JAN ORIGINAL RESEARCH

Effects of a range-of-motion exercise programme

Chien-Ning Tseng1, Cheryl Chia-Hui Chen2, Shiao-Chi Wu3 & Li-Chan Lin4

Accepted for publication 10 July 2006

1
Chien-Ning Tseng MSN RN T S E N G C . N . , C H E N C . C . H . , W U S . C . & L I N L . C . ( 2 0 0 7 ) Effects of a range-of-
Lecturer motion exercise programme. Journal of Advanced Nursing 57(2), 181–191
Cardinal Tien College of Nursing, doi: 10.1111/j.1365-2648.2006.04078.x
Taipei, Taiwan

2 Abstract
Cheryl Chia-Hui Chen DNSc GNP
Assistant Professor Title. Effects of a range-of-motion exercise programme
School of Nursing, Aim. This paper reports an evaluation of a range-of-motion exercise programme
National Taiwan University, aimed at improving joint flexibility, activity function, perception of pain, and
Taipei, Taiwan depressive symptoms in a sample of stroke survivors in long-term care facilities.
Background. The benefits of physical rehabilitation for stroke survivors have been
3
Shiao-Chi Wu PhD well established. There is, however, little empirical data on the effects of a simple
Professor
nurse-led range-of-motion exercise programme in improving function for these
Institute of Health and Welfare Policy,
people.
National Yang-Ming University,
Taipei, Taiwan Method. A randomized controlled trial was conducted in 1999 with 59 bedridden
older stroke survivors in residential care. Participants were randomly assigned to
4
Li-Chan Lin PhD RN usual care or one of two intervention groups. The 4-week, twice-per-day, 6 days-per-
Professor week range-of-motion exercise protocols were similar in both intervention groups,
Institute of Clinical Nursing, and consisted of full range-of-motion exercises of the upper and lower extremities.
National Yang-Ming University, To test the effect of different degrees of staff involvement, in intervention group I, a
Taipei, Taiwan
Registered Nurse was present to supervise participants performing the exercises,
while intervention group II involved a Registered Nurse physically assisting partic-
Correspondence to L.-C. Lin:
e-mail: lichan@ym.edu.tw ipants to achieve maximum range-of-motion within or beyond their present abilities.
Results. Both intervention groups had statistically significant improvement in joint
angles, activity function, perception of pain and depressive symptoms compared with
the usual care group (P < 0Æ05). Post hoc comparison revealed that the joint angles
in intervention group II were statistically significantly wider than in both the other
groups (P < 0Æ01).
Conclusions. A simple nurse-led range-of-motion exercise programme can generate
positive effects in enhancing physical and psychological function of bedridden older
people with stroke. Further studies are needed to investigate the long-term effects of
the programme in maximizing function, reducing care utilization and enhancing
quality of life for this population.

Keywords: depression, nursing, pain, randomized controlled trial, rehabilitation,

residential care, stroke

2000, Kwakkel et al. 2004, American Heart Association

Introduction
2005). In Taiwan, stroke ranks third as a cause of mortality
Stroke is a leading cause of death and one of the most among people aged 65 years and older (Department of
disabling illnesses in developed countries (Miller & Easton Health, the Execution Yuan 2005). Approximately 18Æ9%

