Parkinsonism and Related Disorders 41 (2017) 92e98

Contents lists available at ScienceDirect

Parkinsonism and Related Disorders

journal homepage: www.elsevier.com/locate/parkreldis

Home based training for dexterity in Parkinson's disease: A

randomized controlled trial
Tim Vanbellingen a, b, c, *, Thomas Nyffeler a, b, Julia Nigg a, Jorina Janssens d,
Johanna Hoppe d, Tobias Nef b, Rene  M. Müri b, Erwin E.H. van Wegen c, e,
Gert Kwakkel , Stephan Bohlhalter a
c, e

a
Neurocenter, Luzerner Kantonsspital, Switzerland
b
Gerontechnology and Rehabilitation Group, University of Bern, Bern, Switzerland
c
Dept. of Rehabilitation Medicine, Amsterdam Movement Sciences, VU University Medical Center, Amsterdam, The Netherlands
d
Neurorehabilitation Center, Klinik Bethesda Tschugg, Switzerland
e
Amsterdam Neuroscience, VU University Medical Center, Amsterdam, The Netherlands

a r t i c l e i n f o a b s t r a c t

Article history: Background: Patients with Parkinson's disease exhibit disturbed manual dexterity. This impairment leads
Received 31 January 2017 to difficulties in activities of daily living, such as buttoning a shirt or hand-writing. The aim of the present
Received in revised form study was to investigate the effectiveness of a home-based dexterity program on fine motor skills in a
14 May 2017
single-blinded, randomized controlled trial, in patients with Parkinson's disease.
Accepted 23 May 2017
Methods: One hundred and three patients with Parkinson's disease (aged between 48 and 80 years,
Hoehn & Yahr stage I-IV) were randomized to either a home-based dexterity program (HOMEDEXT) or
Keywords:
Thera-band program. All patients trained over a period of 4 weeks, 5 times/week, 30 min for each
Randomized controlled trial
Home based training
session. A baseline, post-intervention, and follow-up assessment (12 weeks later, time period without
Manual dexterity intervention) were done. The primary outcome measure was dexterity as measured with the Nine Hole
Parkinson's disease peg test (9-HPT). Secondary outcome measures included strength, motor parkinsonian symptoms,
Nine Hole Peg test dexterity-related activities of daily living (ADL) and Health-related Quality of Life (HrQoL).
Results: There was a significant difference in favor of the HOMEDEXT group as compared to the Thera-
band group on the primary outcome 9-HPT (p ¼ 0.006) and dexterity-related ADL (p ¼ 0.02) at post
intervention. No significant differences were found for the other outcomes, nor at follow-up.
Conclusion: This is the first randomized controlled trial showing that an intensive, task specific home-
based dexterity program significantly improved fine motor skills in Parkinson's disease. The effect
generalized to dexterity-related ADL functions. As these improvements did not sustain, the finding
suggest that continuous training is required to maintain the benefit.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction require fine motor skills, such as buttoning a shirt or hand-writing,

Patients with Parkinson's disease (PD) frequently exhibit
disturbed manual dexterity even in early stages of the disease
[1e4]. While dopaminergic treatment improves cardinal symptoms
of PD, such as bradykinesia, impaired dexterity may be less
responsive to pharmacological treatment [1]. Dexterous impair-
ment leads to difficulties in activities of daily living (ADL) that
* Corresponding author. Neurocenter, Luzerner Kantonsspital, Switzerland.
E-mail addresses: tim.vanbellingen@luks.ch, tim.vanbellingen@dkf.unibe.ch
(T. Vanbellingen).
and therefore contribute to the burden of the disease [2]. The exact
mechanisms for the loss of dexterity are unknown. Elementary
motor deficits such as bradykinesia [3], reduced strength and
impaired finger torque production [5] certainly play a role. How-
ever, these deficits do not fully account for the impairment in fine
motor skills. Several authors suggest that an apraxic disorder, called
‘limb kinetic apraxia’, may significantly contribute to the dexterous
deficits observed in patients with PD [1,3,4].
Unfortunately, there is limited data on therapeutic interventions
of dexterous problems in PD [6]. There is growing evidence that
intensity and task-specificity of practice delivered by physical- (PT)
and/or occupational therapists (OT) may be effective and

