up assessments were
limitations in people with hip conducted 4, 10, and 29 months after enrollment by 5 physical therapists who were unaware
osteoarthritis: secondary outcome of group allocations.
analysis of a randomized clinical
trial. Phys Ther. 2016;96:818
827.] Results. No significant group differences were found for ROM, muscle strength, predicted
maximal oxygen consumption, or distance during the 6MWT over the follow-up period, but
2016 American Physical Therapy the exercise group had less pain during the 6MWT than the control group at 10 months (mean
Association difference8.5 mm; 95% confidence interval16.1, 0.9) and 29 months (mean differ-
Published Ahead of Print: ence9.3 mm; 95% confidence interval18.1, 0.6).
December 17, 2015
Accepted: December 6, 2015 Limitations. Limitations of the study were reduced statistical power and 53% rate of
Submitted: February 11, 2015 adherence to the exercise therapy program.

Conclusions. The previously described effect of exercise on self-reported function was


not reflected by beneficial results for ROM, muscle strength, physical fitness, and walking
capacity, but exercise in addition to patient education resulted in less pain during walking in
the long term.
Post a Rapid Response to
this article at:
ptjournal.apta.org

818 f Physical Therapy Volume 96 Number 6 June 2016


Long-Term Effect of Exercise Therapy in People With Hip Osteoarthritis

O steoarthritis (OA) of the hip is


present in 5% to 11% of the gen-
eral adult population, and preva-
lence increases with age.1 4 Pain, stiff-
ness, and functional impairments are the
people in the group receiving exercise
therapy would demonstrate better
results for the selected outcome
measures.
performed in accordance with a
computer-generated randomization list
(block length10, allocation ratio1:1)
designed by a statistician. Participants
were assigned to treatment groups by a
major presenting complaints, often Method research coordinator, who was not
resulting in various degrees of activity Study Design and Participants involved in enrollment or treatment,
limitations and decreased quality of life.5 This exploratory study represents a sec- through drawing of numbered and
Compared with people who are healthy, ondary outcome analysis of a single-blind sealed envelopes. Allocation was con-
people with mild to moderate hip OA randomized controlled trial evaluating, cealed until the baseline assessment and
have demonstrated reduced hip range of for people with hip OA, the effect of patient education were performed. All
motion (ROM), reduced knee extension both exercise therapy and patient educa- participants signed a written informed
muscle strength, and reduced walking tion or patient education only. The consent form before enrollment.
capacity.6,7 results for the primary outcome mea-
sure, the pain subscale of the Western Interventions
Exercise therapy has been recom- Ontario and McMaster Universities All enrolled participants initially
mended as a first-line treatment modality Osteoarthritis Index (WOMAC), at the attended a patient education program
for lower limb OA8 10 and has been dem- 16-month follow-up assessment were developed for people with hip OA and
onstrated to have a beneficial effect on published previously, along with the provided in the form of a hip school
self-reported hip function in patients results for WOMAC stiffness, WOMAC previously described by Klassbo et al.21
with hip OA.1114 Accordingly, a previ- function, and a health-related quality-of- The patient education program consisted
ous publication based on the same trial life questionnaire (Medical Outcomes of 3 group sessions over a 3-week period,
as the present study reported that people Study 36-Item Short-Form Health Survey led by 2 physical therapists educated in
who had hip OA and received exercise questionnaire [SF-36]).13 No additional the method. Thereafter, participants
therapy in addition to patient education beneficial effect was found for WOMAC were randomized to either an exercise
had better self-reported hip function pain. Furthermore, the short-term results group (n55) or a control group
over a 16-month follow-up period than for biomechanical outcome measures (n54). The exercise therapy program
people who had hip OA and received and the long-term results, evaluated as provided to participants in the exercise
patient education only, whereas no the 6-year risk for total hip replacement, group was specifically designed for peo-
effect on self-reported pain was demon- were recently published.18 Inclusion cri- ple with hip OA and was described in
strated.13 However, few studies have teria were age of 40 to 80 years, hip pain detail elsewhere.17 The program con-
evaluated the treatment effect with out- for 3 months or longer, radiographically sisted of warm-up, strengthening exer-
come measures representing specific verified minimum joint space (in accor- cises, functional exercises, and stretch-
impairments and activity limitations, and dance with Danielsson criteria19: 4 mm ing exercises. Participants were asked to
the results regarding whether exercise for people 70 years old and 3 mm for perform the exercise therapy program 2
therapy can improve muscle strength, people 70 years old), and a Harris Hip or 3 times per week for 12 weeks, and
flexibility, and functional performance in Score20 of 60 to 95 points. In people supervision by a physical therapist was
patients with hip OA are conflict- with bilateral hip OA, the more painful provided at least once weekly. There-
ing.12,14 16 Furthermore, because the hip was defined as the index joint. Exclu- fore, the total intervention period, con-
specific exercise therapy program sion criteria were total hip replacement sisting of both the patient education pro-
applied in the present study was aimed at (THR) in the index joint, knee pain or gram and the exercise therapy program,
improving self-perceived function as knee OA, low back pain, rheumatoid was 15 weeks. Adherence to the exer-
well as impairments and activity limita- arthritis, osteoporosis, cancer, cardiovas- cise therapy program was based on train-
tions,17 the evaluation of its effect on cular disease leading to lack of tolerance ing diaries; the cutoff for satisfactory
these specific secondary outcome mea- of exercise, dysfunction in lower extrem- adherence was at least 20 sessions
sures was of particular interest. Hence, ities, pregnancy, or lack of understand- (80%).
the objective of this exploratory study ing of Norwegian.
was to evaluate the long-term effect, for Participants in the control group
people with hip OA, of a 12-week pro- Participants were recruited from the attended a 2-month follow-up visit at a
gram including both exercise therapy orthopedic outpatient clinics of 2 hospi- physical therapy clinic as part of the
and patient education or patient educa- tals, one sports medicine clinic, and gen- patient education program. They did not
tion only on selected secondary out- eral practitioners and through advertise- have access to the exercise therapy pro-
comes of impairments and activity limi- ment in a newspaper. Screening for gram during the intervention period.
tations (including hip ROM, isokinetic inclusion was conducted at a university
concentric knee and hip muscle hospital by an orthopedic surgeon, who Both interventions were carried out at a
strength, indirectly measured oxygen rated all radiographs, and a physical ther- sports medicine clinic. Two physical
consumption, walking capacity, and pain apist, who rated hip symptoms by using therapists were engaged in supervising
during walking). We hypothesized that the Harris Hip Score. Randomization was participants during the exercise therapy

June 2016 Volume 96 Number 6 Physical Therapy f 819


Long-Term Effect of Exercise Therapy in People With Hip Osteoarthritis

and a strap fixating the pelvis.22 Exten-


sion was measured in the position of the
modified Thomas test.22,23 Aggregated
ROM was calculated by adding the mea-
sured degrees of the 6 hip movements.