 2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd 181
C.-N. Tseng et al.
of stroke survivors in Taiwan have moderate levels of
disability, and 26Æ4% present with the most severe motor
disability (Lin et al. 2001). Without assistance, only 34Æ4% of
poststroke patients can walk independently, while the rest
eventually become bedridden (Lin et al. 1999). Such prevent-
able disabilities place an unjustified burden on stroke
survivors and force them to enter long-term care (LTC)
facilities prematurely. Stroke survivors who are confined to
bed in LTC facilities represent one of the most vulnerable and
neglected populations, given that most researchers have
studied and promoted exercise regimes for stroke survivors
mainly in acute care settings. To bridge the gap, the aim of
this study was to test the effects of a simple, 4-week, nurse-led
range-of-motion exercise programme with a sample of
bedridden older stroke survivors in LTC facilities, using a
randomized controlled trial design.
Background
Lack of rehabilitation services for older people with stroke is
not only a unique problem in Taiwan but also a universal
challenge worldwide. Staff shortage, lack of awareness and
reimbursement issues have all contributed to this challenge.
Lack of rehabilitation services could lead to a further decline
in functional status of older residents in LTC (Spector &
Takada 1991). Data also have showed that 23% of patients
with stroke experienced joint tightness and soreness due to
contracture (Pinedo & de la Villa 2001). Such discomfort or
pain may further stop patients from doing basic activities of
daily living (ADL). With prolonged immobile status of joints,
function lessens and emotional stress such as depressive
symptoms soars, which can create a vicious cycle.
Specifically in Taiwan, nearly 47Æ8–86Æ7% of older patients
with stroke are admitted to LTC facilities in their poststroke
time (Yeh et al. 1999, Lin et al. 2000). Without systematic
rehabilitation to restore these patients’ functionality, the
majority are doomed to become bedridden. Substantial
poststroke rehabilitation efforts should continue at LTC
facilities in order to minimize dependency and maximize
functionality. While 87Æ3% of patients with stroke at LTC
facilities require physiotherapy, only 21% receive any. The
two major reasons are that physiotherapy at LTC facilities is
not reimbursed by the National Health Insurance and only
33Æ3% of LTC facilities have physiotherapists on staff (Xiong
& Lin 1999).
The beneficial effects of physical rehabilitation on patients
with stroke have been well established (Taylor et al. 1996,
Akay & Marsh 2001). The effect of range-of-motion
exercise, however, has been less studied. Range of motion
(ROM) is the term used to describe the amount of movement
182  2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
a person has at each joint and every joint in the body has a
‘normal’ ROM. The ROM exercise is the type of movement
or activity that aims to preserve flexibility and mobility of the
joints on which it is performed (Kisner & Colby 1996). Using
the keywords including ROM, ROM exercise and ROM
dance, 19 articles were retrieved from MEDLINE (1966–
2006) and CINAHL (1982–2006) databanks. These 19
articles reported the effects of ROM exercise on various
outcomes and within diverse populations. Specifically, the
importance of ROM exercise in maintaining joint flexibility
(Clough & Maurin 1983, Van Deusen & Harlowe 1987a,b,
Van Deusen & Harlowe 1988, Shaw et al. 1989, Goldsmith
et al. 2002, Lynch et al. 2005), maximizing ADL function
(Valentine-Garzon et al. 1992), reducing depressive symp-
toms and anxiety (Diego et al. 2002), enhancing self-esteem
(Valentine-Garzon et al. 1992), and improving body image
and enjoyment (Van Deusen & Harlowe 1987a,b, Van
Deusen & Harlowe 1988) has been studied in different
populations.
Most of these studies showed that ROM exercise can be
safely instituted as a treatment plan for various patients and
positive effects in physical and psychological health can be
expected. However, a device or more sophisticated phy-
siotherapy were often provided on top of a simple ROM in
these previous studies. For example, a device was applied
when patients completed ROM exercise in Clough’s study
(Clough & Maurin 1983) and a standardized 3Æ5 hours daily
poststroke therapy was provided along with a simple ROM
exercise in Lynch’s study (Lynch et al. 2005). It was not clear,
therefore, if a simple, nurse-led, low-tech ROM exercise
programme would show similar positive effects. In other
words, the independent effect of a simple nurse-led ROM
exercise on patient outcome was unknown. Additionally,
previous studies also suffered from exploring only the angle
changes on extension of upper extremities, instead of
including lower extremities that is often essential to partic-
ipants’ ADL status and mobility. In short, there was a lack of
agreement on the effects of a simple ROM exercise on joint
flexibility, ADL function, perception of pain, or depressive
symptoms in institutionalized bedridden older people with
stroke, particularly regarding questions of frequency, dur-
ation, and dosing issues of the exercise programme, apart
from the staffing issues.
There is evidence suggesting that nurses are redefining and
expanding their role in rehabilitation care and are making an
important contribution (Long et al. 2002). Nurses spend the
most time with patients and have sufficient ability to assist
patients conducting ROM exercise if training is given. Much
of the impetus for studying joint flexibility, ADL function
status, perception of pain and depressive symptoms as
JAN: ORIGINAL RESEARCH
outcome variables emerged from clinical observations that
indicated the high prevalence of these factors and their
disabling effects on poststroke institutional bedridden older
people. With compromised ADL function in their poststroke
time, experiencing pain, depressive symptoms and reducing
joint flexibility will likely disable older people further and
amplify the physical and emotional stress to these patients
(Prencipe et al. 1997, Toso et al. 2004). In many respects,
older people’s ability to function is among the most import-
ant measures of the overall impact of illness and is predictive
of clinical patient outcomes (Yeh & Lo 2003, Chen et al.
2005).
The study
Aim
The aim of the study was to evaluate the effect of a range-
of-motion exercise programme aimed at improving joint
flexibility, activity function, perception of pain and depressive
symptoms in a sample of stroke survivors in LTC facilities.
Design
A randomized controlled trial with a usual care group and
two intervention groups was used. Data were collected in
1999, and a power calculation was not done (see Study
limitations section below).
Randomization
Given the potential impact of age, gender, time since stroke,
baseline function FIMTM ADL scores (see below), and
Brunnstrom stage of extremities (Akay & Marsh 2001) (see
below), a balanced design was adopted to ensure that the
composition of the groups being compared had proportional
representation on these variables. Participants with similar
characteristics were randomly assigned to either the usual
care group or one of the two intervention groups by the toss
of a coin. The coin was tossed to assign the participants to the
group with least number of participants. If no (tails) was
indicated, then the coin was flipped again to see which of the
other two groups the participant should enter.
Participants
Potential participants were screened and recruited from 31
private LTC facilities located throughout Taipei City. ‘Bed-
ridden’ was defined as a person who could not move
independently out of bed. Participants were recruited if they:
 2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
Effects of a ROM exercise
(a) had suffered from hemiplegia for 6 months or more since
stroke; (b) made five errors in the Short Portable Mental
Status Questionnaire (SPMSQ, Pfeiffer 1975); (c) were able
to communicate orally; (d) were aged 55 years or older; (e)
had no limb amputation; (f) had no open wound, thrombo-
phlebitis, or malignancy of the limbs. Of the total of 219
residents screened, 83 met the inclusion criteria, while 65
(78Æ31%) gave consent to participate in the study. Among
those who did not meet the inclusion criteria, 19 residents
were completely independent and free from hemiplegia, with
another 117 unable to communicate because of severe
aphasia or impaired mental status (defined as making more
than five errors in SPMSQ).
Of the 65 participants who gave consent to participate, six
did not complete the study. Five were from the usual care
groups, with one moving out of the facility and the other four
being hospitalized. In intervention group I, one was hospit-
alized. Finally, 59 participants completed the study, with 17
in the usual care group, 21 in intervention group I and 21 in
intervention group II.
Intervention
Intervention group I involved a Registered Nurse (C-N.T.)
supervising participants to perform and complete the ROM
protocol by themselves. Participants in the intervention
group II carried out the same ROM protocol with the
nurse’s presence to help them physically in achieving
maximum ROM within or beyond their present ability
(nurse placed one hand above the participant’s joint, with
another hand below the joint to secure the movement
safely). Participants in the usual care group did not receive
any extra ROM exercise over and above the facility’s
routine. In fact, none of the studied facilities had any ROM
exercise routine in place. No attention control was provided
in this study design. To ensure that participants had equal
access to the additional care, the ROM intervention was
provided for participants in the usual care group at the end
of the data collection.
Participants in both intervention groups completed the
ROM exercise protocol, five times per joint, twice per day
and 6 days per week for 4 weeks with each session lasting
approximately 10–20 minutes. Passive ROM exercise was
done slowly on one joint at a time, with participants using a
stronger limb to assist a weaker limb to perform the exercises.
Full ROM movements in six joints (shoulder, elbow, wrist,
hip, knee and ankle) were included in the protocol, including
flexion, extension, adduction, abduction, internal and exter-
nal rotations, and dorsal and plantar flexions. The ROM
interventions were all scheduled to avoid interfering with