http://dx.doi.org/10.1016/j.parkreldis.2017.05.021
1353-8020/© 2017 Elsevier Ltd. All rights reserved.
 T. Vanbellingen et al. / Parkinsonism and Related Disorders 41 (2017) 92e98
compliment to pharmacological and surgical treatments [7e10].
Notably, task specific training improves the patient's abilities in
ADL and increases levels of participation [8,10]. Activity-dependent
neuroplasticity likely explains the effectiveness of such programs as
shown by short and long-term effects of balance and gait training in
PD [8e10]. Therefore, intensity, specificity, difficulty, and
complexity of practice appear to be driving parameters for neuro-
restorative effects on brain plasticity [11]. In contrast to the sig-
nificant short and long-term improvements of training for gait and
balance, no randomized controlled trial (RCT) has been conducted
to investigate the effectiveness of a specific dexterity training in PD
so far. Therefore, the aim of this study was to explore short-term
and long-term effectiveness of a task-specific home-based dexter-
ity program (HOMEDEXT) in PD. We hypothesized that HOMEDEXT
training would improve fine motor skills both in the short term and
the longterm as compared with Thera-band training. In addition,
we expected a generalization of the HOMEDEXT program to ADL
functioning and Health-related Quality of Life.
2. Methods
2.1. Design
This was an observer-blinded randomized controlled trial (RCT)
with post-intervention measurements at four weeks and 12 weeks.
An independent investigator not involved in the treatment protocol
executed the randomization procedure using a computerized
randomization protocol. Treatment allocation was concealed for
the trained observers by using small, opaque cardboard boxes.
Three trained occupational therapists having longstanding experi-
ence in the field of PD rehabilitation were responsible for the as-
sessments and instruction of both interventions. They were
unaware of the randomization and allocation procedure. All
outcome measures (baseline, post-intervention and follow-up
measures) were videotaped and rated by an independent rater
blinded to group allocation. Every patient was assessed at the same
time of day in the ON phase, approximately 90 min after drug
intake. Patients were advised to have their medical treatment
continued unchanged throughout the study.
2.2. Participants
Patients with PD, as defined by the UK Parkinson's Disease So-
ciety Brain Bank Criteria [12], with Hoehn and Yahr stages I to IV,
aged > 18, 80, and with stable drug usage, were recruited from the
Klinik Bethesda Tschugg as well as the Neurology and Neuro-
rehabilitation Center, Luzerner Kantonsspital, from November 2013
to January 2016. Patients were excluded if they showed significant
physical or psychiatric co-morbidity including dementia as defined
by a Montreal Cognitive Assessment (MOCA) score < 21 [13].
Furthermore, patients not being able to complete the question-
naires (i.e. due to cognitive problems), or those participating in
another intervention trial were also excluded. Handedness was
assessed by the Edinburgh handedness inventory [14]. Prior to
study participation, written informed consent was obtained from
all patients according to the latest Declaration of Helsinki. Ethical
approval was given by the Ethics committee of the State of Bern
(KEK 131/13) Bern, Switzerland. Trial registration was Clinicaltrials.
gov NCT02297893. The study conformed to the CONSORT
(Consolidated Standards of Reporting Trials) statement, http://
www.consort-statement.org.
2.3. Interventions
Two home based treatment conditions were investigated in this
93
study: the experimental condition (HOMEDEXT) and the control
condition (Thera-band). Patients were instructed (total instruction
time approximately thirty minutes) by an occupational therapist
(JN, JJ, or JH) on how to perform the allocated home based training
program. Patients of both training groups trained 5 days per week
(approximately 30 min per day) for 4 weeks. All exercises were
performed with both hands. Patients received a booklet explaining
the exercises with pictures and a short text. This booklet also
contained a diary to document if an exercise was performed and, if
applicable, the time needed to accomplish the exercise. After 1
week of training, patients were contacted by phone call and asked
in a standardized way whether there were any queries related to
their exercises. In this way, the occupational therapist could further
check for adherence to the protocol.
2.3.1. Dexterity intervention (HOMEDEXT)
The HOMEDEXT program was conceptualized to target key
components of manual dexterity: (1) Fine-tuned control of force,
such as in precision grip when picking up small objects (2) Finger
independence, i.e. the ability to move the fingers selectively, (3)
finger coordination, i.e. the ability to synchronize finger move-
ments, and (4) motor sequence performance, i.e. being able to
activate fingers in a temporal sequence [15]. Some of the specific
tasks were adapted from an existing standardized dexterity pro-
gram [16]. In addition, all exercises were aligned with OT expert
opinion [6]. Patients performed six different exercises, either uni-
manually or bi-manually, as illustrated in detail in supplemental
file 1. Both hands were trained equally.
2.3.2. Control intervention (Thera-band)
PD patients allocated to the control intervention received upper
limb Thera-band-exercises. It has been shown that hand strength