Isokinetic concentric muscle strength of


hip and knee flexion and extension was
tested by use of an isokinetic dynamom-
eter (REV9000 [Technogym SpA, Gam-
bettola, Italy] at baseline assessment and
4- and 10-month follow-up assessments;
Biodex 6000 [Biodex Medical Systems
Inc, Shirley, New York] at 29-month
follow-up assessment). Isokinetic con-
centric knee flexion and extension mus-
cle strength was tested with the partici-
pant in a sitting position, with straps
fixating the trunk and thigh (Fig. 1). The
knee joint testing range was set to 5 to 85
degrees of knee flexion. Isokinetic con-
centric hip flexion and extension muscle
strength was tested with the participant
in the supine position, with the opposite
leg extended at the knee and hip, and a
strap fixating the pelvis and opposite
thigh (Fig. 1). The hip joint testing range
was set to 35 to 75 degrees of hip flex-
ion. The test protocol included a
4-repetition warm-up followed by 20 sec-
onds of rest before the 5 test repetitions.
The angular velocity was 60 degrees per
second, and the highest peak torque
value of the 5 test repetitions, measured
in newton-meters, for the index leg was
Figure 1.
used.
Setup for isokinetic muscle strength testing. (Top) Starting position for assessment of knee
extension and flexion muscle strength. (Bottom) Starting position for assessment of hip
extension and flexion muscle strength. Aerobic capacity was assessed by use of
the Astrand test, a submaximal bicycle
ergometer test. We calculated the pre-
dicted maximal oxygen consumption
program. The physical therapists were follow-up assessments at 4 and 10
in accordance with the nomogram
experienced in treating people with months), and at a sports medicine clinic
described by Astrand and Ryhming.24
osteoarthritis; before the start of the (follow-up assessment at 29 months).
The load and sitting position were
study, they underwent training in the Five physical therapists administered all
adjusted for each participant, and the
delivery of the specific exercise outcome assessments after undergoing
bicycle ergometer was calibrated to kilo-
program. training to ensure a standardized ap-
pascals before each test. Heart rate was
proach to the assessments.
corrected for sex and age,25 and pre-
Outcome Measures dicted maximal oxygen consumption
Outcome measures included hip ROM, Hip passive ROM in the index joint was was expressed in liters per minute.
isokinetic concentric muscle strength of measured by use of a half-circle 1-degree-
knee and hip flexion and extension, the increment plastic goniometer with a
In the 6MWT,26 participants walked back
Astrand test, and distance and pain dur- movable arm. Flexion, abduction, and
and forth in a 20-m-long corridor. Partic-
ing the Six-Minute Walk Test (6MWT), as adduction were measured in the supine
ipants were instructed to walk as far as
assessed with a visual analog scale (VAS). position, with a strap fixating the pelvis
possible, without running, over a
Follow-up assessments were conducted and opposite thigh.22 Internal rotation
6-minute period. A stopwatch was used
4, 10, and 29 months after enrollment at and external rotation were measured in
to monitor time, and walking distance
the Norwegian School of Sport Sciences, the prone position, with the hip
was registered in meters. Immediately
Oslo, Norway (baseline assessment and extended, the knee flexed at 90 degrees,

820 f Physical Therapy Volume 96 Number 6 June 2016


Long-Term Effect of Exercise Therapy in People With Hip Osteoarthritis

after the 6MWT, participants were asked Table 1.


to score the hip pain they had experi- Baseline Characteristics of Study Participantsa
enced during walking on a VAS ranging
Exercise Therapy Control
from 0 to 100 mm, with 0 representing Group Group
no pain and 100 representing extreme Characteristic (n55) (n54)
pain.
Age (y), X (SD) 58.4 (10.0) 57.2 (9.8)

Women, n (%) 31 (56.4) 28 (51.9)


Activity level was assessed with the Phys-
ical Activity Scale for the Elderly (PASE). Body mass index (kg/m2), X (SD) 24.6 (3.2) 24.9 (3.8)
The PASE is a self-administered, 7-day Minimum joint space in target jointb (mm), X (SD) 2.1 (1.0) 1.9 (1.1)
recall questionnaire used to assess phys-
Pain duration (mo), X (SD) 47.3 (53.3) 49.5 (50.9)
ical activity.27 The Norwegian version,
which consists of 24 questions yielding a Harris Hip Score,c X (SD) 79.6 (7.7) 76.9 (8.2)
total score of 0 to 315, was used.28 In Bilateral radiographically evident hip OA, n (%) 38 (69.1) 38 (70.4)
addition, participants reported the mean THR of contralateral hip at enrollment, n (%) 4 (7.3) 2 (3.7)
number of times per week they engaged
Hereditary OA or known OA in family, n (%) 17 (30.9) 21 (38.9)
in exercise or physical activity at base-
line and at the 4-month follow-up 12 y of education, n (%) 43 (78.2) 35 (64.8)
assessment. Work status, n (%)

Employed 35 (63.6) 36 (66.7)


