Archives of Physical Medicine and Rehabilitation

journal homepage: www.archives-pmr.org

Archives of Physical Medicine and Rehabilitation 2014;95:1423-32

ORIGINAL ARTICLE

Improvement of Upper Extremity Motor Control and

Function After Home-Based Constraint Induced
Therapy in Children With Unilateral Cerebral Palsy:
Immediate and Long-Term Effects
Hsieh-ching Chen, PhD,a,* Chia-ling Chen, MD, PhD,b,c,* Lin-ju Kang, PhD,c
Ching-yi Wu, ScD, OTR,d Fei-chuan Chen, MS,c,e Wei-hsien Hong, PhDf
From the aDepartment of Industrial Engineering and Management, National Taipei University of Technology, Taipei; bDepartment of Physical
Medicine and Rehabilitation, Chang Gung Memorial Hospital, Linkou; cGraduate Institute of Early Intervention, College of Medicine, Chang
Gung University, Tao-Yuan; dDepartment of Occupational Therapy and Graduate Institute of Behavioral Sciences, College of Medicine, Chang
Gung University, and Healthy Ageing Research Center at Chang Gung University, Taoyuan; eDepartment of Rehabilitation, Taipei Tzu Chi
Hospital, Buddhist Tzu Chi Medical Foundation, Taipei; and fDepartment of Sports Medicine, China Medical University, Taichung, Taiwan.
* H.-c. Chen and C.-l. Chen contributed equally to this work.

Abstract
Objective: To investigate the long-term effects of home-based constraint induced therapy (CIT) on motor control underlying functional change in
children with unilateral cerebral palsy (CP).
Design: Randomized controlled trial.
Setting: Home based.
Participants: Children with unilateral CP (NZ45; aged 6e12y) were randomly assigned to receive home-based CIT (nZ23) or traditional
rehabilitation (TR) (nZ22).
Interventions: Both groups received a 4-week therapist-based intervention at home. The home-based CIT involved intensive functional training
of the more affected upper extremity during which the less affected one was restrained. The TR involved functional unimanual and bimanual
training.
Main Outcome Measures: All children underwent kinematic and clinical assessments at baseline, 4 weeks (posttreatment), and 3 and 6 months (follow-
up). The reach-to-grasp kinematics were reaction time (RT), normalized movement time, normalized movement unit, peak velocity (PV), maximum grip
aperture (MGA), and percentage of movement where MGA occurs. The clinical measures were the Peabody Developmental Motor Scales, Second
Edition (PDMS-2), Bruininks-Oseretsky Test of Motor Proficiency (BOTMP), and Functional Independence Measure for children (WeeFIM).
Results: The home-based CIT group showed a shorter RT (P<.05) and normalized movement time (P<.01), smaller MGA (PZ.006), and fewer
normalized movement units (PZ.014) in the reach-to-grasp movements at posttreatment and follow-up than the TR group. The home-based CIT
group improved more on the PDMS-2 (P<.001) and WeeFIM (P<.01) in all posttreatment tests and on the BOTMP (P<.01) at follow-up than the
TR group.
Conclusions: The home-based CIT induced better spatial and temporal efficiency (smoother movement, more efficient grasping, better movement
preplanning and execution) for functional improvement up to 6 months after treatment than TR.
Archives of Physical Medicine and Rehabilitation 2014;95:1423-32
ª 2014 by the American Congress of Rehabilitation Medicine

Supported by the National Science Council Taiwan, Taiwan (grant nos. NSC 102-2410-H-182- Cerebral palsy (CP) is a major childhood disability.1 The motor
018 and NSC 101-2314-B-182-004-MY3) and Chang Gung Medical Foundation, Taiwan (grant no. disorders of CP are often associated with sensation, perception,
CMRPG 391671-3).
Clinical Trial Registration No.: NCT01076257.
cognition, communication, behavior, and musculoskeletal prob-
Disclosures: none. lems.2 Children with unilateral or hemiparetic CP often present

0003-9993/14/$36 - see front matter ª 2014 by the American Congress of Rehabilitation Medicine
http://dx.doi.org/10.1016/j.apmr.2014.03.025