183
C.-N. Tseng et al.
bath, meal and bed times, and all were completed by the same
nurse (C-N.T.) to ensure reliability.
The two interventions groups were used to test the effect of
staff intensity. Care was given to ensure that participants
were not stretched beyond their limits.
Data collection
All eligible participants were approached by the nurse to
explain the nature and purpose of the study and to invite
them to participate. After obtaining informed consent, they
were randomly assigned to the usual care group or one of the
two intervention groups, as described above. Baseline meas-
ures were taken 1 day prior to the intervention taking place.
Postintervention measures were taken 1 day after the 4-week
intervention was completed. All data collection was pre-
scheduled during off-treatment time. To ensure the objectiv-
ity of the outcome measures, a physiotherapist who was blind
to the randomization obtained all measures on the joint
angles, Brunnstrom stage, and perception of pain for all three
groups.
Baseline measures
The demographic and baseline function data collected were:
age, gender, marital status, education level, time since stroke,
time since hemiplegia, time of residence in LTC facilities,
dominant side, hemiplegic side, muscle tone, SPMSQ and
Brunnstrom stage. These measures were all taken 1 day
before starting the ROM intervention.
The SPMSQ, a simple and widely used scale developed
by Pfeiffer (1975), comprises 10 questions dealing with
orientation, personal history, remote memory and calcula-
tions. The SPMSQ, with demonstrated high internal con-
sistency (alpha coefficient ¼ 0Æ92) and test–retest reliability
(coefficient ¼ 0Æ82), is easy to administer and appears not
to be affected by ethnicity and educational level, making it
suitable to use with this study population (Chen et al.
2004).
Brunnstrom stage is a well-used quantitative method to
classify poststroke motor recovery into six stages. Stage I is
the flaccid stage, stage II is the spastic stage, and stage III is
the synergic stage. Further into stage IV, there is a decrease in
spasticity and the person is capable of performing gross
movement combination with reduced synergy. In stage V,
spasticity is further decreased and the person has the ability
to perform more complex combinations of movement. Lastly,
in stage VI, spasticity disappears, with individual joint
movement becoming possible and coordination approaching
normal (Akay & Marsh 2001).
184  2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
Outcome measures
The outcome measurements were functional independence
(FIMTM-ADL subscale), joint angle, self-reported pain (see
below) and the Chinese version of Geriatric Depression Scale –
Short Form (GDS-15). The selection of these instruments was
undertaken by an expert panel review. Eight field experts in
nursing and physiotherapy with over five years of rehabilit-
ation experience were asked to review the applicability of these
instruments. They used a 4-point Likert scale to rate applica-
bility of the instruments and each item within particular
instruments. Items receiving a score of three or four were
retained, while those with scores of two or less were modified
according to the experts’ suggestions. Finally, all items within
the selected instruments were retained to ensure standardiza-
tion of the measures. A value of 0Æ8 on the content validity
index was obtained, which indicates appropriate use of
measures (Lynn 1986).
Functional status measures
The ADL subscale of FIMTM was used to assess the partici-
pant’s need for assistance in accomplishing the ADL (Keith
et al. 1987). The ADL subscale contains 13 items. Items are
rated on a 7-point Likert scale ranging from 1 (dependent) to
7 (independent). Scores thus range from 13 to 91, with a
higher score indicating more independence. Permission to
translate and use the instrument was granted by the Uniform
Data System for Medical Rehabilitation, USA. A two-stage
translation process was used. Some items were modified to
take into account cultural differences, such as replacing the
use of knife and fork with chopsticks, and deleting items
about spreading butter on bread or cutting steak into pieces.
The Chinese version was then assessed for equivalence and
content validity in comparison with the English version.
Test–retest reliability over a 2-week interval was examined.
Cronbach’s a and test–retest reliability were reported to be
0Æ97 and 0Æ99, respectively. Participants needed to demon-
strate ability to perform these tests in order to be retained in
the study, and no proxy information was used in functional
assessment.
Joint angle and pain measures
A physiotherapist, who was blinded to the study design, took
all 17 joint angle measures in six joints: shoulder flexion,
extension, adduction and abduction; elbow flexion and
extension; wrist flexion and extension; hip flexion, extension
and adduction; knee flexion and extension; ankle dorsal
flexion, plantar flexion, inversion and eversion. At the same
time the therapist assessed self-perception of pain using three
ratings. If there was no oral expression of pain, this was
JAN: ORIGINAL RESEARCH
recorded as ‘painless’ or 0. Conversely, ‘painful’ was recor-
ded as 2 and ‘a little pain’ was recorded as 1. The scores for
the 17 joint angles were summed for analysis purposes.
Depressive symptoms measure
The Chinese GDS-15 was used to measure depressive symp-
toms (Yesavage et al. 1983). Cronbach’s a has been reported
around 0Æ77 (Lin et al. 1998). Dichotomous questions were
used to assess the presence of depressive symptoms. The total
scores ranged from 0 to 15, with higher scores indicating
more depressive symptoms.
Ethical considerations
The study was approved by the institutional review board of
National Yang-Ming University and the 31 facilities involved
all agreed to participate. Written consent was obtained from
every patient enrolled or their legal guardian. Mechanisms
were in place to ensure the safety of patients while conducting
ROM exercises; for example, bed rails were used and
attention was paid to ensure that the nurse was standing on
the weak side of patients and positioned to prevent them
from falling or injury. Additionally, to avoid unjustifiable
pain or suffering, the ROM exercise was immediately
stopped if participants indicated pain, verbally or non-
verbally, or indicated that they did not wish to continue.
Data analysis
Data were analysed using SPSS, version 11 (SPSS, Inc.,
Chicago, IL, USA), and were reviewed and double entered to
ensure accuracy. Descriptive and bivariate statistics, inclu-
ding chi-squared (for categorical variables) and ANOVA (for
continuous variables), were computed to describe the sample
and to compare the outcomes among the three groups. The
change data between pre- and post-intervention measures
were used as continuous variables. Scheffe post hoc compar-
ison was performed to examine the true difference between
groups. The statistical significance level was set at P < 0Æ05.
Results
Participant characteristics at the baseline were examined by
chi-squared and ANOVA to ensure the homogeneity. As shown
in Tables 1 and 2, there was no statistical significance in
demographics or Brunnstrom staging among the groups. A
total of 59 participants completed the study, with the
majority being male (n ¼ 45, 76Æ3%). Ages ranged from 55
to 88 years (mean ¼ 75Æ05; SD ¼ 7Æ77). More than half had
some elementary education (n ¼ 31; 52Æ5%) with another
 2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
Effects of a ROM exercise
33Æ9% (n ¼ 20) being illiterate. All were taking medications
(100Æ0%), ranging from 1 to 9 medications per day with a
median of 5. Eighteen participants (30Æ5%) were taking five
or more medications per day. The SPMSQ scores ranged
from 4 to 10 with a mean of 7Æ19 (SD ¼ 1Æ99), indicative of
mild impaired cognition. Three participants were allowed to
make six errors on the SPMSQ given that these participants
had only a grade school education and were allowed one
more error, according to the SPMSQ education criteria
(Pfeiffer 1975). For 64Æ4% of the participants (n ¼ 38), this
was their first stroke. Left hemiplegia (n ¼ 31) and right
hemiplegia (n ¼ 28) were evenly divided among the groups.
The time since the stroke varied from 10 to 132 months with
a mean of 59Æ81 months. The upper and lower extremities
differed slightly on Brunnstrom stages. Twenty-six partici-
pants (44Æ1%) were at stage 2 (spastic stage) for the upper
extremities, while 24 (40Æ7%) had a stage 3 (synergic stage)
for the lower extremities.
After 4 weeks of the ROM exercise programme, the mean
difference in joint angles among the three groups was
statistically significant (P < 0Æ001). Participants in the usual
care group had a decrease in joint angles, on average 5Æ83
in upper extremities and 3Æ88 in lower extremities. In
contrast, participants in both interventions groups had an
increase in joint angles. The improvements were on average
þ5Æ42 (upper extremity) and þ2Æ14 (lower extremities) for
the intervention group I. An improvement of þ12Æ8 in upper
extremities and þ7Æ92 in lower extremities was found in
intervention group II (see Table 3 for changes on each specific
joint). Scheffe post hoc comparison revealed that both
intervention groups had statistically significant improvements
in joint angles compared with the usual care group. As
illustrated in Table 3, the differences between the two
intervention groups were all statistically significant, with
the best scores (wider angle degree) in the intervention group
II, the only exceptions being elbow extension, knee extension
and ankle dorsal flexion.
In terms of ADL function, the mean FIMTM ADL scores in
three groups did not show obvious changes after the study.
However, examining the data further, it was found that four
participants in the usual care group (23Æ53%) showed a
retrogression in ADL function (lower FIM scores), while two
participants in the intervention group I (9Æ52%) and six in the
intervention group II (28Æ57%) demonstrated a progression
in ADL function by scoring higher on the FIMTM-ADL scales.
Statistically significant differences were found among the
three groups on the pre- and post-intervention FIMTM ADL
scores. A post hoc comparison using the Scheffe test revealed
that FIMTM ADL scores from both intervention groups were
statistically significantly different from those in the usual care
185
C.-N. Tseng et al.