can improve, by means of resistance training with the use of elastic
bands (Thera-band) [17]. These exercises were taken from an
existing standardized training protocol [16]. Patients performed
seven upper extremity strength-training exercises using a Thera-
band®. The exercises with the Thera-band are shown in supple-
mental file 2. In this Thera-band program, both upper limbs were
trained in an equal amount.
2.4. Outcome measures
Primary outcome. Finger and hand function was measured by the
Nine Hole Peg test (9-HPT). The 9HPT is a standardized measure of
hand function and is validated in PD [18]. In a large group of PD
patients, minimal detectable changes have been established: 2.6 s
and 1.3 s for the dominant and non-dominant hands, respectively
[18]. During the 9-HPT patients were seated at a table with a
shallow container holding nine pegs and a plastic block with nine
empty holes. All pegs had to be put into the holes one at a time and
then removed again one at a time into the shallow container. The
time to complete the task was recorded for each hand separately,
and the mean values of the two trials are taken for each hand. The
mean value of both left and right hands was the main primary
outcome.
Secondary outcomes. Health-related Quality of Life was assessed
by a modified version (items 11 to 16 related to fine motor skills) of
the Parkinson's Disease Questionnaire 39 (PDQ-39) [19]. In addi-
tion, a recently validated patient-reported dexterity questionnaire,
DextQ-24, was administered [20]. This questionnaire contains 24
questions, which are divided into five subgroups (“washing/
grooming”; “dressing”; “meals and kitchen”; “everyday tasks”; “TV/
CD/DVD”). For each question, patients had to state whether they
have no problems (1 points), minor problems (2 points), major
problems (3 points), or need aid (4 point) to perform the task.
94 T. Vanbellingen et al. / Parkinsonism and Related Disorders 41 (2017) 92e98
Points are added in each subgroup and summed to a total score.
Score ranged from a minimum of 24 to a maximum of 96 points.
The coin rotation task (CRT) was used to measure coordinated
finger movements. The CRT has been used in a several studies on
dexterity in PD [1,3,4]. For the CRT, patients were instructed to
rotate a Swiss 20-Rappen coin (size, diameter ¼ 20.9 mm;
thickness ¼ 1.6. mm; weight ¼ 4.05 g) between their thumb, index
and middle finger as rapidly as possible. Three trials were reques-
ted, each lasting 10 s. The number of half turns is calculated in
accordance to the following formula: CRT score ¼ half turns e
[(coin drops x 0.1) x half turns].
Isometric hand strength was measured using the JAMAR dyna-
mometer (Sammons Preston Rolyan, 1000 Remington Blvd,
Bolingbrook, IL, 60440). This measurement is a reliable (ICC values
0.85e0.98) and valid test to measure isometric hand strength in
healthy subjects and PD [21]. It was performed in an upright seated
position with 90 flexion of the elbow next to the body. Three
maximum voluntary hand strength movements were performed,
and the mean value in kilograms force was used.
ADL, related to fine motor skills, was measured by a modified
version of the Movement Disorders Society unified Parkinson's
disease rating scale (MDS-UPDRS) subscale II, including items
2.4e2.7 (eating and dressing tasks, personal hygiene and hand
writing) [22]. Parkinsonian motor symptoms were assessed by a
modified version of motor examination subscale III from the MDS-
UPDRS [22]. This modified version consists of the upper limb items,
of which summed scores were calculated: 3.3 (rigidity), 3.4 (index
finger tapping on the thumb 10 times as quickly and as wide as
possible), 3.5 (tight fist and opening the hand 10 times as fully open
Fig. 1. CONSORT diagram demonstrating study flow.
and as quickly as possible), 3.6 (pronation-supination of the hand
10 times as fast and completely as possible), 3.15 (postural tremor),
3.16 (kinetic tremor), 3.17 (rest tremor amplitude) and 3.18 (con-
stancy of rest tremor). This modified version has been used previ-
ously and proved to be valid [3].
2.5. Statistical analysis
A between-group differences 2 (Groups) x 2 (types of therapy)
analysis of covariance tests (ANCOVA) was performed. Baseline
scores were used as covariates. Two-sided 95% confidence intervals
(CI) were calculated, as well as effect size (Cohen's d) [23]. A series
of planned comparisons using the t-statistic were used to investi-
gate improvements from baseline (t0) to the end of the interven-
tion (t1) and to follow up (t2) in each group. The
BenjaminieHochberg procedure was applied to control the false
discovery rate [24]. Descriptive statistics were used to present
baseline characteristics and results of outcome measurements.
Missing values were imputed by means of last observation carried
forward. According to the Intention to treat (ITT) principle every
randomized patient, including the drop-outs, was included for final
evaluation. For all analyses the level of significance was set at
p ¼ 0.05 (two-tailed). Statistical analyses were performed using
PASW for Windows (version 23.0; SPSS, Inc. Chicago, IL). Sample
size calculation was based on the significance level alpha, two-
tailed at p-value of 0.05, to detect 15% difference in 9HPT perfor-
mance between the groups. Based on mean expected baseline
values of 29.1 þ- 7.3 (pooled mean and standard deviations values)
of the whole group, a 15% difference would reflect 4.3 s, which
 T. Vanbellingen et al. / Parkinsonism and Related Disorders 41 (2017) 92e98 95