Downloaded for John Carlo Conge (carloconge@gmail.com) at Our Lady of Fatima University from ClinicalKey.com by Elsevier on July 06, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
1424
with unilateral spasticity, weakness, early asymmetry in move-
ments and/or functional abilities (ie, asymmetrical crawl or early
hand preference), and a high rate of partial seizures.1 The upper
extremities are often more involved than the lower extremities in
children with unilateral CP.2 Children with unilateral CP typically
learn strategies and techniques to perform daily tasks with just the
unaffected hand.3 Children with unilateral CP typically do not use
their more affected upper extremity. They frequently use only
their less affected upper extremity to increase the efficiency of
performing tasks, which is called developmental disregard.3-5
These problems further limit daily activities and participation of
children with unilateral CP.
Constraint induced therapy (CIT) is an intervention approach
for treating developmental disregard in patients with unilateral
motor involvement.6,7 CIT involves intensive functional training
and shaping for the more affected upper extremity, a constraint for
the less affected upper extremity, and a behavioral technique
termed “transfer package.”6,8 The component of the transfer
package in CIT could facilitate turning the therapeutic gains
achieved in the clinic into real-world activities and retention of the
treatment effects 6 months later.8 The aim of CIT is to reverse the
behavioral suppression of the movement of the more affected
upper extremity.3 Several systematic reviews3,7,9 and randomized
controlled trials (RCTs)4,6,10-13 have demonstrated that CIT im-
proves functional performance and activities of daily living in
children with CP.
Home-based CIT places the practice in a natural context and is
the preferred method for the treatment of children with CP over
clinic-based CIT.14 A home environment can enhance motivation,
engagement, and repetition in everyday activities for children.15-17
Home-based therapy can facilitate the transfer of the training ef-
fects of CIT to daily life18 and may also enhance greater transfer
package effects than clinic-based CIT. Furthermore, the home-
based CIT protocol, with its relatively moderate intensity and
shortened constraint time, might balance the effectiveness
and compliance of children and caregivers.13,19,20 Therefore,
home-based CIT may be an effective alternative to conven-
tional CIT.13,19,20
Therapies for upper extremity dysfunction in children with CP
typically target functional outcomes and underlying motor control
impairments. Kinematic analysis can reveal the motor control
strategies underlying the performance of motor tasks. Reach and
reach-to-grasp are fundamental motor tasks that a child needs to
interact with his/her surrounding environment in daily living. The
kinematic parameters of a reach task (eg, reaction time [RT],
normalized movement time, normalized movement unit) can
provide information on movement preplanning, execution, and
smoothness.21-23 The kinematic analysis of a reach-to-grasp task
can provide additional information on grip formation (eg,
maximum grip aperture [MGA], percentage of movement time
List of abbreviations:
BOTMP Bruininks-Oseretsky Test of Motor Proficiency
CIT constraint induced therapy
CP cerebral palsy
MGA maximum grip aperture
PDMS-2 Peabody Developmental Motor Scales, Second Edition
PV peak velocity
RCT randomized controlled trial
RT reaction time
TR traditional rehabilitation
Downloaded for John Carlo Conge (carloconge@gmail.com) at Our Lady of Fatima University from ClinicalKey.com by Elsevier on July 06, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
H.-c. Chen et al
during which MGA occurs). High efficiency of a reach-to-grasp
task is characterized by a short RT and normalized movement
time (movement preplanning and execution), few normalized
movement units (increased movement smoothness), small MGA
(minimal hand opening while grasping), and high percentage of
movement where MGA occurs.23 Research has demonstrated that
CIT produces a greater improvement in motor control on reaching
or reach-to-grasp tasks and functional performance in patients
with stroke than traditional rehabilitation (TR) therapy.21-23 In this
study, the RT, normalized movement time, normalized movement
unit, MGA, and percentage of movement where MGA occurs
were selected for kinematic analysis because these parameters
indicate the ability to complete the functional tasks smoothly and
efficiently.20-23
Previously, we compared home-based CIT with TR and
examined the effects of home-based CIT on the motor perfor-
mance of children with CP over different time frames. Subsets of
data from this study have been published in our previous 2 arti-
cles.19,20 In the first article,19 we used only those children from
this data set who were 6 to 8 years of age and compared the
effectiveness of home-based CIT with TR at posttreatment and 3-
month follow-up. The Peabody Developmental Motor Scales,
Second Edition (PDMS-2)24 and the Pediatric Motor Activity
Log25 were selected as the main functional outcome measures. We
found that home-based CIT improved motor efficacy and induced
greater gains in general functional performance than TR at both
times. In the second article,20 we used all children from this data
set and focused on examining the effects of home-based CIT on
reaching kinematics by using a simple reach task at posttreatment,
but not at follow-up times. The Bruininks-Oseretsky Test of Motor
Proficiency (BOTMP)26 and Pediatric Motor Activity Log were
used as the main functional outcome measures.20 We found that
home-based CIT induced greater temporal and spatiotemporal
efficiency in the short term as demonstrated by shorter RT and
normalized movement time and higher peak velocity (PV) than
TR.20 However, there have been no studies, to our knowledge,
investigating the long-term changes in motor control strategies
after CIT in children with CP using a more complex reach-to-
grasp task (eg, reach and grasp a ball or can) than a simple
reach task (eg, reach and press a bell or button).
The long-term effects of either CIT27-29 or home-based
CIT13,19 on motor function of children with CP have already
been established. To our knowledge, the long-term effects of
home-based CIT on the motor control strategies of children with
CP have not yet been elucidated. Therefore, the current study is a
follow-up RCT to our 2 previous studies.19,20 We compared both
the immediate (posttreatment) and maintained (3- and 6-mo) ef-
fects of home-based CIT and TR therapy by combining the ki-
nematic analysis of a reach-to-grasp task and clinical evaluations.
We will determine whether functional improvement is accompa-
nied by changes in motor control in both short- and long-term time
frames. For this study, the BOTMP26 and PDMS-224 were selected
as the primary outcome measures. The Functional Independence
Measure for children (WeeFIM)30 was used as the secondary
outcome measure. The rationale for using dose-matched in-
terventions for both groups is based on RCTs of CP13,19,20 and
stroke.21-23 We hypothesized that home-based CIT would generate
greater functional gains accompanied by better motor control
strategies (shorter RT and normalized movement time, fewer
normalized movement units, smaller MGA and percentage of
movement where MGA occurs, larger PV) of reach-to-grasp than
TR in both the immediate phase (posttreatment) and maintenance
www.archives-pmr.org
Effects of constraint induced therapy in cerebral palsy
phase (3- and 6-mo follow-up). The study findings will help cli-
nicians elucidate the neuromotor control strategies underlying
functional improvement after home-based CIT.
Methods
Participants
Children with CP aged 6 to 12 years from the department of
physical medicine and rehabilitation of a tertiary medical center
were recruited for this study. The inclusion/exclusion criteria were
reported in one of our previous articles.20 Briefly, the inclusion
criteria included a diagnosis of congenital unilateral spastic CP,
considerable nonuse of the more affected upper extremity, active
extension movement of the wrist and metacarpophalangeal joint