Table 1 Demographic characteristics (n ¼ 59)

Group

Usual care Intervention I Intervention II

Characteristics (n ¼ 17) (n ¼ 21) (n ¼ 21) Total v2 P value

Sexual
Male 13 (22Æ0) 16 (27Æ1) 16 (27Æ1) 45 (76Æ3) 0Æ00 1Æ00
Female 4 (6Æ8) 5 (8Æ5) 5 (8Æ5) 14 (23Æ7)
Age, years
<65 2 (3Æ4) 2 (3Æ4) 3 (6Æ8) 7 (11Æ9) 1Æ54 0Æ95
65–74 7 (11Æ9) 6 (10Æ2) 7 (11Æ9) 20 (33Æ9)
75–84 5 (8Æ5) 10 (16Æ9) 8 (13Æ1) 23 (39Æ0)
>85 3 (5Æ1) 3 (5Æ1) 3 (4Æ9) 9 (15Æ3)
Marital status
Single 8 (13Æ6) 9 (15Æ3) 7 (11Æ9) 24 (40Æ7) 6Æ83 0Æ34
Separate/devoice 2 (3Æ4) 0 (0Æ0) 5 (8Æ5) 7 (11Æ9)
Widower/widow 7 (11Æ9) 11 (18Æ6) 8 (13Æ6) 26 (44Æ1)
Married 1 (1Æ7) 1 (1Æ7) 1 (1Æ7) 2 (3Æ4)
Education
Illiterate 3 (5Æ1) 8 (13Æ1) 9 (15Æ3) 20 (33Æ9) 4Æ69 0Æ33
Elementary 10 (16Æ9) 10 (16Æ9) 11 (18Æ6) 31 (52Æ5)
High school 4 (6Æ8) 3 (5Æ1) 1 (1Æ7) 8 (13Æ6)
Medication
No 0 (0Æ0) 0 (0Æ0) 0 (0Æ0) 0 (0Æ0)
Yes 17 (28Æ8) 21 (35Æ6) 21 (35Æ6) 61 (100Æ0)
Times of stroke
1 11 (18Æ6) 13 (22Æ0) 14 (23Æ7) 38 (64Æ4) 2Æ02 0Æ92
2 4 (6Æ8) 4 (6Æ8) 4 (6Æ8) 12 (20Æ3)
3 2 (3Æ4) 3 (5Æ1) 3 (5Æ1) 8 (13Æ6)
4 0 (0Æ0) 1 (1Æ7) 0 (0Æ0) 1 (1Æ7)
Dominant side
Right 15 (25Æ4) 20 (33Æ9) 21 (35Æ6) 56 (94Æ9) 2Æ70 0Æ26
Left 1 (1Æ7) 1 (1Æ7) 1 (1Æ7) 3 (5Æ1)
Hemiplegia
Right 8 (13Æ6) 9 (15Æ3) 11 (18Æ6) 28 (47Æ5) 0Æ38 0Æ83
Left 9 (15Æ3) 12 (20Æ3) 10 (16Æ9) 31 (52Æ5)
Muscle tone
Upper extremity
1 2 (3Æ4) 2 (3Æ4) 1 (1Æ7) 5 (8Æ5) 9Æ95 0Æ27
1þ 5 (8Æ5) 5 (8Æ5) 5 (8Æ5) 15 (25Æ4)
2 7 (11Æ9) 9 (15Æ3) 11 (18Æ6) 27 (45Æ8)
3 2 (3Æ4) 3 (5Æ1) 2 (3Æ4) 7 (11Æ9)
4 1 (1Æ6) 2 (3Æ3) 2 (3Æ3) 5 (8Æ2)
Lower extremity
1 3 (5Æ1) 4 (6Æ8) 4 (6Æ8) 11 (18Æ6) 9Æ76 0Æ28
1þ 6 (10Æ2) 4 (6Æ8) 5 (8Æ5) 15 (25Æ4)
2 4 (6Æ8) 6 (10Æ2) 4 (6Æ8) 14 (23Æ7)
3 1 (1Æ7) 3 (5Æ1) 4 (6Æ8) 8 (13Æ6)
4 3 (5Æ1) 4 (6Æ8) 4 (6Æ8) 11 (18Æ6)
Brunnstrom stage
Upper extremity
2 8 (13Æ6) 9 (15Æ3) 9 (15Æ3) 26 (44Æ1) 1Æ24 0Æ97
3 4 (6Æ8) 6 (10Æ2) 6 (10Æ2) 16 (27Æ1)
4 4 (6Æ8) 5 (8Æ5) 5 (8Æ5) 14 (23Æ7)
5 1 (1Æ7) 1 (1Æ7) 1 (1Æ7) 3 (5Æ1)