exceeds the minimal detectable change (MDC) of 2.6 s found in the
literature [18]. Following Cohen's guidelines the probability to
prevent type II error was set at 0.20 assuming a 1-beta value of 0.80
[23]. The power analysis, with the above mentioned parameters,
resulted in a total sample size of 90 participants, i.e. 45 per group.
Taking into account a drop-out rate of 15% we aimed to recruit at
least 100 patients in total.
3.1. Primary outcome
At post-intervention PD patients with HOMEDEXT training
(p ¼ 0.006) performed significantly faster in 9-HPT as compared to
the PD patients that received Thera-band training. The difference
between both groups did not sustain at follow-up (p ¼ 0.51).
3.2. Secondary outcomes

HOMEDEXT training also significantly (p ¼ 0.02) improved

3. Results
The flow of the participants throughout the whole trial is pre-
sented in Fig. 1. Patients were mostly excluded because of cognitive
problems (n ¼ 55), the presence of psychiatric co-morbidity
(n ¼ 24), no commitment to the study protocol (n ¼ 14), partici-
pation in another study (n ¼ 10), instable drug use (n ¼ 8).
A total of 103 patients (52 in the HOMEDEXT group and 51 in the
Thera-band group) were recruited. The overall drop-out rate was
13%, and no significant difference (c2 ¼ 2.3, p ¼ 0.13) with respect
to the drop rate between both groups was found. There were no
significant baseline differences between the groups, except for the
secondary outcome measure CRT right (Table 1).
Based on the diaries, adherence to the training programs was
high with an average of 88% of the performed exercises in the
HOMEDEXT and 84% in the Thera-band group. The total training
time (5 times per week, for 4 weeks) was consistent and did not
differ between the groups (602 ± 30.5 min for HOMEDEXT vs
588 ± 36.2 min for Thera-band, p ¼ 0.78). In addition, there were no
reported adverse events in both groups.
The changes of the primary and secondary outcomes from
baseline (t0) to post-intervention (t1) and follow-up (t2) for each
group (HOMEDEXT and Thera-band) and the between group dif-
ferences adjusted for baseline values are presented in detail in
Tables 2 and 3.
dexterity related ADL, according to the DextQ-24, in PD patients as
compared to those who received Thera-band training. This effect
was not present at follow-up. For the other secondary outcomes, no
significant group differences were found between the HOMEDEXT
and Thera-band group at post-intervention and at follow-up. With
respect to the ADL subscale of the PDQ-39, there was only a trend in
favor of the HOMEDEXT training (p ¼ 0.08).
4. Discussion
Dexterous difficulties may considerably impede manual ADL
function in PD, even at early stages of the disease [4]. Impaired
dexterity is typically less responsive to dopaminergic treatment [1].
As a consequence, the development of standardized rehabilitation
protocols for dexterity in PD is an important need. Herein, we
developed a home-based program (HOMEDEXT) that is easily
implemented in the daily routine, because it only requires thirty
minutes training, 5 times per week, over a period of four weeks. In
addition, the use of items is straightforward. Task specificity of the
HOMEDEXT program has been achieved by including exercises that
target distinct components of dexterity such as fine-tuned force
control and coordinated finger movements. Finally, the home-
based approach may generally be more cost effective as supervi-
sion by specialized therapists is minimized.
This present proof-of-concept trial showed that HOMEDEXT

Table 1
Demographic and baseline characteristics of PD patients allocated to HOMEDEXT, Thera-band intervention.

HOMEDEXT-Group Theraband-Group P-value

n ¼ 52 n ¼ 51

Patients' characteristics
Age (years) 67.15 ± 7.94 68.16 ± 7.38 0.51
Gender, female/male (% female) 18/34 (35%) 22/29 (43%) 0.38
Handedness, right/left (%) 51/1 (98%) 49/2 (96%) 0.55
Disease duration (y) 6.12 ± 3.52 6.35 ± 3.99 0.75
H&Y stage 1.94 ± 0.90 2.00 ± 0.82 0.73
Levodopa equivalent 741.63 ± 471.8 745.43 ± 502.69 0.97
MOCA 26.65 ± 1.78 26.59 ± 2.30 0.87
Primary outcome
9-HPT_mean both hands, sec 29.83 ± 7.09 30.82 ± 7.81 0.50
Left 31.20 ± 8.48 31.26 ± 9.50 0.97
Right 28.42 ± 7.19 30.37 ± 7.38 0.18
Secondary outcomes
PDQ39_Total 24.52 ± 12.30 26.48 ± 14.71 0.46
PDQ39_ADL 30.77 ± 19.16 30.77 ± 19.16 0.24
DextQ-24 37.73 ± 8.75 38.33 ± 10.79 0.76
CRT_mean both hands, sec 8.41 ± 3.21 7.31 ± 3.22 0.09
Left 7.78 ± 3.63 7.22 ± 3.77 0.44
Right 9.04 ± 3.53 7.41 ± 3.66 0.02
JAMAR_mean both hands, kg 28.05 ± 7.93 27.75 ± 8.81 0.85
Left 26.77 ± 8.25 26.95 ± 8.88 0.92
Right 29.34 ± 8.08 28.78 ± 9.26 0.75
Mod.MDS-UPDRS II_Total 4.73 ± 2.34 5.31 ± 2.93 0.27
Mod.MDS-UPDRS III_Total 15.67 ± 5.16 16.12 ± 6.78 0.71

Values are means ± standard deviation or as otherwise indicated; n: number of patients; y: years; H&Y: Hoehn and Yahr; MOCA: Montreal Cognitive Assessment; 9HPT:
Nine Hole Peg Test; sec: seconds; PDQ39: Parkinson disease questionnaire 39; DextQ-24: Dexterity Questionnaire 24; ADL: Activities of Daily Living; CRT: Coin Rotation
Task; kg: kilogram; Mod.MDS-UPDRS II & III: modified Movement Disorders Society Unified Parkinson disease Rating Scale II & III.
96 T. Vanbellingen et al. / Parkinsonism and Related Disorders 41 (2017) 92e98

Table 2
Efficacy endpoints. Changes baseline to post intervention.