10 , and no excessive muscle tone of the more affected upper
extremity. The exclusion criteria included a severe cognitive or
sensory disorder, severe concurrent illness or disease not typically
associated with CP, active medical conditions, and any major
surgery or nerve blockage within 6 months prior to the exposure to
CIT. Ethical approval was provided by the Institutional Review
Board for Human Studies of Chang Gung Memorial Hospital (no.
97-1823B). Caregivers of all participants gave written
informed consent.
Design
This was a single-blinded RCT (fig 1) as reported in our 2 pre-
vious articles.19,20 All children who participated in the study were
randomly assigned to either the home-based CIT or the TR group.
Among the 75 eligible candidates, a total of 45 children with
spastic unilateral CP (23 in the home-based CIT group, 22 in the
TR group) were included in the final analysis. Children were first
stratified by age into 2 strata (6e8y, 9e12y). The cut-off age was
8 years because the fine finger force coordination in children
grasping objects approximated that of adults by 8 years old.31,32
The randomization procedure was reported in one of our pre-
vious articles.20 Each age strata had a set of sealed envelopes to
facilitate the randomized allocation of 45 children into either the
home-based CIT or TR group. There were 2 sets of envelopes in
total, and each set contained 30 sealed envelopes labeled from 1 to
30 with a group allocation (home-based CIT or TR).
All children underwent kinematic analysis (reach-to-grasp
movement) and functional measures (primary and secondary
outcomes) in a laboratory setting before the intervention (pre-
treatment). An occupational therapist blinded to the group
assignment was trained to administer these outcome measures. All
children underwent these assessments again at 4 weeks immedi-
ately after the intervention (posttreatment), at a 3-month follow-
up, and at a 6-month follow-up. The children’s demographic
characteristics, more affected upper extremity, level of bimanual
fine motor function,33 and constraint time (in the home-based CIT
group only) were recorded.
Interventions
Details of the interventions have been described in one of our
previous articles.20 Both groups received individualized home-
based interventions of 3.5 to 4h/d, twice weekly for 4 weeks.
One certified physical therapist provided treatments for all the
children. The home-based CIT approach may reduce the treatment
dosage and constraint time needed to achieve benefits by
enhancing participant compliance.19,20 The home-based CIT
www.archives-pmr.org
Downloaded for John Carlo Conge (carloconge@gmail.com) at Our Lady of Fatima University from ClinicalKey.com by Elsevier on July 06, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
1425
program focused on the functional training of the more affected
upper extremity by applying the principles of shaping and repet-
itive task practice. Children in the home-based CIT group were
required to wear an elastic bandage and restraint mitten on the less
affected upper extremity for 3.5 to 4h/d for 4 weeks, and their
parents were asked to document the number of restraint hours.
Children in the TR group were engaged in functional unilateral or
bilateral upper extremity training using the principles of activity-
oriented approach, neurodevelopment treatment techniques, and
motor learning and control. Both groups continued their usual
clinic-based rehabilitation programs. Between treatment sessions,
children in both groups were encouraged to exercise or perform
daily activities under parental supervision (eg, reaching or
grasping objects, manipulating their own toys, eating, writing,
drawing, tool usage). Children in the home-based CIT group were
encouraged to exercise or perform daily functional activities with
the more affected upper limb, whereas those in the TR group
performed daily activities with unilateral or bilateral upper limbs.
Outcome measures
Primary outcomes
Subtest 8 of the BOTMP,26 which includes 8 items, was used to
evaluate upper extremity speed and dexterity. Subtest 8 assesses
the ability of the child to move and manipulate everyday objects
in timed tasks and involves both unimanual and bimanual tasks
(eg, placing pennies, sorting cards, stringing beads, displacing
pegs, drawing lines, marking dots).26 The subtest 8 of the
BOTMP was selected because the subtest measures upper
extremity speed and dexterity during a variety of reach-to-pinch
and hand manipulating movements. The unimanual tasks on the
BOTMP were measured with the more affected upper extremity.
The fine motor domain of the PDMS-2 was used to assess
upper extremity functional ability.24 The grasping subscale, which
includes 26 items, was selected to assess grasping, releasing, and
manipulating skills. Unimanual tasks were performed with the
more affected upper extremity. The BOTMP and PDMS-2, in
which higher raw scores indicate better performance, have well-
established psychometric properties for the total scale and each
subscale in children with motor disabilities.24,26 The PDMS-2
grasping subscale was selected because it measures functional
grasping skills that correspond to the reach-to-grasp kine-
matic task.
Secondary outcomes
The WeeFIM30 includes 3 functional subscales: self-care,
mobility, and cognition. The level of independence is rated by a
7-level ordinal rating system ranging from 7 (complete indepen-
dence) to 1 (total assistance). The WeeFIM self-care subscale,
which pertains to upper limb function, was selected to measure
performance in daily activities. The responsiveness of the Wee-
FIM instrument has been shown to document changes in the
functional abilities of children with chronic developmental
disabilities.30
Kinematic analysis
Details of the experimental setups for the kinematic analyses have
been described in one of our previous articles.20 During the reach-
to-grasp task (fig 2), the child sat on a chair adjacent to a table
with his/her trunk secured to the chair by a harness. The child
rested the affected hand on a pressure-sensitive hand switch
located in line with the child’s affected side. A ball 8cm in
1426 H.-c. Chen et al

Fig 1 Flow diagram of the randomization procedure.

diameter was positioned along the child’s midsagittal plane. The
distance to reach the ball was standardized to the child’s arm
length. If the maximum distance the child could reach was less
than his/her arm length, the distance to reach the ball was adjusted
to the maximum reachable distance. The child was instructed to
reach, grasp, and lift the ball at his/her own speed. A start signal
was used to start the task.
Reference markers were placed on the distal interphalangeal
joints of the thumb and index finger, styloid process of the ulna,
proximal end of the second metacarpal at the base of the second
finger, and ball. An 8-camera motion analysis systema was used to
capture the movement of these reference markers and to simul-
taneously collect 1 channel of analog signals. The analog signal
connected to the hand switch was used to identify movement
onset. The reference marker attached to the ball was used to
determine movement offset. Movement recording began when the
child’s hand moved off of the switch and terminated when the
marker attached to the ball moved.23
Data reduction and data analysis
The kinematic data from the reach-to-grasp movements were
processed by an analytical program coded in LabVIEW.b The
parameters were as follows: RT, normalized movement time,
normalized movement unit, PV, MGA, and percentage of
movement where MGA occurs.23 The RT is the interval from the
start signal to movement onset,34 movement time is the interval
between movement onset and movement completion (ie, lifting
the ball),23,35,36 and movement unit is used to characterize
movement smoothness and the control strategy of reaching that is
extracted from the ulnar styloid data. One movement unit com-
prises 1 acceleration phase and 1 deceleration phase. Normali-
zation of movement time and movement unit was performed to
adjust for differences in reaching distance.23 PV is defined as the
highest instantaneous velocity during the movement. MGA, a
measure of the extent of hand opening, was defined as the
maximum distance between the thumb and index finger. Per-
centage of movement where MGA occurs is a measure of the
online correction of target grasping during enclosure.37 A high
percentage of movement where MGA occurs indicates greater
temporal efficiency because less time is needed for the online
correction of target grasping.37 A small MGA indicates greater
spatial efficiency caused by less hand opening needed for
grasping objects.37 A small MGA is often considered to be
better because the child can grasp the objects with relatively
small aperture.

