ARTICLE IN PRESS

Parkinsonism and Related Disorders 13 (2007) 195202

www.elsevier.com/locate/parkreldis

Review

Clinical relevance of rehabilitation programs for Parkinsons disease.

I: Non-symptom-specic therapeutic approaches
Michael Jobges, Heike Spittler-Schneiders, Caroline I.E. Renner, Horst Hummelsheim
Neurological Rehabilitation Centre, Neurology Department, Muldentalweg 1, 04828, Bennewitz, Saxony, Germany
Received 2 March 2006; received in revised form 20 July 2006; accepted 25 July 2006

Abstract
The idiopathic Parkinson syndrome (IPS) affects multiple structures of the central nervous system and exhibits a broad variety of
clinical symptoms that are only partially treatable by pharmacological treatment. Therefore, non-pharmacological approaches are highly
warranted. The aim of this study was to identify, categorize and rate studies on rehabilitative therapeutic approaches for IPS regarding
not only study design and statistics, but also clinical relevance. For reasons of homogeneity, only studies applying a rehabilitative
program targeting three or more symptoms were included and studies treating isolated symptoms were excluded. Study design and
statistics were rated by using an established rating system. To evaluate clinical relevance a rating system was developed considering
effectiveness, everyday life relevance, long-term effect, therapy frequency and setting, duration of the therapy units, effects on quality of
life, and assessment. Applying this rating system to 17 studies, we found four studies with a Level I of statistical quality and four studies
with high clinically relevant information. This novel method of analysing may help to gain valuable data from studies that may lack
strong methodology, and may help to promote further research.
r 2006 Elsevier Ltd. All rights reserved.
Keywords: Rehabilitation; Idiopathic Parkinson syndrome; Parkinson disease; Physiotherapy; Occupational therapy; Speech therapy; Evidence based
medicine; Clinical relevance

Contents
1.
2.

3.
4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1. Clinical relevance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2. Statistics and study design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3. Symptom-specicnon-symptom-specic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4. Selection of the studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction
The idiopathic Parkinson syndrome (IPS) is a neurodegenerative disorder affecting several predisposed types of
Corresponding author. Tel.: +403425888291; fax: +493414888190.

E-mail address: emjoebges@aol.com (M. Jobges).

1353-8020/$ - see front matter r 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.parkreldis.2006.07.019

195
196
196
196
196
196
197
197
202

neurones in specic regions of the brain. The underlying

neuropathological process is characterized by intraneuronal proteinaceous inclusion bodies. It begins in clearly
dened induction sites and advances in a topographically
predictable sequence. During early stages synuclein accumulations can be detected in the medulla oblongata and
olfactory bulb, later in the substantia nigra. In addition

ARTICLE IN PRESS
196

M. Jobges et al. / Parkinsonism and Related Disorders 13 (2007) 195202

other nuclear substances of the midbrain and basal

forebrain are the focus of initially subtle and then, severe
changes. At this point the disease reaches its symptomatic
phase. The pathological process encroaches upon the
telencephalic cortex during the end-stages [1].
This widespread distribution of neuropathologcial lesions explains the manifold clinical symptoms of IPS.
Furthermore, it explains, why the substitution of one
transmitter alone is unlikely to treat all of the symptoms of
IPS. For these reasons, non-pharmacological therapeutic
approaches are highly warranted. Unfortunately, six
Cochrane systematic reviews on physiotherapy, occupational therapy, and speech and language therapy bore the
result, that the trials had marked methodological aws. In
summary, they failed to nd conclusive evidence for the
benet of any form of the non-pharmacological therapeutic
approaches sufcient to recommend them in routine
clinical practice [2].
Besides these methodological aws the evaluation studies
possibly contain relevant therapeutical interventions disguised by shortcomings in study design. Therefore we
developed a novel method of analysis to glean gather
valuable data from papers that may lack strong methodology. The results of this analysis may help to direct further
research re-evaluating clinical relevant therapeutic approaches corresponding to the principles of evidence-based
medicine.

2. Methods
2.1. Clinical relevance
The clinical relevance of a study is estimated to be sufcient if the
following seven criteria are met:

1. Was the effectiveness of the non-pharmacologic therapeutic approach

reected by the improvement of outcome measures? (effectiveness)
2. Was an everyday life relevance of the respective therapy assessed by
appropriate scores? (everyday life relevance)
3. Was a long-term effect of the obtained improvement demonstrated by
re-examination after a dened time interval? (long-term effect)
4. Were therapy frequency and setting dened? (therapy frequency+setting)
5. What was the duration of the individual therapy units? (duration of
therapy units)
6. Was a change of the quality of life assessed by appropriate scores?
(quality of life)
7. Were the assessments performed at the same time of day to minimize
the extent of on-off uctuations that may affect patients responses?
(timing of assessment+medication)

Questions 13 analyse the therapeutic effect of the studies in detail. In

our opinion a clinically relevant study should describe an effective
approach; the effect should be relevant for everyday life and lead to a
long-term improvement.
Questions 4 and 5 pertaining to the frequency, the setting and the
duration of the therapy units should be answered in detail by the
respective study, for the therapy to be reproducible in any other clinical
environment.