Sample Size On sick leave 8 (14.5) 5 (9.3)
Sample size calculations were based on
the pain subscale of the WOMAC (VAS Retired 12 (21.8) 9 (16.7)
version), the predefined primary out- a
OAosteoarthritis, THRtotal hip replacement.
b
come measure. On the basis of a The minimum joint space in the hip joint was assessed in accordance with Danielsson criteria.19
Radiographically evident hip OA was defined as a minimum joint space of 4 mm for people 70
between-group difference of greater years old and 3 mm for people 70 years old.
c
than or equal to 15 points, a standard The Harris Hip Score is a clinician-administered tool for evaluating hip pain, hip function, and hip
deviation of 23, a 5% significance level, range of motion and is scored from worst to best on a scale of 0 100.20
90% power, and a dropout rate of 10%, at
least 54 participants per group were
needed.13 Calculations to evaluate statis-
tical power for the secondary outcome
Data Analysis specific between-group comparisons
Analyses were performed with IBM SPSS (Student t test) to identify at which occa-
measures were not conducted before the
Statistics, version 19.0 (IBM Corp, sions the groups differed.
start of the study.
Armonk, New York). The significance
level was set to .05. All analyses were Because many participants had under-
Masking intention-to-treat analyses and included gone THR during the follow-up period,
The researchers, the 5 outcome asses- all participants enrolled in the study, we conducted a missing-value analysis to
sors, and the physical therapists who regardless of adherence to the patient evaluate the pattern of missing observa-
provided the patient education program education program or the exercise ther- tions. The analysis revealed that missing
remained unaware of group allocations apy program. data were not completely at random.
throughout the trial. The participants Therefore, we conducted a sensitivity
and the 2 physical therapists who pro- analysisa linear mixed model with the
We applied a repeated-measures study
vided the exercise therapy program last-observation-carried-forward tech-
design with one between-group factor
were aware of group allocations. Partic- niqueto account for the poorer preop-
(intervention at 2 levels: patient educa-
ipants in the control group did not have erative results for participants who had
tion, patient education plus exercise
access to the exercise therapy program undergone THR.
therapy) and one within-subject factor
during the intervention period and there-
(occasion at 4 levels: baseline, 4 months,
fore could not cross over to the exercise
therapy group. After completion of the
10 months, 29 months). The interven- Role of the Funding Source
tion occasion interaction term evalu- This study was supported by EXTRA
4-month follow-up assessment, partici-
ated the extent to which intervention funds from the Norwegian Foundation
pants in the control group could visit any
effectiveness differed over the total for Health and Rehabilitation, through
physical therapist they wanted for treat-
follow-up period (linear mixed model; the Norwegian Rheumatism Association,
ment and supervision, and the specific
variance component covariance struc- and by grants from Oslo University
exercise therapy program used in this
ture with occasion and occasion inter- Hospital.
trial was provided if participants
vention as fixed effects and occasion as
requested it.
random-effect intercept and slope). If Results
this interaction was statistically signifi- Between April 2005 and October 2007,
cant, then we performed occasion- 59 women and 50 men with a mean age

June 2016 Volume 96 Number 6 Physical Therapy f 821


Long-Term Effect of Exercise Therapy in People With Hip Osteoarthritis

measures analysis. No between-group


difference was present at the 4-month
follow-up assessment, but the exercise
therapy group reported less pain during
walking at the 10-month follow-up
assessment (P.043) and the 29-month
follow-up assessment (P.043).

Participants in both groups reported that


they were engaged in exercise or physi-
cal activity 3.2 and 3.7 times per week at
the baseline and 4-month assessments,
respectively. No group difference was
found for the total PASE score over the
29-month follow-up period (P.397).

Discussion
The main findings of the present study
were that a 12-week exercise therapy
program given in addition to patient edu-
cation provided no beneficial long-term
effect over patient education only for
ROM, muscle strength, indirectly mea-
sured maximal oxygen uptake, or dis-
tance covered in the 6MWT. Participants
who received both exercise therapy and
patient education reported significantly
less pain during the 6MWT at the 10- and
29-month follow-up assessments than
participants who received patient educa-
tion only.

In line with our findings, Juhakoski et


al14 and Bennell et al16 reported that
exercise therapy had no significant short-
term or long-term effect on hip ROM or
lower limb muscle strength. Further-
more, no effect of exercise has been
Figure 2. demonstrated for various walking tests
Enrollment, randomization, and follow-up assessments of study participants. or stair walking tests.12,14 16 On the
other hand, French et al15 reported that
patients who were given exercise ther-
icant intervention occasion interac- apy achieved better aggregated ROM
of 57.8 years (SD9.9) were enrolled in
than people in a control group. In gen-
the study (Tab. 1). The number of par- tion for pain on the VAS during the
eral, the lack of an additional effect of
ticipants attending each follow-up assess- 6MWT (P.018) (Tab. 2). No significant
exercise therapy on ROM, muscle
ment and the reasons for not attending differences were found for the remaining
strength, and walking ability in the pres-
are shown in Figure 2. Participants in the secondary outcome measures. Overall,
ent study may be partially explained by
exercise group completed a median of the estimated means for each group
the relatively small deficits presented by
20 (first to third quartiles16 24) exer- improved or remained relatively
participants at baseline relative to people
cise sessions over the 12-week period; unchanged over the follow-up period
who were healthy and normative
53% completed at least 20 sessions (Tab. 3). The results of the sensitivity
data.6,7,29 33 Together with participants
(80% adherence). One participant dis- analysis corresponded to the findings of
continued exercise after 3 sessions the main analysis (data not shown). high levels of physical activity at base-
because of increasing pain. No other line, these data indicate that the potential
adverse events occurred. for improvement was somewhat limited.
We conducted occasion-specific
The exercise effect may also have been
between-group comparisons for pain on
compromised by the fact that only 53%
The results of the repeated-measures the VAS during walking on the basis of
of participants were adherent to the
analysis revealed that there was a signif- the significant finding of the repeated-

822 f Physical Therapy Volume 96 Number 6 June 2016


Long-Term Effect of Exercise Therapy in People With Hip Osteoarthritis

Table 2.
Estimated Mean Differences Between Exercise Therapy Group and Control Groupa

Estimated Mean Difference


(95% Confidence Interval)
Between Exercise Therapy Group and Control Groupb at:

Variable 4 mo 10 mo 29 mo Pc

Hip ROM in index joint ()