www.archives-pmr.org

Downloaded for John Carlo Conge (carloconge@gmail.com) at Our Lady of Fatima University from ClinicalKey.com by Elsevier on July 06, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
Effects of constraint induced therapy in cerebral palsy
Fig 2 Setup of the reach-to-grasp kinematic task (trunk restraint is not shown).
Statistical analysis
All statistical analyses were conducted using SPSS version 12.0
software.c To determine the comparability of the demographic and
clinical characteristics at baseline, an independent 2-sample t test
was applied for continuous variables, and a chi-square test was
applied for categorical variables. Analysis of covariance, while
controlling for pretest differences, was used to determine whether
the home-based CIT group performed significantly better than the
TR group for each outcome variable (reach-to-grasp kinematics
and scores for the PDMS-2, BOTMP, WeeFIM). During each
analysis, the pretest performance was controlled as a covariate,
group was the independent variable, and posttreatment perfor-
mance was the dependent variable. The significance level was set
at .05. The effect size (h2) was calculated for each outcome var-
iable to reflect the magnitude of the between-group differences. A
large effect is represented by an eta-square of at least .138, a
moderate effect by an eta-square of .059, and a small effect by an
eta-square of .01.38
Results
Among the 48 children who initially participated in this study, 1
child in the TR group was unable to complete the follow-up
because of a tight family schedule and lack of transportation to the
site of the posttest assessments (see fig 1). One child in each of the
home-based CIT and TR groups was excluded from the analysis
because their motor ability was insufficient to complete the
standardized study procedure of the reach-to-grasp kinematic
analysis. Consequently, a total of 45 children, 23 in the home-
based CIT group and 22 in the TR group, completed the inter-
vention, posttest, and follow-up measures.
www.archives-pmr.org
Downloaded for John Carlo Conge (carloconge@gmail.com) at Our Lady of Fatima University from ClinicalKey.com by Elsevier on July 06, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
1427
Demographic and baseline characteristics
The demographic characteristics of these 2 groups were compa-
rable (table 1). The pretest results indicated that no differences
existed between the 2 groups in the scores for functional measures
and reach-to-grasp kinematics, except for the normalized move-
ment unit and percentage of movement where MGA occurs scores.
The home-based CIT group had greater normalized movement
unit and lower percentage of movement where MGA occurs than
the TR group (PZ.007e.037).
Primary outcomes
The PDMS-2 grasping subscale and subtest 8 of the BOTMP scores
at all posttreatment times for both groups were higher than those at
baseline (table 2). The analysis of covariance results indicated that
the home-based CIT group improved more than the TR group on
the PDMS-2 grasping subscale not only at the posttreatment but
also at both follow-ups (3 and 6mo), with large effects (h2Z.604e
.658, P<.001). Furthermore, the home-based CIT group has also
shown greater improvement than the TR group on the subtest 8 of
the BOTMP at the 3- and 6-month follow-ups, with large effects
(h2Z.167e.193, PZ.003e.006), but not at posttreatment.
Secondary outcomes
For the daily activity assessments, the WeeFIM self-care subscale
scores for both groups improved at posttreatment and both follow-
ups (see table 2). The analysis of covariance results indicated that
the home-based CIT group improved more than the TR group on
the WeeFIM self-care subscale at posttreatment and both follow-
ups, with large effects (h2Z.195e.264, P<.001 to PZ.003).
1428 H.-c. Chen et al

Table 1 Demographic and baseline characteristics for

.658

.193

.264
h2
participants
Home-Based CIT TR Group

ANCOVA

<.001

.003

<.001
Variables Group (nZ23) (nZ22) P

P
Age (y) 8.731.90 8.662.00 .909*
Sex

80.87

10.05

15.05
F1,42
.873y

6-mo Follow-Up
Boys 11 (48) 10 (46)
Girls 12 (52) 12 (54)
More affected arm

43.553.40

10.141.42

43.413.44
TR (nZ22)
Right 12 (52) 11 (52) .884y
Left 11 (48) 11 (48)
BFMF
Level I 3 (13) 3 (14) .953y

hCIT (nZ23)
Level II 20 (87) 19 (86)

47.7802.78

11.7822.11

46.8305.27
Reach-to-grasp
kinematics
RT 0.570.01 0.570.01 .954*
nMT 0.610.05 0.630.03 .081*
nMU 0.290.03 0.270.03 .037*

.634

.167

.202
h2
PV 65.116.09 65.163.14 .953*
MGA 9.380.95 9.390.88 .955*

ANCOVA

<.001

.006

.002
PMGA 74.8118.93 88.6913.24 .007*
PDMS-2

P
Grasping 39.652.95 38.732.68 .277*
BOTMP

8.41
72.91

10.61
F1,42
3-mo Follow-Up
Subtest 8 4.911.35 4.641.14 .461*
WeeFIM

42.593.25

9.271.61

41.773.61
Self-care 39.576.33 37.413.67 .169*

TR (nZ22)
NOTE. Values are expressed as mean  SD, n (%), or as otherwise
indicated.
Abbreviations: BFMF, bimanual fine motor function; nMT, normalized
movement time; nMU, normalized number of movement unit; PMGA,
hCIT (nZ23)

percent time where MGA occurs. 46.042.73

10.872.16

44.785.72
* t test.
y
Chi-square test.
Descriptive statistics and results of ANCOVA for clinical measures

.604

.058

.195
Reach-to-grasp kinematics
h2

Abbreviations: ANCOVA, analysis of covariance; hCIT, home-based CIT.

Both groups had improved reach-to-grasp kinematics at post-
ANCOVA

<.001

.116

.003

treatment and follow-ups (table 3). Relative to the TR group, the

home-based CIT group improved more in RT (h2Z.133e.221,
P

PZ.001e.015) and normalized movement time (h2Z.158e.601,

NOTE. Descriptive statistics were presented as mean  SD.
2.57
61.15

10.19

P<.001 to PZ.008) at posttreatment and follow-ups, with large

F1,42
Posttreatment

effects. Additionally, the home-based CIT group improved more

than the TR group with respect to normalized movement unit at
41.273.27

7.861.64

39.733.66

the 6-month follow-up, with a moderate effect (h2Z.136,

TR (nZ22)

PZ.014), but not at posttreatment or 3-month follow-up. More-

over, the home-based CIT group improved more than the TR
group with respect to MGA at posttreatment, with a large effect
(h2Z.169, PZ.006), but not at the 3- and 6-month follow-ups.
hCIT (nZ23)

43.962.72

8.872.26

42.616.24

The PV and percentage of movement where MGA occurs scores

did not differ between the 2 groups.