186  2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
JAN: ORIGINAL RESEARCH Effects of a ROM exercise

Table 1 (Continued)

Group

Usual care Intervention I Intervention II

Characteristics (n ¼ 17) (n ¼ 21) (n ¼ 21) Total v2 P value

Lower extremity
2 6 (10Æ2) 7 (11Æ9) 8 (13Æ6) 21 (35Æ6) 3Æ37 0Æ50
3 6 (10Æ2) 9 (15Æ3) 9 (15Æ3) 24 (40Æ7)
4 6 (10Æ2) 5 (8Æ5) 4 (6Æ8) 14 (23Æ7)

Value is number (percentage) unless otherwise indicated.

Participants in usual care group received no treatment from the researchers in study sessions. Participants in intervention I group worked all
extremity joints to the possible ROM by themselves under the researchers’ supervision. Participants in intervention II group worked all joints as
intervention I group did, and the researchers offered help to achieve the maximum ROM and to complete some joint motions beyond their
ability.

Table 2 Comparison of baseline scores for age, time of hemiplegia, time of residence, SPMSQ, FIMTM and perception of pain (n ¼ 59)

Group

Usual care Intervention I Intervention II

(n ¼ 17) (n ¼ 21) (n ¼ 21) Total Mean (SD ) F (2,56) P value

Age, years 74Æ06 (8Æ47) 75Æ86 (6Æ62) 74Æ43 (8Æ49) 75Æ05 (7Æ77) 0Æ23 0Æ799
Time of hemiplegia, months 58Æ24 (31Æ65) 61Æ67 (29Æ82) 59Æ24 (30Æ56) 59Æ81 (33Æ13) 0Æ05 0Æ948
Time of residence, months 52Æ35 (32Æ26) 51Æ76 (24Æ22) 52Æ10 (40Æ04) 52Æ05 (32Æ29) 0Æ01 0Æ998
SPMSQ 7Æ29 (1Æ96) 7Æ14 (2Æ26) 7Æ14 (1Æ82) 7Æ19 (1Æ99) 0Æ03 0Æ857
FIMTM ADL 36Æ17 (7Æ63) 35Æ48 (8Æ72) 34Æ29 (7Æ44) 35Æ24 (7Æ87) 0Æ28 0Æ759
GDS-15 10Æ18 (1Æ85) 9Æ95 (1Æ83) 9Æ71 (1Æ96) 9Æ93 (2Æ33) 0Æ17 0Æ841
Pain 13Æ88 (7Æ03) 13Æ19 (7Æ42) 16Æ14 (8Æ05) 14Æ44 (7Æ53) 0Æ87 0Æ425

Data are given as mean (SD ) except where noted.

SPMSQ, Short Portable Mental Status Questionnaire; FIMTM, Functional Independence Measure; GDS-15, Geriatric Depression Scale-Short
Form; ADL, activities of daily living.
Participants in usual care group received no treatment from the researchers in study sessions. Participants in intervention I group worked all
extremity joints to the possible ROM by themselves under the researchers’ supervision. Participants in intervention II group worked all joints as
intervention I group did, and the researchers offered help to achieve the maximum ROM and to complete some joint motions beyond their
ability.

group. However, the differences between the two interven-
tion groups were not statistically significant (see Table 4).
The results suggested that, with or without physical help
from nursing staff, doing the ROM might have the most
measurable effects.
Pain scores in both intervention groups revealed a tendency
for reduction. The pain scores used for analysis was the sum
of pain scores on 17 joint angles. The average pre- and post-
intervention scores for pain increased 5Æ41 in the usual care
group. However, pain scores dropped on average 7Æ62 in the
intervention group I, and decreased by 10Æ00 in the interven-
tion group II. A comparison of changes in pre- and
post-intervention pain scores among the three groups
showed a statistically significant difference (F(2,56) ¼ 42Æ82
P < 0Æ001) between the usual care group and both interven-
tion groups, but no significant difference was found between
the two intervention groups by post hoc analysis.
Depressive symptoms measured by the GDS-15 also
showed a statistically significant difference among the three
groups. GDS scores revealed a tendency for decrease in both
intervention groups. It is quite striking that the mean
difference of GDS score was þ2Æ35 in the usual care group,
4Æ76 in intervention group I, and 4Æ77 in intervention
group II. The Scheffe test revealed that the difference was
statistically significant between the usual care group and both
intervention groups (P < 0Æ05). However, no statistically
significant difference was found between intervention groups
I and II (Table 4).
Discussion
Participants in both intervention groups showed an increase
in all joint angles with the subjects in intervention group II
showing greater improvement in joint flexibility. The greater

 2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd 187
C.-N. Tseng et al.

Table 3 Comparison of changes (mean difference) in joint angles among groups (n ¼ 59)

Group

Joint angle Usual carea (n ¼ 17) Intervention Ib (n ¼ 21) Intervention IIc (n ¼ 21) F (2,56) P value Scheffe Post hoc test