Aggregated 15.6 (3.2, 34.4) 8.5 (12.1, 29.1) 5.4 (28.8, 18.0) .252

Extension 0.1 (3.1, 3.3) 1.7 (5.2, 1.8) 0.9 (4.9, 3.0) .888

Flexion 4.7 (1.4, 10.9) 2.1 (4.6, 8.8) 0.5 (7.2, 8.1) .072

Internal rotation 3.0 (2.9, 8.9) 4.7 (1.7, 11.1) 0.1 (7.2, 7.4) .543

External rotation 5.4 (0.5, 10.3) 0.1 (5.4, 5.2) 2.9 (9.0, 3.1) .116

Abduction 2.3 (1.0, 5.5) 1.8 (1.7, 5.3) 0.4 (4.4, 3.6) .393

Adduction 1.4 (1.2, 4.0) 1.7 (1.1, 4.6) 1.7 (4.9, 1.6) .468

Isokinetic muscle strength in index leg, peak torque (Nm)

Knee extension 3.3 (12.3, 18.9) 11.6 (28.5, 5.2) 5.7 (25.5, 14.1) .672

Knee flexion 1.8 (11.3, 7.7) 7.1 (17.4, 3.2) 4.1 (16.2, 7.9) .647

Hip extension 4.9 (27.3, 17.5) 23.1 (49.4, 3.2) 6.0 (35.4, 23.4) .450

Hip flexion 0.3 (12.3, 11.8) 12.2 (26.3, 1.9) 1.9 (17.7, 13.9) .472

Predicted VO2max (L/min) 0.1 (0.2, 0.4) 0.1 (0.2, 0.4) 0.2 (0.1, 0.6) .464

Walking capacity and pain

6MWT for distance (m) 0.2 (40.7, 40.3) 1.8 (46.6, 42.9) 8.7 (42.7, 60.0) .801

Pain on VAS during 6MWT (mm) 4.4 (11.3, 2.4) 8.5 (16.1, 0.9) 9.3 (18.1, 0.6) .018d
a
ROMrange of motion, VO2maxmaximal oxygen consumption, 6MWTSix-Minute Walk Test, VASvisual analog scale.
b
Mean for control group subtracted from mean for exercise group.
c
Linear mixed model (variance component model) with time and time group as fixed effects and time as random-effect intercept and slope. P values are
for time group.
d
Significance level: P.05.

exercise program. Additionally, the exer- and has been found to be associated with repetitions, aimed at 70% to 80% of the
cises included in the program may have hip function and disability.6,7,34 There- 1-repetition maximum.36 The lack of an
been ineffective for initiating changes in fore, flexibility and stretching exercises effect on isokinetic muscle strength may
the selected secondary outcome mea- were included in the exercise therapy have been related to inadequate dosage
sures, and the dosage, progression, and program. The lack of a manual compo- and progression of the strengthening
execution of the exercises may have nent during stretching may have resulted exercises. Only 53% of participants were
been inadequate. Furthermore, the PASE in an inadequate stimulus and, therefore, adherent to the exercise therapy pro-
score and the reported engagement in may account for the lack of effect on gram, suggesting that many did not exer-
exercise or physical activity were similar ROM. Hoeksma et al35 reported that man- cise frequently enough, as 2 or more ses-
in the 2 groups and did not increase ual therapy and stretching had a better sions per week seem to be required to
substantially from the baseline assess- effect on ROM than exercise, whereas achieve increased muscle strength.37
ment to the 4-month follow-up assess- French et al15 reported that manual ther- Additionally, resistance and progression
ment. These data may suggest that par- apy in addition to exercise was as effec- may have been insufficient, resulting in
ticipants in the exercise group did not tive as exercise only. an intensity below the required 60% to
increase their overall exercise dosage by 80% of the 1-repetition maximum.37
adding the specific exercise therapy pro- Despite the fact that resistance exercises However, Fukumoto et at38 found that
gram to their usual activities but rather represented a key component of the high-velocity strength training and low-
that they replaced some of their previous exercise therapy program, no between- velocity strength training were equally
weekly activities with 2 weekly sessions group difference in isokinetic concentric beneficial for improving strength in
of exercise therapy. muscle strength of the knee and hip was patients with hip OA. Furthermore, the
found. The exercise therapy program strengthening component of the func-
Impaired ROM has been demonstrated in applied in the present study comprised 7 tional exercises included in the exercise
people with hip OA relative to people resistance exercises,17 and participants therapy program may have been inade-
who are healthy6,7 and normative data29 were instructed to perform 3 sets with 8

June 2016 Volume 96 Number 6 Physical Therapy f 823


Long-Term Effect of Exercise Therapy in People With Hip Osteoarthritis

Table 3.
Outcome Measures at Baseline and Follow-Up Assessments for Exercise Therapy Group and Control Groupa

X (SD) for:

Exercise Therapy Group Control Group

Baseline 4 mo 10 mo 29 mo Baseline 4 mo 10 mo 29 mo
Variable (n55) (n54) (n48) (n40) (n54) (n50) (n39) (n28)

Hip ROM ()

Aggregated 234.6 (43.8) 235.3 (49.4) 231.9 (55.2) 216.8 (44.7) 222.0 (46.6) 219.6 (43.6) 223.4 (52.5) 222.1 (51.9)

Extension 1.5 (7.9) 1.8 (8.3) 1.9 (9.7) 7.3 (6.5) 1.6 (8.6) 1.7 (7.9) 3.6 (8.6) 8.2 (7.1)

Flexion 126.5 (13.8) 123.3 (15.3) 119.8 (16.4) 116.0 (13.4) 119.1 (17.3) 118.5 (15.3) 117.7 (17.6) 115.5 (17.0)

Internal rotation 33.2 (14.4) 35.3 (14.5) 33.5 (15.9) 28.5 (14.0) 32.8 (15.6) 32.3 (15.7) 28.8 (16.0) 28.3 (13.7)

External rotation 27.7 (13.4) 31.9 (12.7) 32.0 (15.5) 24.4 (10.5) 24.4 (11.2) 26.5 (11.7) 32.1 (13.0) 27.3 (9.5)

Abduction 23.9 (8.6) 23.7 (7.5) 24.2 (7.9) 23.3 (8.4) 22.2 (9.0) 21.4 (7.1) 22.4 (9.3) 23.7 (8.6)

Adduction 21.9 (6.1) 20.7 (6.4) 20.5 (7.0) 17.4 (7.8) 21.9 (6.8) 19.3 (5.7) 18.8 (6.3) 19.0 (7.9)