Discussion
This study is the first, to our knowledge, to reveal the long-term
Subtest 8
Grasping

Self-care

effects of home-based CIT on motor control strategies by using

Variables
Table 2

WeeFIM
PDMS-2

BOTMP

reach-to-grasp kinematics in combination with functional

improvement in children with CP. Other related studies usually
focused only on functional improvement. Compared with TR, the

www.archives-pmr.org

Downloaded for John Carlo Conge (carloconge@gmail.com) at Our Lady of Fatima University from ClinicalKey.com by Elsevier on July 06, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
Effects of constraint induced therapy in cerebral palsy 1429

home-based CIT group improved more on the functional out-

.137
.601
.136
.013
.008
.005
comes, as measured by the PDMS grasping subscale, subtest 8 of

h2
the BOTMP, and WeeFIM self-care subscale, at most posttreat-

ANCOVA

.014
<.001
.014
.463
.564
.659
ment and follow-up time points. The home-based CIT group also
performed better than the TR group on the reach-to-grasp kine-

Abbreviations: ANCOVA, analysis of covariance; hCIT, home-based CIT; nMT, normalized movement time; nMU, normalized number of movement unit; PMGA, percent time where MGA occurs.
matics at posttreatment or the 3- and 6-month follow-ups,

6.65
63.36
6.64
0.55
0.34
0.20
demonstrating fewer normalized movement units, smaller MGA,
F1,42
6-mo Follow-up

shorter RTs, and shorter normalized movement times. That is,

home-based CIT induced greater temporal and spatial efficiencies

81.9710.39
0.460.04
0.530.03
0.220.03
70.961.96
8.620.94
of reach-to-grasp movements in children with CP than TR in terms
TR (nZ22)

of increasing movement preplanning and execution, movement

smoothness, and hand grasping efficiency during both the imme-
diate and maintenance phases. Home-based CIT induced better
motor control changes underlying greater functional improvement
hCIT (nZ23)

79.7112.59 than TR in both the short and long term. The current study offers
0.430.03
0.460.03
0.220.04
71.251.94
8.441.04

valuable kinematic biomarkers that support the neuromotor

models proposed to account for the home-based CIT effects on the
functional outcomes in children with CP. The kinematic parame-
ters, particularly RT and normalized movement unit, are the most
.221
.494
.137
.006
.084
.013

useful for creating biomarkers to examine the benefits of CIT. The

2
h

study results are significant for translating kinematic measures

into clinical effects from home-based CIT in children with CP.
.001
<.001
.049
.608
.057
.454
ANCOVA

Home-based CIT generated greater gains in both the grasping

P

skills (measured by the PDMS-2 grasping subscale) and daily

activities (measured by the WeeFIM self-care subscale) at post-
3.006
11.89
40.98

0.27
3.84
0.57
F1,42

treatment and both follow-ups than TR in children with CP.

3-mo Follow-Up

Home-based CIT led to greater gains than TR in upper extremity

speed and dexterity (as measured by the subtest 8 of the BOTMP)
77.2514.90
0.490.03
0.560.03
0.240.03
69.562.48
8.401.08
TR (nZ22)

at the follow-ups, but not at posttreatment. Enhanced motivation,

repetition, shaping, persistence in exploring familiar environ-
ments, and transfer package effects may be the reasons that the
effects were maintained after home-based CIT. A home envi-
ronment also provides a rich natural context to facilitate moti-
hCIT (nZ23)

85.2515.61
0.460.03
0.490.03
0.240.04
69.772.02
7.800.94

vation and engagement using familiar objects in everyday

functional activities.12,16,17 However, home-based CIT requires
Descriptive statistics and results of ANCOVA for reach-to-grasp kinematics

more time to induce upper extremity speed and dexterity changes.

Previous studies have demonstrated that clinic-based CIT can
lead to significant long-term improvements in unimanual capac-
.133
.158
.027
.004
.165
.055

ity, bimanual performance, and individualized outcomes28,29 or

2
h

across multiple functional performance measures.27 A previous

ANCOVA

.015
.008
.291
.670
.006
.125

follow-up RCT study also suggested that home-based CIT has

P

beneficial effects on unilateral grasping skills and unilateral/

bilateral functional performance at 6 months.13 The reinforced
6.44
7.88
1.15
0.18
8.32
2.44
F1,42

use of the more affected upper extremity by constraining the less

Posttreatment

NOTE. Descriptive statistics were presented as mean  SD.

affected upper extremity decreases neglect of the more affected

83.4116.08
0.530.02
0.590.03
0.240.04
67.503.06
8.660.94

one, thereby reducing developmental disregard and improving

TR (nZ22)

upper extremity skills.4,6 The improvement in upper extremity

skills might further enhance functional performance and self-care
activities in a child’s daily life.
The beneficial effects of home-based CIT over TR leading to
hCIT (nZ23)

better motor control strategies were retained for 3 to 6 months.

0.500.02
0.550.05
0.260.03
67.851.90
7.890.84
86.177.91

The home-based CIT group demonstrated shorter RTs and

normalized movement times and fewer normalized movement
units at the follow-ups than the TR group. The enhanced control
strategy after home-based CIT may be a result of the intensive
Kinematic Variables

practice of the more affected upper extremity during functional

nMU (times/mm)

tasks. The most likely explanation for this retention is that learned
nonuse and use-dependent cortical reorganization39 is overcome
nMT (s/mm)

PV (mm/s)

PMGA (%)
MGA (cm)

through enhanced motor control strategies. Cortical reorganization

Table 3

was maintained after CIT at the 6-month follow-up.40 Further-

RT (s)

more, the resulting increase in the use of the more affected upper
extremity may cause expansion of the contralateral cortical area