Shoulder
Flexion 8Æ06 (7Æ34) 8Æ43 (8Æ35) 17Æ95 (7Æ23) 54Æ46 0Æ001 (a) < (b) < (c)
Extension 5Æ87 (5Æ71) 5Æ76 (7Æ97) 13Æ95 (6Æ02) 36Æ93 0Æ001 (a) < (b) < (c)
Abduction 5Æ73 (6Æ61) 6Æ00 (5Æ89) 15Æ53 (11Æ36) 27Æ50 0Æ001 (a) < (b) < (c)
Adduction 5Æ00 (6Æ22) 4Æ67 (4Æ83) 13Æ57 (9Æ41) 31Æ79 0Æ001 (a) < (b) < (c)
Elbow
Flexion 4Æ59 (5Æ65) 3Æ57 (4Æ20) 10Æ38 (4Æ32) 33Æ84 0Æ001 (a) < (b) < (c)
Extension 2Æ76 (4Æ79) 3Æ24 (7Æ90) 6Æ86 (9Æ86) 6Æ89 0Æ002 (a) < (b),(c)
Wrist
Flexion 6Æ47 (5Æ93) 5Æ86 (5Æ49) 10Æ48 (4Æ83) 18Æ33 0Æ001 (a) < (b) < (c)
Extension 8Æ12 (10Æ82) 5Æ81 (4Æ38) 13Æ71 (4Æ77) 46Æ69 0Æ001 (a) < (b) < (c)
Hip
Flexion 7Æ06 (8Æ27) 3Æ62 (4Æ01) 12Æ95 (5Æ66) 50Æ44 0Æ001 (a) < (b) < (c)
Abduction 4Æ76 (5Æ14) 0Æ81 (3Æ53) 7Æ11 (5Æ41) 28Æ58 0Æ001 (a) < (b) < (c)
Adduction 3Æ59 (7Æ74) 2Æ14 (2Æ59) 7Æ68 (5Æ15) 19Æ61 0Æ001 (a) < (b) < (c)
Knee
Flexion 2Æ65 (3Æ66) 2Æ38 (3Æ19) 8Æ90 (6Æ98) 25Æ74 0Æ001 (a) < (b) < (c)
Extension 3Æ94 (5Æ10) 2Æ00 (3Æ78) 9Æ57 (11Æ39) 14Æ83 0Æ001 (a) < (b),(c)
Ankle
Dorsal flexion 3Æ53 (6Æ06) 4Æ00 (6Æ17) 8Æ24 (8Æ03) 13Æ93 0Æ001 (a) < (b),(c)
Plantar flexion 3Æ59 (3Æ24) 2Æ05 (3Æ02) 6Æ05 (4Æ99) 28Æ78 0Æ001 (a) < (b) < (c)
Eversion 2Æ82 (2Æ04) 1Æ38 (3Æ71) 4Æ71 (2Æ92) 29Æ21 0Æ001 (a) < (b) < (c)
Inversion 2Æ82 (2Æ69) 0Æ86 (2Æ65) 4Æ25 (3Æ24) 27Æ70 0Æ001 (a) < (b) < (c)

Change means ‘mean difference’; post-test value minus pretest value and data are given as mean (SD ) except where noted.
Participants in usual care group received no treatment from the researchers in study sessions. Participants in intervention I group worked all
extremity joints to the possible ROM by themselves under the researchers’ supervision. Participants in intervention II group worked all joints as
intervention I group did, and the researchers offered help to achieve the maximum ROM and to complete some joint motions beyond their
ability.

Table 4 Comparison of changes in FIMTM, GDS-15 and pain scores by group (n ¼ 59)

Group

Measure Usual carea (n ¼ 17) Intervention Ib (n ¼ 21) Intervention IIc (n ¼ 21) F (2,56) P value Scheffe Post hoc test

FIMTM ADL 1Æ01 (2Æ81) 0Æ67 (1Æ71) 0Æ81 (1Æ72) 4Æ49 0Æ015 (a)<(b),(c)
GDS-15 2Æ35 (1Æ99) 4Æ76 (2Æ26) 4Æ77 (2Æ86) 52Æ06 0Æ000 (a)>(b),(c)
Pain 5Æ41 (4Æ15) 7Æ62 (4Æ91) 10Æ00 (6Æ42) 46Æ29 0Æ001 (a)>(b),(c)

Change means ‘mean difference’; post-test value minus pretest value. Data are given as mean (SD ) except where noted.
Participants in usual care group received no treatment from the researchers in study sessions. Participants in intervention I group worked all
extremity joints to the possible ROM by themselves under the researchers’ supervision. Participants in intervention II group worked all joints as
intervention I group did, and the researchers offered help to achieve the maximum ROM and to complete some joint motions beyond their
ability.
FIMTM, Functional Independence Measure; GDS-15, Geriatric Depression Scale-Short Form; ADL, activities of daily living.

improvement in intervention group II indicated that, if a
ROM exercise could be completed with nurses’ assistance in
achieving maximum joint angles, greater effects could be
expected. This finding might be partly because the majority of
participants in this sample were in a spastic or synergic stage
and many were unable to complete or achieve maximum
ROM alone. Therefore, more instrumental help from staff is
needed to achieve maximal benefits. Future studies are
needed to examine such staffing effects in the broader
context.
In terms of joint angles, similar to our findings, Clough and
Maurin (1983) studied four subjects who underwent a