Isokinetic muscle strength, peak torque (Nm)

Knee extension 116.7 (35.9) 123.4 (38.9) 120.2 (38.3) 136.8 (38.8) 120.6 (44.2) 120.1 (43.8) 131.1 (41.9) 142.5 (40.1)

Knee flexion 67.7 (24.4) 69.6 (25.1) 69.2 (24.3) 76.9 (23.0) 68.7 (24.4) 71.4 (25.5) 76.0 (26.4) 81.1 (25.1)

Hip extension 157.0 (58.4) 158.5 (56.4) 152.3 (49.2) 145.6 (51.1) 163.7 (65.9) 163.4 (60.4) 173.6 (64.1) 147.3 (61.5)

Hip flexion 91.4 (31.2) 90.7 (27.7) 84.0 (27.8) 85.0 (25.4) 96.1 (36.5) 90.9 (35.5) 96.8 (32.6) 84.3 (31.8)

Predicted VO2max (L/min) 2.3 (0.1) 2.4 (0.1) 2.3 (0.1) 2.3 (0.1) 2.4 (0.1) 2.5 (0.1) 2.4 (0.1) 2.6 (0.1)

Walking capacity and pain

6MWT for distance (m) 632.6 (103.8) 668.1 (99.5) 688.3 (16.6) 669.6 (88.4) 657.4 (95.0) 668.3 (100.4) 690.2 (101.3) 660.9 (141.1)

Pain on VAS during 23.1 (20.2) 15.1 (16.7) 12.8 (17.6) 14.4 (16.7) 19.4 (14.9) 19.6 (16.0) 21.3 (20.5) 23.7 (19.7)
6MWT (mm)
a
ROMrange of motion, VO2maxmaximal oxygen consumption, 6MWTSix-Minute Walk Test, VASvisual analog scale.

quate to improve maximum isokinetic improve walking capacity in patients clinically important improvement on the
concentric muscle strength. with lower limb OA.38,43 VAS in patients with hip OA was esti-
mated to be 15.3 mm45 but was found
The Astrand bicycle ergometer test and The previously reported results for the to be affected by the degree of symptom
the 6MWT for distance can both be con- primary outcome measure of the ran- severity. In patients with less pain at
sidered to represent levels of physical domized trial on which the present study baseline, comparable to that of partici-
fitness. No between-group difference was based revealed that exercise therapy pants with mild to moderate symptoms
was found for either of them over the had no additional effect on WOMAC pain at the time of enrollment in the present
study period. Walking speed and dis- over patient education only.13 Although study, the threshold was estimated to be
tance have been suggested to be some- the WOMAC can be considered to rep- 7.2 mm.45 These data indicate that the
what decreased in patients with hip OA resent an overall measure of self- mean estimated between-group differ-
relative to people who are healthy.6,7,39 perceived pain, including pain at rest, ences at the 10- and 29-month follow-up
However, the 6MWT results for partici- the assessment of pain during the 6MWT assessments were clinically relevant.
pants in the present study were compa- was included to evaluate activity-related
rable to normative values for adults who pain. Activity-related pain is an important Although no additional effect of exercise
are healthy,3133,40 suggesting that the component of OA,44 but previous studies therapy was revealed for the primary out-
potential for improvement was some- reported conflicting results regarding the come measure (WOMAC pain), partici-
what limited. Furthermore, specific exer- effect of exercise therapy on levels of pants who received exercise therapy
cises targeting walking activities or spe- activity-related pain.12,15,16 In the present and patient education had better self-
cific types of training aimed at increasing study, exercise therapy seemed to pro- reported physical function than those
cardiovascular fitness were not included vide a beneficial long-term effect on pain who received patient education
in the exercise therapy program. Results during walking, with the exercise group only.13,46 Furthermore, we demonstrated
regarding the importance of quadriceps having 8.5 mm and 9.3 mm less pain on a lower 6-year risk for THR surgery in the
muscle strength for walking speed are the VAS during the 6MWT at the 10- and exercise group,46 indicating a favorable
inconsistent,41,42 but it has been sug- 29-month follow-up assessments, respec- effect of exercise therapy on disease pro-
gested that resistance training may tively. The overall threshold for minimal gression. The lack of an effect of exercise