www.archives-pmr.org

Downloaded for John Carlo Conge (carloconge@gmail.com) at Our Lady of Fatima University from ClinicalKey.com by Elsevier on July 06, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
1430
controlling the movement of the more affected upper extremity
and recruitment of new ipsilateral areas.39
Home-based CIT induced better temporal efficiency in terms
of more preprogrammed and efficient movement execution stra-
tegies during both the short- and long-term phases than TR as
shown by the shorter RTs and normalized movement times for
reach-to-grasp tasks at posttreatment and both follow-ups. Lin
et al23 also demonstrated that CIT induced more efficient pre-
planning of reaching and grasping (shorter RTs) and a shift toward
feed-forward control (a higher percentage of movement time when
PV occurs) in patients with stroke than TR. The improvement in
the temporal efficiency is consistent with our findings of better
grasping skills and daily activities during both the short-term and
long-term phases after home-based CIT than after TR. The
intensive practice for the more affected upper extremity provided
by CIT may provide sufficient proprioceptive and visual feedback
to develop internal models for feed-forward movement con-
trol23,41,42 and learning effects for executing movement with
increased efficiency. Therefore, motor control changes may help
children with CP learn to preplan motor patterns and execute
motor tasks more efficiently after home-based CIT during both the
short- and long-term phases.
Home-based CIT induced better spatiotemporal efficiency by
increasing the long-term, rather than the short-term, movement
smoothness (fewer normalized movement units) during reach-
to-grasp tasks. The reason for this improvement may be that follow-
up visits are long enough to detect the differences in the movement
smoothness between the 2 groups. The results are compatible with
those in some studies involving patients with stroke,9,43 but they are
not consistent with those obtained by another stroke study.23 The
improvement in movement smoothness is consistent with our
finding of long-term, rather than short-term, enhanced movement
speed and dexterity, as measured by the BOTMP, after home-based
CIT compared with TR. The increase in movement smoothness
indicates a decreased amount of error correction,44,45 and move-
ment increasingly depends on feed-forward control.34,46
Home-based CIT enhanced spatial efficiency in hand grasping
(smaller MGA) during the reach-to-grasp tasks at posttreatment,
but not at the follow-ups, compared with TR. The improvement in
the spatial efficiency of hand grasping is partially consistent with
our finding of better grasping skills at both posttreatment and
follow-ups compared with TR. The findings that the long-term
effects of home-based CIT can be detected in PDMS-2, but not in
MGA, may be explained by the following. The increased grasping
skills at posttreatment can be maintained at the follow-ups
because the child had already developed these skills. However,
the inability to maintain the spatial efficiency of hand grasping
that is enhanced at posttreatment may be because of the lack of
continuing intensive and repetitive training of the more affected
upper extremity. This finding is compatible with the results of a
case report43 but is not consistent with that of a RCT involving
patients with stroke.23 A hand opened widely may maximize the
likelihood of successfully grasping an object.37 Our results that
children with CP achieved grasping movement without opening
the hand widely after the home-based CIT indicate a well pre-
planned or skillful grasping movement.23,37 The enhanced spatial
efficiency in hand grasping over the short term after home-based
CIT may be attributed to repeated functional practice and
shaping techniques for training the affected upper extremity.
Both home-based CIT and TR therapy generated comparable
performance in the online correction of target grasping (measured
by percentage of movement where MGA occurs), which is
Downloaded for John Carlo Conge (carloconge@gmail.com) at Our Lady of Fatima University from ClinicalKey.com by Elsevier on July 06, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
H.-c. Chen et al
compatible with a study of patients with stroke.23 In combination
with the results of MGA, our results suggest that the home-based
CIT protocol was more effective in promoting the child’s effi-
ciency to perform hand opening while grasping a ball but not in
promoting online correction during hand transportation. A
possible explanation is that the accuracy demands of the grasp-a-
ball task were relatively low because the ball is stationary and the
round shape is familiar to the children.23 The percentage of
movement where MGA occurs indicates the online correction
(temporal efficiency) during hand transportation, which may be
related to the accuracy demands of the task. However, the MGA
indicates the spatial efficiency of hand grasping, which may be
related to the child’s ability to perform wrist and finger extension
during grip formation. Children with CP may preserve the ability
to grasp a familiar target with little online correction of hand
posture during hand enclosure.23,47,48 Therefore, the grasp-a-ball
task may be too simple to detect between-group differences in the
percentage of movement where MGA occurs after the in-
terventions. The use of differently sized objects or unfamiliar
objects may be more applicable for differentiating between
possible changes after treatment.
Children in both groups performed comparably with respect to
force control (measured by PV) during the reach-to-grasp task
immediately after the intervention and during the follow-ups. A
high PV amplitude indicates a high force produced at movement
initiation and good force control while reaching toward a
target.45,49 The nonsignificant differences for PV amplitude be-
tween the 2 interventions may be because home-based CIT did not
focus on force production or control training.
The strength of this study protocol is the home-based inter-
vention implemented by a therapist. Home-based interventions
have advantages over clinic-based interventions. Compared with
the clinic-based group, the home-based group showed significant
improvement in motor proficiency and functional performance
after CIT in the follow-up sessions.14 The home-based CIT may
transfer greater training effects to daily life and induce greater
transfer package effects than clinic-based CIT. The transfer
package allows children to transfer the treatment gains to real-
world activities.8 In combination with the home environment,
home-based CIT facilitates the learning processes,16 enhances
daily function,23 promotes family participation, generalizes