188  2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
JAN: ORIGINAL RESEARCH
4 weeks, twice per day, and three times per joint passive
ROM exercise programme and they found a greater effect in
the joint angles than those obtained in our study. However, a
device was used in their study to maintain the joint angles in
extension. Conversely, Kaegi et al. (1995) found no improve-
ment in the change of joint angles in their study. The low
frequency of their ROM exercise protocol (only twice a
week) might be the reason. Nevertheless, these results suggest
that an effective ROM exercise should be carried out multiple
times during the day and over a period of time to generate
positive outcomes and benefits.
Activities of daily living function decreased in the usual
care group during the 4-week study period while it improved
in both intervention groups, with statistical significance
found only in intervention group II. This finding further
supported the importance of nurses’ presence, physically and
emotionally, in helping participants complete the ROM
exercise. One previous study has shown a greater effect on
ADL scores (mean difference ¼ 14Æ3, n ¼ 163) compared to
our results (Lin et al. 1999). Given that their extensive
intervention (including gait training, postural balance train-
ing and ADL performance training) and a younger sample
(64Æ1 years mean age, time since stroke 26Æ9 days), better
results should not be surprising. In other words, participants
in Lin’s study were better positioned in the golden time for
physical rehabilitation and some spontaneous recovery might
also be captured as time goes along. Nevertheless, our finding
adds to the literature which supports the effect of a simple
nurse-led ROM exercise in maintaining functional status
even when the participants are older and have suffered from
stroke related disability for some period of time.
Our findings also indicated that joint tightness and soreness
caused by contracture from stroke could be decreased by this
simple ROM exercise protocol. Perception of pain is
important because it may deter the participants from doing
ROM and create a vicious cycle of immobility. Observation
of the participants in the usual care which indicated a higher
perception of pain and a reduction on joint angles is
consistent with this concept. The finding was similar to the
report of Wasilewski et al. (1990), in which a ROM exercise
alleviated the perception of pain for patients who underwent
total knee arthoplasty.
Effects of lessening depressive symptoms have been shown
in many exercise studies (Zigmond & Snaith 1983, Van
Deusen & Harlowe 1987b). Our finding is consistent with
previous works showing that a simple nurse-led ROM
exercise could improve both physical and psychological
well-being of patients following stroke. The improvement in
depressive symptoms is statistically significant in this study.
However, it is not known if improvement in depressive
 2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
Effects of a ROM exercise
symptoms was from the ROM exercise per se, or it was
because of the attention effect of the researcher’s presence, or
both. This 4-week, simple, nurse-led ROM exercise did
statistically significantly improve the joint flexibility, ADL
functional status and perception of pain in this sample of
institutionalized bedridden older people with stroke. There-
fore, depressive symptoms might also be alleviated because of
the mitigation of the abovementioned physical health. One
way or the other, such simple, nurse-led ROM exercise
protocols should be promoted in the LTC facilities where
high-technological and specialized physical rehabilitation is
not feasible under the current reimbursement system.
In summary, a simple ROM exercise programme, comple-
ted by participants alone or with assistance from a nurse,
enhanced joint flexibility, functional status, and lessened
perception of pain and depressive symptoms in this sample of
59 bedridden older stroke survivors living in LTC facilities.
The use and effects of ROM exercise should not be
overlooked for institutionalized bedridden older people who
are beyond the optimum time for stroke rehabilitation.
Additionally, ROM exercise is beneficial not only in improv-
ing participants’ physical health, but also in enhancing their
psychological health by reducing depressive symptoms.
It might be asked if there are role conflicts between nurses
and physiotherapists in this important work of minimizing
dependency for bedridden stroke survivors. Arguments could
be made that including an exercise regimen as part of nursing
care is worthwhile and raises no conflict. First, in many parts
of the world, formal physiotherapy is not available to
bedridden patients with stroke under current healthcare
systems and less than one-third of LTC facilities in Taiwan
have physiotherapists on staff. Nurses are usually the most
educated staff in LTC facilities and may be the only hope for
these patients. Secondly, human contact and touch is one of
the core values of nursing since the time of Florence
Nightingale. Assisting bedridden patients conducting ROM
provides human contact and amplifies the trust and intimacy
that these patients desperately need. Apart from ROM
exercise, the physiotherapist’s expertise is more focused on
elements such as gait training, postural balance training, and
ADL performance training. While resources are limited,
nurses and physiotherapists should work in partnership to
ensure maximal benefits for patient outcomes.
Study limitations
Despite its careful design, the study had several limitations.
First, the sample size was small and because of selection
criteria, a relatively small proportion of patients with stroke
in LTC facilities were actually entered in the trial. Thus, the
189
C.-N. Tseng et al.
What is already known about this topic
• Physical rehabilitation, including physiotherapy and
exercise programmes, has demonstrated positive out-
comes with stroke survivors.
• Many older stroke survivors in long-term care facilities
have limited access to formal physical rehabilitation.
• The effectiveness of range-of-motion exercise in
improving function has not been fully studied for older
stroke survivors.
What this paper adds
• Participants, who received a 4-week, twice-per-day,
6 days-per-week range-of-motion exercise programme
showed statistically significant improvements in joint
angles, activity function, perception of pain, and
depressive symptoms compared with a usual care
group.
• Greater improvement was achieved in the group
receiving a nurse’s assistance in achieving maximum
joint angles.
• Further studies with larger samples and a longitudinal
design are needed to investigate the long-term effects of
such exercise programmes in maximizing function,
reducing healthcare utilization, and enhancing quality
of life for stroke survivors.
power of the study was unknown. Secondly, how long the
positive effects of ROM exercise programme will last remains
to be seen. Thirdly, the data were collected in 1999 and the
continued validity of the findings might be questioned.
However, despite the advance of medical technology, little
has changed for this severely affected group. This study offers
a solid beginning for a simple ROM exercise programme to
be studied and implemented for bedridden older stroke
survivors in LTC facilities.
Conclusion
Stroke is an international problem and nurses have an
important role to play (Mitchell & Moore 2004). For
institutionalized, bedridden older people with stroke, the
positive effects of a 10–20 minutes nurse-led ROM exercise
programme can be significant. In current cost-constrained
healthcare systems worldwide, there are many competing
needs for nursing time, and spending more time at the bedside
with patients is not easy. Given the promising effects of a
simple nurse-led ROM exercise for bedridden stroke