824 f Physical Therapy Volume 96 Number 6 June 2016


Long-Term Effect of Exercise Therapy in People With Hip Osteoarthritis

therapy on ROM, muscle strength, indi- tions in both groups at all follow-up ment.13 Post hoc power calculations
rectly measured maximum oxygen con- assessments and the wide confidence based on estimated smallest detectable
sumption, and the 6MWT for distance intervals for the estimated means indi- changes or minimal important changes
and the results reported in previous pub- cated that large individual differences for aggregated ROM,53 isokinetic knee
lications for this trial appear to be some- were present at all follow-up assessments extension muscle strength,57 the
what inconsistent. In summary, benefi- and for the treatment effect. 6MWT,50,58 and pain on the VAS sug-
cial effects of exercise therapy were gested that the numbers of participants
demonstrated for self-reported function, Self-reported outcome measures are fre- needed in the groups were 66, 60, 42 to
pain during walking, and the need for quently used to evaluate functional 119, and 80, respectively. Hence, the
THR (ie, less need), whereas no addi- impairments and the treatment effect for study was underpowered for detecting
tional effects on self-reported pain, ROM, OA, but according to Wright et al,48 long-term group differences in the sec-
muscle strength, aerobic fitness, or walk- observer-based outcome measures of ondary outcome measures. During the
ing capacity were found. Perceived pain function can provide supplementary follow-up period, 18% of participants in
during activity may influence self- information concerning physical func- the exercise group and 32% of partici-
reported physical function,47 whereas tion. Hence, self-reported and observer- pants in the control group underwent
pain is taken into account to a lesser based outcome measures of function can THR. Therefore, the estimates for the
extent in observer-driven clinical and be considered to be complementary long-term follow-up assessments must be
performance-based tests.48 Joint pain in rather than competing.49 The secondary interpreted cautiously, as the presum-
early OA is typically described as being outcome measures used in the present ably poorer preoperative results for par-
exacerbated by activity and relieved by study were chosen on the basis of the ticipants who underwent THR were not
rest.44 Hence, the favorable effect on reduced ROM, isokinetic knee extension taken into account. This situation may
pain during walking in the present study muscle strength, and walking capacity in have led to an overestimation of the
may be associated with the previously patients with hip OA relative to people mean estimates at the latest follow-up
demonstrated effect on self-reported who are healthy7 and normative assessments in both groups. More impor-
physical function.13 However, the bene- data.29,30 Acceptable reliability has been tantly, the fact that more participants
ficial results were not reflected by demonstrated for the 6MWT50 and isoki- (with poorer preoperative results) in the
changes in ROM, muscle strength, or the netic knee muscle strength assess- control group underwent THR may have
6MWT. It has been suggested that even if ments,51,52 whereas poorer intrarater resulted in an underestimation of the
high-intensity strength training produces and interrater reliability and relatively group differences in favor of the exercise
larger maximal strength adaptations, it large measurement errors have been therapy group. This situation, in addition
does not necessarily improve physical reported for ROM assessments.5355 One to the overall reduced statistical power
function more than low-intensity train- previous study reported acceptable reli- because of the small sample size and a
ing.37 Furthermore, the task specificity of ability for isokinetic hip muscle strength considerable number of missing values
the functional exercises in the exercise assessments, but the protocol in that during the long-term follow-up period,
therapy program may have been decisive study differed from our protocol in that it may have increased the risk of type II
for the improvement of self-perceived had a larger hip joint testing range.56 errors. Furthermore, multiple tests be-
physical function but may have had a Recently, a consensus-derived set of cause of the repeated use of several sec-
limited effect on the specific measures of performance-based tests for assessing ondary outcome measures over a
ROM and muscle strength. physical function in patients with hip or 29-month follow-up periodincreased
knee OA were suggested as suitable out- the risk of type I errors. Hence, the pres-
The findings of the present study are come measures for use in clinical trials of ent study should be interpreted as an
applicable to people with symptomatic patients with lower limb OA: the 30-s exploratory study, and future studies are
and radiographically evident hip OA, chair stand test, the 40-m fast-paced walk needed to verify or reject the study
with mild to moderate symptoms, but test, a stair-climb test, the Timed Up & hypothesis.
without concomitant knee or back pain. Go Test, and the 6MWT.49 Further stud-
ies evaluating the appropriateness of Changes in isokinetic muscle strength
In participants with symptomatic and clinical and performance-based outcome during the follow-up period must be
radiographically evident hip OA, increas- measures in clinical trials and studies interpreted with caution, as a different
ing impairments and a mean decline in evaluating the effect of exercise interven- device was used for isokinetic strength
physical performance over time could tions with various outcome measures in assessment at the 29-month follow-up
have been expected. However, the esti- patients with hip OA are therefore assessment. For the assessment of isoki-
mated mean results suggested that both encouraged. netic muscle strength of the hip, the test-
groups remained relatively stable over ing range was small because of the tech-
time. For a comparison of the treatment The present study had some limitations. nical limitations of the REV6000; the
effect with the time-dependent course of The power calculation was based on the testing range was set at 35 to 75 degrees
the disease, a control group offered no predefined primary outcome of this ran- of hip flexion and, therefore, repre-
treatment should have been included. domized clinical trial, the WOMAC pain sented only about 32% of the mean total
Furthermore, the high standard devia- score at the 16-month follow-up assess- ROM in the sagittal plane. Furthermore,