HOMEDEXT-Group Theraband-Group Adjusted Group Difference HOMEDEXT e Theraband (95% CI) P* d

Change (t1-t0) Pþ Change (t1-t0) P#

Primary outcome
9-HPT 2.11 ± 4.20 0.001 0.58 ± 6.83 0.58 3.01 (0.87e5.15) 0.006 0.47
Left 2.75 ± 4.83 <0.001 0.90 ± 6.68 0.39 3.46 (1.20e5.71) 0.003 0.63
Right 1.56 ± 4.89 0.03 0.89 ± 6.99 0.40 3.03 (0.65e5.41) 0.01 0.41
Secondary outcomes
PDQ39_Total 4.15 ± 6.26 <0.001 2.84 ± 6.91 0.009 1.65 (0.88 to 4.18) 0.19 0.20
PDQ39_ADL 7.99 ± 13.06 <0.001 4.92 ± 10.60 0.004 3.89 (0.53 to 8.31) 0.08 0.26
DextQ-24 3.22 ± 5.00 <0.001 0.71 ± 5.43 0.39 2.39 (0.33e4.44) 0.02 0.48
CRT 0.56 ± 1.32 0.004 0.35 ± 2.15 0.29 0.35 (1.06 to 0.36) 0.33 0.12
Left 0.53 ± 1.79 0.04 0.16 ± 2.13 0.64 0.41 (1.18 to 0.37) 0.30 0.19
Right 0.60 ± 1.56 0.01 0.41 ± 2.47 0.28 0.45 (1.28 to 0.38) 0.28 0.09
JAMAR 0.30 ± 3.13 0.50 0.43 ± 2.70 0.30 0.12 (1.10 to 1.33) 0.85 0.04
Left 0.54 ± 3.53 0.29 0.71 ± 2.91 0.12 0.18 (1.17 to 1.54) 0.79 0.05
Right 0.07 ± 3.37 0.89 0.18 ± 3.23 0.71 0.07 (1.28 to 1.43) 0.91 0.03
MDS-UPDRS II 0.59 ± 1.43 0.006 0.27 ± 0.90 0.05 0.34 (0.15 to 0.84) 0.17 0.27
MDS-UPDRS III 0.10 ± 2.69 0.79 0.46 ± 3.30 0.37 0.35 (1.59 to 0.88) 0.57 0.12

Values are mean ± SD or as otherwise indicated; n: number of patients; sec: seconds; 9HPT: Nine Hole Peg Test CRT: PDQ39: Parkinson disease questionnaire 39; CRT: Coin
Rotation Task; Mod.MDS-UPDRS II & III: modified Movement Disorders Society Unified Parkinson disease Rating Scale II & II; þp-value change baseline to post-intervention
HOMEDEXT-Group; #p-value change baseline to post-intervention Theraband-Group; *p-value change baseline to post-intervention HOMEDEXT-Group vs. Theraband-Group.
Cohen's d; Adjusted Group Difference HOMEDEXT e Theraband: This is the difference in the change of the outcome measure between both groups (HOMEDEXT e Theraband
group) controlled for the baseline value of the corresponding outcome measure; 95% CI: 95% Confidence interval.

Table 3
Efficacy endpoints. Changes baseline to follow up.

HOMEDEXT-Group Theraband-Group Adjusted Group Difference HOMEDEXT e Theraband (95% CI) P* d

Change (t2-t0) Pþ Change (t2-t0) P#

Primary outcome
9-HPT_mean 0.47 ± 4.15 0.43 0.12 ± 4.37 0.86 0.58 (1.14 to 2.30) 0.51 0.08
Left 1.25 ± 4.95 0.09 0.27 ± 5.16 0.74 1.02 (0.98 to 3.02) 0.31 0.19
Right 0.25 ± 5.29 0.74 0.04 ± 4.28 0.95 0.18 (1.81 to 2.18) 0.86 0.06
Secondary outcomes
PDQ39_Total 3.26 ± 8.74 0.01 2.35 ± 6.35 0.02 1.50 (1.81 to 4.82) 0.37 0.12
PDQ39_ADL 5.21 ± 14.78 0.02 4.66 ± 9.44 0.003 1.28 (3.54 to 6.10) 0.60 0.04
DextQ-24 2.56 ± 6.04 0.005 0.79 4.75± 0.29 1.48 (0.60 to 3.56) 0.16 0.33
CRT 0.49 ± 2.03 0.10 0.63 ± 1.71 0.02 0.01 (0.77 to 0.80) 0.98 0.07
Left 0.64 ± 2.03 0.03 0.53 ± 2.18 0.13 0.13 (1.0 to 0.74) 0.76 0.05
Right 0.35 ± 2.49 0.34 0.54 ± 1.31 0.01 0.004 (0.86 to 0.85) 0.99 0.10
JAMAR 0.14 ± 3.74 0.79 0.35 ± 2.75 0.41 0.47 (1.84 to 0.89) 0.49 0.15
Left 0.13 ± 4.08 0.83 0.42 ± 3.35 0.42 0.24 (1.83 to 1.34) 0.76 0.08
Right 0.41 ± 4.12 0.49 0.36 ± 3.68 0.53 0.77 (2.35 to 0.82) 0.34 0.20
MDS-UPDRS II 0.23 ± 1.42 0.27 0.14 ± 1.03 0.37 0.10 (0.42 to 0.63) 0.69 0.07
MDS-UPDRS III 0.06 ± 3.32 0.90 0.07 ± 2.63 0.86 0.04 (1.27 to 1.19) 0.95 0.003