certain therapeutic events to a child’s daily activities, and in-
creases the adherence of parents and therapists.50 The home-based
intervention may also facilitate the development of exploratory
behavior in children because the parents serve as a secure base for
exploration.51,52 In addition, this study provided the same treat-
ment dosage and one-on-one therapy sessions at home for both
groups. This study design may reduce bias from different treat-
ment dosages and enhance home-based CIT treatment effects in
natural environments for children with CP.
Considering the practical feasibility and compliance of the
children and caregivers, we chose 3.5 to 4h/d for the restraint of
the more affected upper extremity with a mitten. A systemic re-
view indicated the restraint time and treatment frequency of CIT
in children with CP varied among studies.9 The intervention time
for pediatric CIT varied from 1h/wk to 7h/d for 4 weeks to 8
months. Most school children with hemiplegia could not write or
complete homework after school with the more affected upper
extremity. School children often had 6 to 8 free hours after school.
Therefore, this study designated 3.5 to 4 hours for intensive
training and shaping of the more affected upper extremity, with
restraint of the less affected one. The children were allowed 1 to 3
www.archives-pmr.org
Effects of constraint induced therapy in cerebral palsy
hours for their school work with the less affected upper extremity.
In addition, a mitten, rather than splinting or a univalved or
bivalved cast, was selected as the restraint to enhance the
compliance because it is more convenient, lighter, and easier
to use.
Study limitations
Several study limitations should be considered. The inclusion
criteria included mild to moderate impairment of upper limb
function. Children with severely impaired upper limb function
were not included. Therefore, the study results cannot be gener-
alized to all children with CP. Using a uniform, familiar object (a
ball) as the grasping target to assess treatment efficacy may be
another limitation. Using smaller objects than that used in this
study may increase sensitivity for differentiating between possible
changes after treatment. Despite these limitations, this study
demonstrates the beneficial effects of home-based CIT on motor
control strategies and functional changes relative to those of TR.
Conclusions
Home-based CIT induced better spatial and temporal control
strategies underlying functional improvement when compared
with TR. Home-based CIT generated smoother movement, more
efficient hand grasping, and better movement preplanning and
execution than TR. Furthermore, the beneficial effects on the
motor control strategies, upper limb efficacy, and functional out-
comes were maintained for up to 6 months after treatment.
Suppliers
a. Vicon MX 3-D; Oxford Metrics Ltd, 14 Minns Estate, West
Way, Oxford OX2 0JB, UK.
b. National Instruments Corp, 11500 N MoPac Expwy, Austin,
TX 78759-3504.
c. SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606-6412.
Keywords
Cerebral palsy; Kinematics; Randomized controlled trial as topic;
Rehabilitation
Corresponding author
Chia-ling Chen, MD, PhD, Department of Physical Medicine and
Rehabilitation, Chang Gung Memorial Hospital, Linkou, 5 Fu-
hsing St, Kwei-shan, Tao-yuan 333, Taiwan. E-mail address:
clingchen@gmail.com.
References
1. Jones MW, Morgan E, Shelton JE, Thorogood C. Cerebral palsy:
introduction and diagnosis (part I). J Pediatr Health Care 2007;21:
146-52.
2. Rosenbaum P, Paneth N, Leviton A, et al. A report: the definition and
classification of cerebral palsy April 2006. Dev Med Child Neurol
Suppl 2007;109:8-14.
3. Hoare B, Imms C, Carey L, Wasiak J. Constraint-induced movement
therapy in the treatment of the upper limb in children with
www.archives-pmr.org
Downloaded for John Carlo Conge (carloconge@gmail.com) at Our Lady of Fatima University from ClinicalKey.com by Elsevier on July 06, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
1431
hemiplegic cerebral palsy: a Cochrane systematic review. Clin
Rehabil 2007;21:675-85.
4. Deluca SC, Echols K, Law CR, Ramey SL. Intensive pediatric
constraint-induced therapy for children with cerebral palsy: ran-
domized, controlled, crossover trial. J Child Neurol 2006;21:931-8.
5. Kuhtz-Buschbeck JP, Sundholm LK, Eliasson AC, Forssberg H. Quan-
titative assessment of mirror movements in children and adolescents with
hemiplegic cerebral palsy. Dev Med Child Neurol 2000;42:728-36.
6. Taub E, Ramey SL, DeLuca S, Echols K. Efficacy of constraint-
induced movement therapy for children with cerebral palsy with
asymmetric motor impairment. Pediatrics 2004;113:305-12.
7. Charles J, Gordon AM. A critical review of constraint-induced
movement therapy and forced use in children with hemiplegia.
Neural Plast 2005;12:245-61.
8. Taub E, Griffin A, Nick J, Gammons K, Uswatte G, Law CR. Pe-
diatric CI therapy for stroke-induced hemiparesis in young children.
Dev Neurorehabil 2007;10:3-18.
9. Huang HH, Fetters L, Hale J, McBride A. Bound for success: a
systematic review of constraint-induced movement therapy in chil-
dren with cerebral palsy supports improved arm and hand use. Phys
Ther 2009;89:1126-41.
10. Aarts PB, Jongerius PH, Geerdink YA, van Limbeek J, Geurts AC.
Modified constraint-induced movement therapy combined with
bimanual training (mCIMT-BiT) in children with unilateral spastic
cerebral palsy: how are improvements in arm-hand use established?
Res Dev Disabil 2011;32:271-9.
11. Aarts PB, Jongerius PH, Geerdink YA, van Limbeek J, Geurts AC.
Effectiveness of modified constraint-induced movement therapy in
children with unilateral spastic cerebral palsy: a randomized controlled
trial. Neurorehabil Neural Repair 2010;24:509-18.
12. de Brito Brandao M, Mancini MC, Vaz DV, Pereira de Melo AP,
Fonseca ST. Adapted version of constraint-induced movement ther-
apy promotes functioning in children with cerebral palsy: a ran-
domized controlled trial. Clin Rehabil 2010;24:639-47.
13. Lin KC, Wang TN, Wu CY, et al. Effects of home-based constraint-
induced therapy versus dose-matched control intervention on func-
tional outcomes and caregiver well-being in children with cerebral
palsy. Res Dev Disabil 2011;32:1483-91.
14. Rostami HR, Malamiri RA. Effect of treatment environment on
modified constraint-induced movement therapy results in children
with spastic hemiplegic cerebral palsy: a randomized controlled trial.
Disabil Rehabil 2012;34:40-4.
15. Al-Oraibi S, Eliasson AC. Implementation of constraint-induced
movement therapy for young children with unilateral cerebral palsy
in Jordan: a home-based model. Disabil Rehabil 2011;33:2006-12.
16. Eliasson AC, Krumlinde-sundholm L, Shaw K, Wang C. Effects of