June 2016 Volume 96 Number 6 Physical Therapy f 825


Long-Term Effect of Exercise Therapy in People With Hip Osteoarthritis

the fact that 5 different testers per- assessments and research coordinator Kristin 9 Zhang W, Nuki G, Moskowitz RW, et al.
formed the outcome assessments Blstad for study management. They thank OARSI recommendations for the manage-
ment of hip and knee osteoarthritis, part
increased the risk of measurement physical therapist and postdoctoral fellow III: changes in evidence following system-
errors. Ingrid Eitzen for valuable discussions during atic cumulative update of research pub-
the preparation of the article. They also lished through January 2009. Osteoarthri-
acknowledge the Norwegian Sports Medi- tis Cartilage. 2010;18:476 499.
In general, masking of patients and care cine Clinic, Oslo, Norway, for supporting the 10 Fernandes L, Hagen KB, Bijlsma JW, et al.
providers and adequate placebo treat- Norwegian Research Center for Active Reha- EULAR recommendations for the non-
ments are difficult to apply in studies bilitation with rehabilitation and test facilities pharmacological core management of hip
evaluating the effect of exercise interven- and knee osteoarthritis. Ann Rheum Dis.
and research staff. 2013;72:11251135.
tions, including the present study. The
The study was approved by the Regional 11 Fransen M, McConnell S, Hernandez-
rationale for providing the patient edu-
Medical Research Ethics Committee and was Molina G, Reichenbach S. Exercise for
cation program to all participants in the carried out in accordance with the Declara- osteoarthritis of the hip. Cochrane Data-
present study, including participants in base Syst Rev. 2014;4:CD007912.
tion of Helsinki.
the control group, was primarily based 12 Tak E, Staats P, Van HA, Hopman-Rock M.
on ethical considerations. The placebo This study was supported by EXTRA funds The effects of an exercise program for
effect of exercise treatment has not been from the Norwegian Foundation for Health older adults with osteoarthritis of the hip.
and Rehabilitation, through the Norwegian J Rheumatol. 2005;32:1106 1113.
fully estimated but has been suggested to
Rheumatism Association, and by grants from 13 Fernandes L, Storheim K, Sandvik L, et al.
be of some significance.16 However, the Oslo University Hospital. Efficacy of patient education and super-
efficacy of patient education interven- vised exercise vs patient education alone
tions has been found to be negligible or Original trial registration: ClinicalTrials.gov in patients with hip osteoarthritis: a single
NCT00319423; additional trial registration blind randomized clinical trial. Osteoar-
small9; therefore, patient education may thritis Cartilage. 2010;18:12371243.
be considered to have served as a pla- for the long-term follow-up study: Clinical-
Trials.gov NCT01063777. 14 Juhakoski R, Tenhonen S, Malmivaara A,
cebo in the present study. et al. A pragmatic randomized controlled
DOI: 10.2522/ptj.20140520 study of the effectiveness and cost conse-
quences of exercise therapy in hip osteo-
The present study was carried out at a arthritis. Clin Rehabil. 2010; 25:370 383.
sports medicine clinic, with a limited
References 15 French HP, Cusack T, Brennan A, et al.
number of physical therapists involved Exercise and manual physiotherapy arthri-
1 Grotle M, Hagen KB, Natvig B, et al. Prev-
in treating participants. These physical alence and burden of osteoarthritis: results tis research trial (EMPART) for osteoarthri-
therapists had a particular interest in the from a population survey in Norway. tis of the hip: a multicenter randomized
J Rheumatol. 2008;35:677 684. controlled trial. Arch Phys Med Rehabil.
field of active rehabilitation and were 2013;94:302314.
trained in providing the applied interven- 2 Quintana JM, Arostegui I, Escobar A, et al.
Prevalence of knee and hip osteoarthritis 16 Bennell KL, Egerton T, Martin J, et al.
tions. Therefore, the findings of the pres- Effect of physical therapy on pain and
and the appropriateness of joint replace-
ent study cannot be generalized to all ment in an older population. Arch Intern function in patients with hip osteoarthri-
physical therapists. However, the avail- Med. 2008;168:1576 1584. tis: a randomized clinical trial. JAMA.
2014;311:19871997.
ability of thorough descriptions of both 3 Dagenais S, Garbedian S, Wai EK. System-
the patient education program21 and the atic review of the prevalence of radio- 17 Fernandes L, Storheim K, Nordsletten L,
graphic primary hip osteoarthritis. Clin Risberg MA. Development of a therapeutic
exercise therapy program17 enables ther- exercise program for patients with osteo-
Orthop Relat Res. 2009;467:623 637.
apists to acquire knowledge about these arthritis of the hip. Phys Ther. 2010;90:
treatment modalities. 4 Pereira D, Peleteiro B, Araujo J, et al. The 592 601.
effect of osteoarthritis definition on prev-
alence and incidence estimates: a system- 18 Eitzen I, Fernandes L, Nordsletten L, Ris-
atic review. Osteoarthritis Cartilage. berg MA. No effects of a 12-week super-
In conclusion, exercise therapy in addi- vised exercise therapy program on gait in
2011;19:1270 1285.
tion to patient education provided no patients with mild to moderate osteoar-
long-term benefits over patient educa- 5 Dieppe PA, Lohmander LS. Pathogenesis thritis: a secondary analysis of a random-
and management of pain in osteoarthritis. ized trial. J Negat Results Biomed. 2015;
tion only for hip ROM, muscle strength, Lancet. 2005;365:965973. 14:5.
aerobic fitness, and walking capacity, but
6 Arokoski MH, Haara M, Helminen HJ, 19 Danielsson LG. Incidence and prognosis of
participants who attended the exercise Arokoski JP. Physical function in men with coxarthrosis. Acta Orthop Scand Suppl.
therapy program reported significantly and without hip osteoarthritis. Arch Phys 1964;66(suppl 66):1114.
less pain during walking at the 10- and Med Rehabil. 2004;85:574 581.
20 Harris WH. Traumatic arthritis of the hip
29-month follow-up assessments. 7 Rydevik K, Fernandes L, Nordsletten L, after dislocation and acetabular fractures:
Risberg MA. Functioning and disability in treatment by mold arthroplastyan end-
patients with hip osteoarthritis with mild result study using a new method of result
to moderate pain. J Orthop Sports Phys evaluation. J Bone Joint Surg Am. 1969;
The authors express their thanks to the peo- Ther. 2010;40:616 624. 51:737755.
ple who participated in this trial. The authors
8 Hochberg MC, Altman RD, April KT, et al. 21 Klassbo M, Larsson G, Harms-Ringdahl K.
thank physical therapists Karin Rydevik, Sig-
American College of Rheumatology 2012 Promising outcome of a hip school for
mund Ruud, Emilie Jul-Larsen Aas, and Marie recommendations for the use of nonphar- patients with hip dysfunction. Arthritis
Moltubakk for their contributions in leading macologic and pharmacologic therapies in Rheum. 2003;49:321327.
the patient education program, supervising osteoarthritis of the hand, hip, and knee.
Arthritis Care Res (Hoboken). 2012;64: 22 Norkin CC, White DJ. Measurement of
participants during the exercise therapy Joint Motion: A Guide to Goniometry.
465 474.
intervention, and conducting the outcome Philadelphia, PA: FA Davis Co; 2003.