Values are mean ± SD or as otherwise indicated; n: number of patients; sec: seconds; 9HPT: Nine Hole Peg Test CRT: PDQ39: Parkinson disease questionnaire 39; CRT: Coin
Rotation Task; Mod.MDS-UPDRS II & III: modified Movement Disorders Society Unified Parkinson disease Rating Scale II & II; þp-value change baseline to post-intervention
HOMEDEXT-Group; #p-value change baseline to post-intervention Theraband-Group; *p-value change baseline to post-intervention HOMEDEXT-Group vs. Theraband-Group.
Cohen's d; Adjusted Group Difference HOMEDEXT e Theraband: This is the difference in the change of the outcome measure between both groups (HOMEDEXT e Theraband
group) controlled for the baseline value of the corresponding outcome measure; 95% CI ¼ 95% Confidence interval.

significantly improved dexterous function (9-HPT) and self-
reported ADL (DextQ-24) in patients with PD when compared
with an unspecific Thera-band program of the upper limb.
Furthermore, the positive effect on 9-HPT but not MDS-UPDRS part
III speaks for the specificity of the finding on dexterous function,
which is much less captured by the motor score of the MDS-UPDRS
III. Similarly, HOMEDEXT improved DextQ-24, but not MDS-UPDRS
II. DextQ-24 has been developed and validated to specifically cover
different dexterity related ADL [20], whereas the modified MDS-
UPDRS II is more symptom specific monitoring of ADL relevant
arm functions. For the 9-HPT, the benefits exceeded the minimal
detectable change [18]. It generally exceeded the clinically mean-
ingful 10% reported for other programs [9], although being below
the expected 15% change based on our power calculation. A
possible reason for this discrepancy may be the lower variability
(lower standard deviations) of the primary outcome 9-HPT scores
than actually expected. However, the effects did not sustain three
months after completion of the training. This finding is in line with
previous RCT's [8,9]. Both programs improved quality of life, sup-
porting the beneficial effects of exercise in general, as demon-
strated previously [8].
According to our knowledge this is the first and largest RCT
demonstrating a significant effect for home-based dexterity
training on motor skill in PD. Furthermore, the generalization to
self-reported dexterity-related ADL such as writing a shopping list
or to dial on a mobile phone underscored the clinical relevance of
the findings for neurorehabilitation. So far, evidence-based thera-
pies are missing mainly due to lack of methodologically well-
conducted trials in this field [6]. Our findings are in agreement
with previous home-based interventions demonstrating significant
short-term effects for other functions such as parkinsonian motor
symptoms [25], physical and psychological health [26]. The lack of
 T. Vanbellingen et al. / Parkinsonism and Related Disorders 41 (2017) 92e98
long term improvement of the HOMEDEXT has been reported for
other home-based training programs [9,25], suggesting that for
home rehabilitation generally continuous training is needed to
maintain function. The intensity of the HOMEDEXT training could
have been also too low for some patients, explaining the absence of
long term effects.
Adherence to the protocol was high, being on average 86%,
which is comparable with figures reported in the literature [27].
The main reasons why patients could not fully complete their
protocol were fatigue, daily stress, and lack of motivation or
sickness.
Our findings result from PD patients with good cognitive func-
tions, who were able to perform the home-based training without
caregiver support. For cognitively impaired patients the present
protocol remains to be evaluated, which is important as difficulties
with fine motor skills are even more prevalent [28]. We expect that
patients with more pronounced cognitive deficits will need more
supervision by their caregivers to stick to the home-based protocol.
Future research may also address how specific training of fine
motor skill improving dexterity stimulates brain plasticity in PD.
Interestingly, in healthy subjects training of repetitive sequential
finger movement not only improved the practised task but also
generalized to other non-related finger tasks, possibly by increasing
cortical sensorimotor representations [29].
A limitation of the study may be that the supervision during the
home-based training was done not on a weekly basis, but only after
the first week of training. Therefore, it was difficult to control
whether all patients actually performed their exercises as pre-
scribed in the booklet. However, the patients generally appreciated