constraint-induced movement therapy in young children with hemi-
plegic cerebral palsy: an adapted model. Dev Med Child Neurol
2005;47:266-75.
17. Eliasson AC, Shaw K, Berg E, Krumlinde-Sundholm L. An ecological
approach of constraint induced movement therapy for 2-3-year-old
children: a randomized control trial. Res Dev Disabil 2011;32:2820-8.
18. Takebayashi T, Koyama T, Amano S, et al. A 6-month follow-up
after constraint-induced movement therapy with and without trans-
fer package for patients with hemiparesis after stroke: a pilot quasi-
randomized controlled trial. Clin Rehabil 2013;27:418-26.
19. Hsin YJ, Chen FC, Lin KC, Kang LJ, Chen CL, Chen CY. Efficacy of
constraint-induced therapy on functional performance and health-
related quality of life for children with cerebral palsy: a random-
ized controlled trial. J Child Neurol 2012;27:992-9.
20. Chen CL, Kang LJ, Hong WH, Chen FC, Chen HC, Wu CY. Effect of
therapist-based constraint-induced therapy at home on motor control,
motor performance and daily function in children with cerebral
palsy: a randomized controlled study. Clin Rehabil 2013;27:236-45.
21. Wu CY, Chen CL, Tang SF, Lin KC, Huang YY. Kinematic and
clinical analyses of upper-extremity movements after constraint-
induced movement therapy in patients with stroke: a randomized
controlled trial. Arch Phys Med Rehabil 2007;88:964-70.
1432
22. Wu CY, Lin KC, Chen HC, Chen IH, Hong WH. Effects of modified
constraint-induced movement therapy on movement kinematics and
daily function in patients with stroke: a kinematic study of motor
control mechanisms. Neurorehabil Neural Repair 2007;21:460-6.
23. Lin KC, Wu CY, Wei TH, Lee CY, Liu JS. Effects of modified
constraint-induced movement therapy on reach-to-grasp movements
and functional performance after chronic stroke: a randomized
controlled study. Clin Rehabil 2007;21:1075-86.
24. Folio MR, Fewel RR. Peabody Developmental Motor Scales. 2nd ed.
Austin: ProEd; 2000.
25. Wallen M, Bundy A, Pont K, Ziviani J. Psychometric properties of
the Pediatric Motor Activity Log used for children with cerebral
palsy. Dev Med Child Neurol 2009;51:200-8.
26. Bruininks RH. Bruininks-Oseretsky Test of Motor Proficiency. Ex-
aminer’s manual. Circle Pines: American Guidance Service; 1978.
27. Case-Smith J, DeLuca SC, Stevenson R, Ramey SL. Multicenter
randomized controlled trial of pediatric constraint-induced move-
ment therapy: 6-month follow-up. Am J Occup Ther 2012;66:15-23.
28. Sakzewski L, Ziviani J, Abbott DF, Macdonell RA, Jackson GD,
Boyd RN. Equivalent retention of gains at 1 year after training with
constraint-induced or bimanual therapy in children with unilateral
cerebral palsy. Neurorehabil Neural Repair 2011;25:664-71.
29. Sakzewski L, Ziviani J, Abbott DF, Macdonell RA, Jackson GD,
Boyd R. Randomized trial of constraint-induced movement therapy
and bimanual training on activity outcomes for children with
congenital hemiplegia. Dev Med Child Neurol 2011;53:313-20.
30. Ottenbacher KJ, Taylor ET, Msall ME, et al. The stability and
equivalence reliability of the functional independence measure for
children (WeeFIM). Dev Med Child Neurol 1996;38:907-16.
31. Eliasson AC, Gordon AM, Forssberg H. Impaired anticipatory con-
trol of isometric forces during grasping by children with cerebral
palsy. Dev Med Child Neurol 1992;34:216-25.
32. Gordon AM, Charles J, Wolf SL. Efficacy of constraint-induced
movement therapy on involved upper-extremity use in children
with hemiplegic cerebral palsy is not age-dependent. Pediatrics 2006;
117:e363-73.
33. Himmelmann K, Beckung E, Hagberg G, Uvebrant P. Gross and fine
motor function and accompanying impairments in cerebral palsy.
Dev Med Child Neurol 2006;48:417-23.
34. Brooks VB, Watts SL. Adaptive programming of arm movements. J
Mot Behav 1988;20:117-32.
35. Levin MF. Interjoint coordination during pointing movements is
disrupted in spastic hemiparesis. Brain 1996;119:281-93.
36. Trombly CA. Deficits of reaching in subjects with left hemiparesis: a
pilot study. Am J Occup Ther 1992;46:887-97.
Downloaded for John Carlo Conge (carloconge@gmail.com) at Our Lady of Fatima University from ClinicalKey.com by Elsevier on July 06, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
H.-c. Chen et al
37. Grosskopf A, Kuhtz-Buschbeck JP. Grasping with the left and right
hand: a kinematic study. Exp Brain Res 2006;168:230-40.
38. Cohen J. Statistical power analysis for the behavioral sciences. 2nd
ed. Hillsdale: Lawrence Erlbaum Associates; 1988.
39. Morris DM, Taub E. Constraint-induced therapy approach to
restoring function after neurological injury. Top Stroke Rehabil 2001;
8:16-30.
40. Sutcliffe TL, Gaetz WC, Logan WJ, Cheyne DO, Fehlings DL.
Cortical reorganization after modified constraint-induced movement
therapy in pediatric hemiplegic cerebral palsy. J Child Neurol 2007;
22:1281-7.
41. Desmurget M, Grafton S. Forward modeling allows feedback control
for fast reaching movements. Trends Cogn Sci 2000;4:423-31.
42. Sabes PN. The planning and control of reaching movements. Curr
Opin Neurobiol 2000;10:740-6.
43. Hakim RM, Kelly SJ, Grant-Beuttler M, Healy B, Krempasky J,
Moore S. Case report: a modified constraint-induced therapy (mCIT)
program for the upper extremity of a person with chronic stroke.
Physiother Theory Pract 2005;21:243-56.
44. Kamper DG, McKenna-Cole AN, Kahn LE, Reinkensmeyer DJ.
Alterations in reaching after stroke and their relation to movement
direction and impairment severity. Arch Phys Med Rehabil 2002;83:
702-7.
45. Ma HI, Trombly CA, Tickle-Degnen L, Wagenaar RC. Effect of one
single auditory cue on movement kinematics in patients with Par-
kinson’s disease. Am J Phys Med Rehabil 2004;83:530-6.
46. Nagasaki H. Asymmetric velocity and acceleration profiles of human
arm movements. Exp Brain Res 1989;74:319-26.
47. Harvey M, Jackson SR, Newport R, Kramer T, Morris DL, Dow L. Is
grasping impaired in hemispatial neglect? Behav Neurol 2001;13:
17-28.
48. Michaelsen SM, Jacobs S, Roby-Brami A, Levin MF. Compensation
for distal impairments of grasping in adults with hemiparesis. Exp
Brain Res 2004;157:162-73.
49. Wu CY, Chuang LL, Lin KC, Chen HC, Tsay PK. Randomized trial
of distributed constraint-induced therapy versus bilateral arm training
for the rehabilitation of upper-limb motor control and function after
stroke. Neurorehabil Neural Repair 2011;25:130-9.
50. Naylor CE, Bower E. Modified constraint-induced movement therapy
for young children with hemiplegic cerebral palsy: a pilot study. Dev
Med Child Neurol 2005;47:365-9.
51. Thompson RA. The legacy of early attachments. Child Dev 2000;71:
145-52.
52. Waters E, Cummings EM. A secure base from which to explore close
relationships. Child Dev 2000;71:164-72.
www.archives-pmr.org