190  2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
survivors, routine ROM exercises should be seriously con-
sidered in LTC facilities where older people with hemiplegia
from stroke reside. Future studies are needed to evaluate the
cost-effectiveness of such ROM exercise programmes in strict
monetary terms.
Nevertheless, a simple, nurse-led ROM exercise seems to
provide beneficial effects for both physical and psychosocial
health of institutional bedridden older people with stroke.
This study offers a solid starting point for such ROM
interventions to be studied, designed and implemented.
Institutional bedridden older people with stroke, a vulnerable
and underserved population, are in need of good nursing
care. Future studies with a larger sample and longitudinal
design are needed to investigate the long-term effects of such
ROM exercise programmes in maximizing function, reducing
healthcare utilization, and enhancing quality of life of this
very needy population.
Author contributions
CNT and LCL were responsible for the study conception and
design. CNT, CHCC and LCL were responsible for the
drafting of the manuscript. CNT performed the data collec-
tion. CNT and SCW performed the data analysis. CNT
obtained funding. CNT, CHCC and LCL provided adminis-
trative support. SCW provided statistical expertise. CNT,
CHCC and LCL made critical revisions to the paper. CNT,
CHCC and LCL supervised the study.
References
Akay M. & Marsh A. (2001) Rehabilitation and Treatment. Wiley,
New York.
American Heart Association (2005) Heart Disease and Stroke
Statistics: 2005 Update. American Heart Association, Dallas, TX,
USA.
Chen C.C.-H., Kenefick A.L., Tang S.T. & McCorkle R. (2004)
Utilization of a comprehensive geriatric assessment in cancer
patients. Critical Reviews in Oncology/Hematology 49(1), 53–
67.
Chen C.C.-H., Chang C.K., Chyun D.A. & McCorkle R. (2005)
Dynamics of nutritional health in a community sample of Amer-
ican elders. Advances in Nursing Science 28, 376–389.
Clough D.M. & Maurin J.J. (1983) ROM versus NRx: care of a CVA
patient. Journal of Gerontological Nursing 9(5), 278–286.
Department of Health, the Execution Yuan. (2005) Annual statistical
report. Department of Health website. Retrieved from http://
www.doh.gov.tw/statistic/data/causeofdeathreport/93/table9.xls
on 25 May 2006.
Diego M.A., Field T., Hernandez-Reif M., Hart S., Brucker B.,
Field T. & Burman I. (2002) Spinal cord patients benefit from
massage therapy. International Journal of Neuroscience 112(2),
133–142.
JAN: ORIGINAL RESEARCH
Goldsmith J.R., Lidtke R.H. & Shott S. (2002) The effects of range-
of-motion therapy on the planter pressure of patients with diabetes
mellitus. Journal of the American Pediatric Medical Association
92(9), 483–490.
Kaegi C., Lapointe M.F., Giroux F. & Bourbonnais D. (1995)
Absence of change in the passive joint movements of long term care
patients following the application of a passive range of motion
exercise program. Physical and Occupational Therapy in Ger-
iatrics 13(1/2), 81–100.
Keith R.A., Granger C.V., Hamilton B.B. & Sherwin F.S. (1987) The
functional independence measure: a new tool for rehabilitation.
Advances in Clinical Rehabilitation 1, 6–18.
Kisner C. & Colby L.A. (1996) Therapeutic Exercise: Foundations
and Techniques. F.A. Davis, Philadelphia.
Kwakkel G., van Peppen R., Wagenaar R.C., Dauphinee S.W.,
Richards C., Ashburn A., Miller K., Lincoln N., Partridge C.,
Wellwood I. & Langhorne P. (2004) Effects of augmented exercise
therapy time after stroke: a meta-analysis. Stroke 35, 2529–2536.
Lin C.Y., Lu C.H., Yu S. & Yang Y.Y. (1998) Correlations between
scores on Chinese versions of long and short forms of the Geriatric
Depression Scale among elderly Chinese. Psychological Reports
82(1), 211–214.
Lin J.H., Liu J.K., Lin Y.T. & Huang M.H. (1999) Clinical data and
functional restoration of stroke patients receiving rehabilitation
therapy. Chinese Journal of Physical Therapy 24(2), 29–34.
Lin J.H., Lee C.H., Huang M.H., Liu C.K. & Lin Y.T. (2000)
Factors influencing long-term care institution utilization for
stroke patients after in-patient rehabilitation. Physical Therapy
25(4), 1–9.
Lin J.H., Chang C.M., Huang M.H., Liu C.K. & Lin Y.T. (2001)
A follow-up study of first stroke patients after inpatient re-
habilitation. Formosan Journal of Physical Therapy 26(4), 196–
204.
Long A.F., Kneafsey L.R., Ryan J. & Berry J. (2002) The role of the
nurse within the multi-professional rehabilitation team. Journal of
Advanced Nursing 37, 70–78.
Lynch D., Ferraro M., Krol J., Trudell C.M., Christos P. & Volpe
B.T. (2005) Continuous passive motion improvement shoulder
joint integrity following stroke. Clinical Rehabilitation 19(6), 594–
599.
Lynn M.R. (1986) Determination and quantification of content
validity. Nursing Research 35(6), 382–385.
Miller E. & Easton K. (2000) Perceived losses following stroke.
Rehabilitation Nursing 25(5), 192–195.
Mitchell E. & Moore K. (2004) Stork: holistic care and management.
Nursing Standard 33(18), 43–52.
Pfeiffer E. (1975) A Short Portable Mental Status Questionnaire for
the assessment of organic brain deficit in elderly patients. Journal
of the American Geriatric Society 23(10), 433–441.
 2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
Effects of a ROM exercise
Pinedo S. & de la Villa F.M. (2001) Complications in the hemiplegic
patient in the first year after the stroke. Revista de Neurologia
32(3), 206–209.
Prencipe M., Ferretti C., Casini A.R., Santini M., Giubilei F. &
Culasso F. (1997) Stroke, disability, and dementia: results of a
population survey. Stroke 28, 531–536.
Shaw D.K., Deutsch D.T. & Bowling R.J. (1989) Efficacy of shoulder
range of motion exercise in hospitalized patients after coronary
artery bypass graft surgery. Heart and Lung 18(4), 364–369.
Spector W.D. & Takada H.A. (1991) Characteristics of nursing
homes that affect resident outcomes. Journal of Aging Health 3(4),
427–454.
Taylor T.N., Davis P.H., Torner J.C., Holmes J., Meyer J.W. &
Jacobson M.F. (1996) Lifetime cost of stroke in United States.
Stroke 27(9), 1459–1466.
Toso V., Gandolfo C., Paolucci S., Provinciali L., Torta R., Grass-
ivaro N. & DESTRO Study Group (2004) Post-stroke depression:
research methodology of a large multicentre observational study.
Neurological Sciences 25(3), 138–144.
Valentine-Garzon M.A., Maynard M. & Selznick S.Z. (1992) ROM
dance program effects on frail elderly women in an adult day-care
center. Physical and Occupational Therapy in Geriatrics 11(1),
63–83.
Van Deusen J. & Harlowe D. (1987a) A comparison of the ROM
dance home exercise/rest program with traditional routines. The
Occupational Therapy Journal of Research 7, 349–360.
Van Deusen J. & Harlowe D. (1987b) The efficacy of ROM dance
program for adults with rheumatoid arthritis. American Journal of
Occupational Therapy 41(2), 90–95.
Van Deusen J. & Harlowe D. (1988) One year following-up results
of ROM dance research. The Occupational Therapy Journal of
Research 8(1), 52–54.
Wasilewski S.A., Woods L.C. & Torgerson W.L. (1990) Value of
continuous passive motion in total knee arthoplasty. Orthopedics
13(3), 291–295.
Xiong J.L. & Lin G.C. (1999) A survey of physical therapy’s need of
residents with stroke in long term care facilities in Taipei. The
Journal of Long Term Care 3(1), 45–52.
Yeh S.C.J. & Lo S.K. (2003) Is rehabilitation associated with change
in functional status among nursing home residents? Journal of
Nursing Care Quality 19(1), 58–66.
Yeh L., Wen M.J., Tu S.L., Lin T. & Dai Y.S. (1999) Functional
status of long-term care residents in Tainan city. Public Health
26(3), 183–192.
Yesavage J.A., Brink T.L. & Rose T.L. (1983) Development and
validation of a geriatric depression scale. Journal of Psychiatric
Research 17, 31–49.
Zigmond A.S. & Snaith R.P. (1983) The hospital anxiety and
depression scale. Acta Psychiatrica Scandinavica 67(6), 361–370.
191