the well-structured protocol, which included simple explanations
and pictures, rendering the program highly comprehensible.
Furthermore, the diaries of the training in almost all patients sug-
gested high therapeutic adherence.
In future studies, online monitoring using tele-rehabilitation
services may be evaluated for home-based training. Such e-health
techniques would not only allow to give regular feedback to the
patients but also to increasingly adapt the difficulty of the tasks.
Specifically, wrist worn sensors, that measure the amount and
quality of hand use, would allow to explore a dose response rela-
tionship between dexterity training and outcome during ADL [30].
5. Conclusion
In sum, we demonstrated that the HOMEDEXT program is
effective in improving dexterity and dexterity related ADL in
cognitively non-impaired PD patients. The present findings suggest
that the comprehensible and easy implementable protocol allowed
successful home training of dexterity without extensive instruction,
or continuous face-to-face guidance from a therapist. We therefore
recommend health care professionals to instruct their PD patients
suffering from dexterous dysfunction, to apply the HOMEDEXT
program. Furthermore, the applicability of HOMEDEXT is important
as long term improvement of dexterity can only be maintained by
continuous training, particularly if taking into account the pro-
gression of the disorder.
Authors' roles
1) Research project: A. Conception, B. Organization, C. Execution;
2) Statistical Analysis: A. Design, B. Execution, C. Review and
Critique;
3) Manuscript: A. Writing of the first draft, B. Review and Critique.
Tim Vanbellingen: 1 ABC; 2 ABC; 3 AB
Thomas Nyffeler: 1 AB; 2 C; 3 B
97
Julia Nigg: 1 C; 2 C; 3 B
Jorina Janssens: 1 C; 2 C; 3 B
Johanna Hoppe: 1 C, 2 C, 3 B
Tobias Nef: 2 C; 3 B
Rene M. Müri; 2 C; 3 B
Erwin E. H. van Wegen: 2 BC; 3 B
Gert Kwakkel: 2 C; 3 B
Stephan Bohlhalter: 1 AB; 2 AC; 3 B
Financial disclosures
None.
Conflicts of interests
The authors declare that they have no competing interests.
Acknowledgements
We wish to thank the clinical board of the Neurorehabilitation
center of the Klinik Bethesda Tschugg and the Luzerner Kant-
onsspital, supporting this trial. In addition, we are very grateful to
the clinical trial unit (CTU) of the university of Bern providing us
their expert advice in methodological statistical issues. Finally,
special thanks go to all the patients who participated in this study.
This study was supported by the Jacques and Gloria Gossweiler
Foundation.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.parkreldis.2017.05.021.
References
[1] A. Gebhardt, T. Vanbellingen, F. Baronti, B. Kersten, S. Bohlhalter, Poor dopa-
minergic response of impaired dexterity in Parkinson's disease: bradykinesia
or limb kinetic apraxia? Mov. Disord. 23 (2008) 1701e1706.
[2] S.L. Pohar, C.A. Jones, The burden of Parkinson disease (PD) and concomitant
comorbidities, Arch. Gerontol. Geriatr. 49 (2009) 317e321.
[3] T. Vanbellingen, B. Kersten, M. Bellion, P. Temperli, F. Baronti, R. Müri,
S. Bohlhalter, Impaired finger dexterity in Parkinson's disease is associated
with praxis function, Brain Cogn. 77 (2011) 48e52.
[4] T. Foki, T. Vanbellingen, C. Lungu, W. Pirker, S. Bohlhalter, T. Nyffeler,
J. Kraemmer, D. Haubenberger, F.P. Fischmeister, E. Auff, M. Hallett,
R. Beisteiner, Limb-kinetic apraxia affects activities of daily living in Parkin-
son's disease: a multi-center study, Eur. J. Neurol. 23 (2016) 1301e1307.
[5] M.A. Oliveira, A.M. Rodrigues, R.M. Caballero, R.D. Petersen, J.K. Shim, Strength
and isometric torque control in individuals with Parkinson's disease, Exp.
Brain Res. 184 (2008) 445e450.
[6] A.K. Rao, Enabling functional independence in Parkinson's disease: update on
occupational therapy intervention, Mov. Disord. 25 (2010) S146eS151.
[7] M.A. Van der Marck, J.G. Kalf, J.H. Sturkenboom, M.J. Nijkrake, M. Munneke,
B.R. Bloem, Multidisciplinary care for patients with Parkinson's disease, Park.
Relat. Disord. 15 (2009). S219eS23.
[8] M.E. Morris, R. Iansek, B. Kirkwood, A randomized controlled trial of move-
ment strategies compared with exercise for people with Parkinson's disease,
Mov. Disord. 15 (2009) 64e71.
[9] A. Nieuwboer, G. Kwakkel, L. Rochester, D. Jones, E. van Wegen, A.M. Willems,
F. Chavret, V. Hetherington, K. Baker, I. Lim, Cueing training in the home
improves gait-related mobility in Parkinson's disease: the RESCUE trial,
J. Neurol. Neurosurg. Psychiatry 78 (2007) 134e140.
[10] M.A. Hirsch, T. Toole, C.G. Maitland, R.A. Rider, The effects of balance training