826 f Physical Therapy Volume 96 Number 6 June 2016


Long-Term Effect of Exercise Therapy in People With Hip Osteoarthritis

23 Eland DC, Singleton TN, Conaster RR, 38 Fukumoto Y, Tateuchi H, Ikezoe T, et al. 49 Dobson F, Hinman RS, Roos EM, et al.
et al. The iliacus test: new information Effects of high-velocity resistance training OARSI recommended performance-based
for the evaluation of hip extension dys- on muscle function, muscle properties, tests to assess physical function in people
function. J Am Osteopath Assoc. 2002; and physical performance in individuals diagnosed with hip or knee osteoarthritis.
102:130 142. with hip osteoarthritis: a randomized con- Osteoarthritis Cartilage. 2013;21:1042
trolled trial. Clin Rehabil. 2014;28:48 58. 1052.
24 Astrand PO, Ryhming I. A nomogram for
calculation of aerobic capacity (physical 39 Judd DL, Thomas AC, Dayton MR, Stevens- 50 Kennedy DM, Stratford PW, Wessel J, et al.
fitness) from pulse rate during sub- Lapsley JE. Strength and functional deficits Assessing stability and change of four per-
maximal work. J Appl Physiol. 1954;7: in individuals with hip osteoarthritis com- formance measures: a longitudinal study
218 221. pared to healthy, older adults. Disabil evaluating outcome following total hip
Rehabil. 2014;36:307312. and knee arthroplasty. BMC Musculosk-
25 Astrand I. Aerobic work capacity in men
and women with special reference to age. 40 Gibbons WJ, Fruchter N, Sloan S, Levy RD. elet Disord. 2005;6:3.
Acta Physiol Scand Suppl. 1960;49:192. Reference values for a multiple repetition 51 Carpenter MR, Carpenter RL, Peel J, et al.
6-minute walk test in healthy adults older The reliability of isokinetic and isometric
26 Enright PL. The six-minute walk test. than 20 years. J Cardiopulm Rehabil. leg strength measures among individuals
Respir Care. 2003;48:783785. 2001;21:8793. with symptoms of mild osteoarthritis.
27 Washburn RA, Smith KW, Jette AM, Janney J Sports Med Phys Fitness. 2006;46:585
41 Hulens M, Vansant G, Claessens AL, et al.
CA. The Physical Activity Scale for the 589.
Predictors of 6-minute walk test results in
Elderly (PASE): development and evalua-
tion. J Clin Epidemiol. 1993;46:153162. lean, obese and morbidly obese women. 52 Kean CO, Birmingham TB, Garland SJ,
Scand J Med Sci Sports. 2003;13:98 105. et al. Minimal detectable change in quad-
28 Loland NW. Reliability of the Physical riceps strength and voluntary muscle acti-
42 Madsen OR, Bliddal H, Egsmose C, Sylvest
Activity Scale for the Elderly (PASE). Eur vation in patients with knee osteoarthritis.
J Sports Sci. 2002;2:112. J. Isometric and isokinetic quadriceps Arch Phys Med Rehabil. 2010;91:1447
strength in gonarthrosis; inter-relations
29 Soucie JM, Wang C, Forsyth A, et al. Range 1451.
between quadriceps strength, walking
of motion measurements: reference values ability, radiology, subchondral bone den- 53 Holm I, Bolstad B, Lutken T, et al. Reliabil-
and a database for comparison studies. sity and pain. Clin Rheumatol. 1995;14: ity of goniometric measurements and
Haemophilia. 2011;17:500 507. 308 314. visual estimates of hip ROM in patients
30 Danneskiold-Samsoe B, Bartels EM, Bulow with osteoarthrosis. Physiother Res Int.
43 Jan MH, Lin JJ, Liau JJ, et al. Investigation
PM, et al. Isokinetic and isometric muscle 2000;5:241248.
of clinical effects of high- and low-
strength in a healthy population with spe- resistance training for patients with knee 54 Theiler R, Stucki G, Schutz R, et al. Para-
cial reference to age and gender. Acta osteoarthritis: a randomized controlled tri- metric and non-parametric measures in
Physiol (Oxf). 2009;197(suppl 673):1 68. al. Phys Ther. 2008;88:427 436. the assessment of knee and hip osteoar-
31 Casanova C, Celli BR, Barria P, et al. The thritis: interobserver reliability and corre-
44 Hunter DJ, McDougall JJ, Keefe FJ. The
6-min walk distance in healthy subjects: lation with radiology. Osteoarthritis Car-
symptoms of osteoarthritis and the genesis
reference standards from seven countries. tilage. 1996;4:35 42.
of pain. Rheum Dis Clin North Am. 2008;
Eur Respir J. 2011;37:150 156. 34:623 643. 55 Poulsen E, Christensen HW, Penny J,
32 Troosters T, Gosselink R, Decramer M. Six et al. Reproducibility of range of motion
45 Tubach F, Ravaud P, Baron G, et al. Evalu-
minute walking distance in healthy elderly and muscle strength measurements in
ation of clinically relevant changes in
subjects. Eur Respir J. 1999;14:270 274. patients with hip osteoarthritis: an inter-
patient reported outcomes in knee and rater study. BMC Musculoskelet Disord.
33 Camarri B, Eastwood PR, Cecins NM, et al. hip osteoarthritis: the minimal clinically 2012;13:242.
Six minute walk distance in healthy sub- important improvement. Ann Rheum Dis.
jects aged 5575 years. Respir Med. 2006; 2005;64:29 33. 56 Arokoski MH, Arokoski JP, Haara M, et al.
Hip muscle strength and muscle cross sec-
100:658 665. 46 Svege I, Nordsletten L, Fernandes L, Ris- tional area in men with and without hip
34 Steultjens MP, Dekker J, van Baar ME, et al. berg MA. Exercise therapy may postpone osteoarthritis. J Rheumatol. 2002;29:
Range of joint motion and disability in total hip replacement surgery in patients 21852195.
patients with osteoarthritis of the knee or with hip osteoarthritis: a long-term
hip. Rheumatology (Oxford). 2000;39: follow-up of a randomised trial. Ann 57 Lienhard K, Lauermann SP, Schneider D,
955961. Rheum Dis. 2015;74:164 169. et al. Validity and reliability of isometric,
isokinetic and isoinertial modalities for the
35 Hoeksma HL, Dekker J, Ronday HK, et al. 47 Stratford PW, Kennedy DM. Does parallel assessment of quadriceps muscle strength
Comparison of manual therapy and exer- item content on WOMACs pain and func- in patients with total knee arthroplasty. J
cise therapy in osteoarthritis of the hip: a tion subscales limit its ability to detect Electromyogr Kinesiol. 2013;23:1283
randomized clinical trial. Arthritis Rheum. change in functional status? BMC Muscu- 1288.
2004;51:722729. loskelet Disord. 2004;5:17.
58 Jakobsen TL, Kehlet H, Bandholm T. Reli-
36 Kraemer WJ, Adams K, Cafarelli E, et al. 48 Wright AA, Cook CE, Baxter GD, et al. ability of the 6-min walk test after total
American College of Sports Medicine posi- Relationship between the Western knee arthroplasty. Knee Surg Sports Trau-
tion stand: progression models in resis- Ontario and McMaster Universities Osteo- matol Arthrosc. 2013;21:26252628.
tance training for healthy adults. Med Sci arthritis Index Physical Function Subscale
Sports Exerc. 2002;34:364 380. and physical performance measures in
patients with hip osteoarthritis. Arch Phys
37 Steib S, Schoene D, Pfeifer K. Dose- Med Rehabil. 2010;91:1558 1564.
response relationship of resistance train-
ing in older adults: a meta-analysis. Med Sci
Sports Exerc. 2010;42:902914.

June 2016 Volume 96 Number 6 Physical Therapy f 827


Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.

You might also like