and high-intensity resistance training on persons with idiopathic Parkinson's
disease, Arch. Phys. Med. Rehabil. 84 (2003) 1109e1117.
[11] G.M. Petzinger, B.E. Fisher, J.E. Van Leeuwen, M. Vukovic, G. Akopian,
C.K. Meshul, D.P. Holschneider, A. Nacca, J.P. Walsh, M.W. Jakowec, Enhancing
neuroplasticity in the basal ganglia: the role of exercise in Parkinson's disease,
Mov. Disord. 25 (2010) S141eS145.
[12] A.J. Hughes, Y. Ben-Shlomo, S.E. Daniel, A.J. Lees, What features improve the
accuracy of clinical diagnosis in Parkinson's disease: a clinicopathologic study,
Neurology 57 (2001) S34eS38.
[13] D.J. Gill, A. Freshman, J.A. Blender, B. Ravina, The Montreal Cognitive Assess-
ment as a screening tool for cognitive impairment in Parkinson's disease, Mov.
98 T. Vanbellingen et al. / Parkinsonism and Related Disorders 41 (2017) 92e98
Disord. 23 (2008) 1043e1046.
[14] R.C. Oldfield, The assessment and analysis of handedness: the edinburgh in-
ventory, Neuropsychologia 9 (1971) 97e113.
[15] M. Tere
metz, F. Colle, S. Hamdoun, M.A. Maier, P.G. Lindberg, A novel method
for the quantification of key components of manual dexterity after stroke,
J. Neuroeng. Rehabil. 2 (2015) 64.
[16] C.P. Kamm, H.P. Mattle, R.M. Müri, M.R. Heldner, V. Blatter, S. Bartlome,
J. Lüthy, D. Imboden, G. Pedrazzini, S. Bohlhalter, R. Hilfiker, T. Vanbellingen,
Home-based training to improve manual dexterity in patients with multiple
sclerosis: a randomized controlled trial, Mult. Scler. 21 (2015) 1546e1556.
[17] S. Mateos-Toset, I. Cabrera-Martos, I. Torres-S anchez, A. Ortiz-Rubio,
E. Gonz nez, M.C. Valenza, Effects of a single hand-exercise session on
alez-Jime
manual dexterity and strength in persons with Parkinson disease: a ran-
domized controlled trial, PM R 8 (2016) 115e122.
[18] G.M. Earhart, J.T. Cavanaugh, T. Ellis, M.P. Ford, K.B. Foreman, L. Dibble, The 9-
hole PEG test of upper extremity function: average values, test-retest reli-
ability, and factors contributing to performance in people with Parkinson
disease, J. Neurol. Phys. Ther. 35 (2011) 157e163.
[19] V. Peto, C. Jenkinson, R. Fitzpatrick, Determining minimally important dif-
ferences for the PDQ-39 Parkinson's disease questionnaire, Age Ageing 30
(2001) 299e302.
[20] T. Vanbellingen, T. Nyffeler, T. Nef, G. Kwakkel, S. Bohlhalter, E.E. van Wegen,
Reliability and validity of a new dexterity questionnaire (DextQ-24) in Par-
kinson's disease, Park. Relat. Disord. 33 (2016) 78e83.
[21] M.J. Peters, S.I. van Nes, E.K. Vanhoutte, M. Bakkers, P.A. van Doorn,
I.S. Merkies, C.G. Faber, PeriNomS Study group, Revised normative values for
grip strength with the Jamar dynamometer, J. Peripher Nerv. Syst. 16 (2011)
47e50.
[22] C.G. Goetz, B.C. Tilley, S.R. Shaftman, et al., Movement Disorder Society UPDRS
Revision Task Force, Movement Disorder Society sponsored revision of the
Unified Parkinson's Disease Rating Scale (MDS-UPDRS): scale presentation
and clinimetric testing results, Mov. Disord. 23 (2008) 2129e2170.
[23] J. Cohen, A power primer, Psychol. Bull. 112 (1992) 155e159.
[24] Y. Benjamini, Y. Hochberg, Controlling the false discovery rate: a practical and
powerful approach to multiple testing, J. R. Stat. Soc. 57 (1995) 289e300.
[25] V. Lun, N. Pullan, N. Labelle, C. Adams, O. Suchowersky, Comparison of the
effects of a self-supervised home exercise program with a physiotherapist-
supervised exercise program on the motor symptoms of Parkinson's dis-
ease, Mov. Disord. 20 (2005) 971e975.
[26] C.G. Canning, C. Sherrington, S.R. Lord, J.C. Close, S. Heritier, G.Z. Heller,
K. Howard, N.E. Allen, M.D. Latt, S.M. Murray, S.D. O'Rourke, S.S. Paul, J. Song,
V.S. Fung, Exercise for falls prevention in Parkinson disease: a randomized
controlled trial, Neurology 84 (2015) 304e312.
[27] R.M. Pickering, C. Fitton, C. Ballinger, L. Fazakarley, A. Ashburn, Self reported
adherence to a home-based exercise programme among people with Par-
kinson's disease, Park. Relat. Disord. 19 (2013) 66e71.
[28] P. Dahdal, A. Meyer, C. Menorca, K. Nowak, A.D. Roesch, P. Fuhr,
U. Gschwandtner, Fine motor function skills in patients with Parkinson dis-
ease with and without mild cognitive impairment, Dement. Geriatr. Cogn.
Disord. 42 (2016) 127e134.
[29] P. Hlustík, A. Solodkin, D.C. Noll, S.L. Small, Cortical plasticity during three-
week motor skill learning, J. Clin. Neurophysiol. 21 (2004) 180e191.
[30] W. Maetzler, J. Domingos, K. Srulijes, J.J. Ferreira, B.R. Bloem, Quantitative
wearable sensors for objective assessment of Parkinson's disease, Mov. Dis-
ord. 28 (2013) 1628e1637.