Question 6 relates to the specic assessment scores. The quality of life

of the patients is affected by specic IPS symptoms that vary individually.
They can only be quantied by using adequate assessments.
The last question examined whether the temporal relationship between
the time of medication and the time of assessment was considered.
We classied a study as clinically highly relevant, when it permitted
positive answers to all these questions. If individual questions were not
answered and/or only negatively, e.g. pertaining to the long-term effect,
the study was given the rank of medium clinical relevance. We rated
studies as low clinically relevant, if they hardly permitted any positive
answer, or if all questions were negatively answered. The question of
effectiveness was not paramount to other questions, as in our opinion the
report of the ineffectiveness of a method may cause to reject this
ineffective method and this is clinically important, too.

2.2. Statistics and study design

Our second emphasis concerns the statistics and the design of the
studies. For this purpose we employed a slightly modied internationally
accepted systematic classication [3]:
Level I studies:
Level II studies:

Level III studies:

Randomized, controlled studies

Controlled clinical trials or observational
controlled studies such as cohort or case control
studies
Non-controlled studies like case series

Common inclusion criteria for all studies were that patients were
diagnosed with IPS by a neurologist and that the studies were
internationally accessible. Furthermore, we excluded therapy recommendations, which were published as expert opinions referring to nonpublished investigations. At least two patients had to be included for the
studies to be considered.

2.3. Symptom-specificnon-symptom-specific
In order to improve the comparability of the studies, we subdivided
them into symptom-specic and non symptom-specic studies. Symptomspecic studies investigated the effect of the application of one nonpharmacological therapeutic approach to improve one or two target
symptoms. Non symptom-specic studies evaluate the effect of a
rehabilitation program on three or more target symptoms. This approach
may appear somewhat arbitrary, but symptom-specic and non symptomspecic studies address different aspects of IPS rehabilitation. Symptomspecic interventions can be employed to tailor an individualized
rehabilitation program. Non-symptom specic-studies evaluate an effect
of rehabilitation on IPS in general.

2.4. Selection of the studies

The literature search was started on the basis of the Cochrane
Systematic Review [2] for IPS with the association of physiotherapy,
occupational therapy and speech therapy. Eleven physiotherapy studies
were found, which examined a single physiotherapeutic approach. Seven
studies compared two therapeutic approaches with another. Searching for
occupational therapy and speech therapy two and three studies were
found, respectively. One speech therapy study examined a single speech
therapy approach and two studies compared two therapeutic approaches.
Apart from these studies and the cross references mentioned, we applied a
systematic search in Medline using the terms: rehabilitation, physiotherapy, occupational therapy, speech therapy AND (Boolean operator) Parkinsons disease. Our goal was to seize all studies in international
medical journals, published in English from 1980 to August 2005, which
examined non-pharmacological approaches for the treatment of IPS.
The results concerning the non-symptom specic approaches are
presented, analyzed and discussed below.

ARTICLE IN PRESS
M. Jobges et al. / Parkinsonism and Related Disorders 13 (2007) 195202

3. Results
17 studies applied a non-symptom-specic therapy regimen. Table 1 describes the clinical relevance and Table 2
summarizes the clinical relevance and the study design.
4. Discussion
Fifteen of 17 clinical studies described a positive effect of
the applied therapy regimens on Parkinsonian symptoms.
Gibberd et al. [4] reported a different outcome. Examining
this crossover study in detail, it gives rise to criticism.
Concerning its clinical relevance, the effect of the therapy
regimen was evaluated by using arbitrarily chosen parameters. The assessment instruments were not validated and
cannot be compared to common standardized scores like
the UPDRS [5]. The effects on every day life and/or quality
of life were not addressed. In addition it is not possible to
reproduce this study, because the durations of the therapy
units were not documented. Finally, the time of assessment
and the time of drug intake were not specied. As a
consequence, the assessment at the beginning of the study
could have been performed in the on- and the assessment
at the end in the off-phase or vice versa. This possible
bias by the uncontrolled timing of medication restricts the
power of the study. In summary, the study presented by
Gibberd et al. [4] lacks a sufcient study design and the
clinical relevance is low.
The second study reporting a negative inuence of a
rehabilitation program on IPS was carried out by Pedersen
et al. [6]. Surprisingly, the gait analysis showed a signicant
decrease in stride length after training. As the list of items
in the training program is not precise, it gives rise to
speculations about the exact content of the training and
why it is the only study reporting a negative inuence on
IPS. The study design was rated as Level III, there was no
control group or baseline phase.
On the contrary, Comella et al. [7] performed a Level I
study as far as the statistical and study design is concerned.
In addition, the results were of high clinical relevance. The
assessments were all carried out during the on-phase.
Validated and standardized assessment instruments were
applied, which reect everyday life relevance. Quality of life
was not explicitly tested, but a geriatric depression scale
was used instead; it detected no signicant improvement.
After six months, a long-term effect was not observed. The
authors concluded that physical disability in moderately
advanced IPS objectively improved with a regular physical
rehabilitation program, but this improvement did not
persist beyond the end of the therapy.
Other studies were able to reproduce these results. In 118
patients, Trend et al. [8] studied the efcacy of a therapy
regimen consisting of individual and group therapy. The
assessments were accomplished at the same time of day to
minimize the extent to which on-off- uctuations might
affect patients responses. They used both validated
assessment scores and self-dened parameters and found

197

signicant improvements in depression, mobility, voice,

articulation and speech after therapy. A follow-up study by
Wade et al. [9] reported that 24 weeks after discontinuing
the therapy a trend towards a better stand-walk-sit score
and worse general and mental health was observed. In
conclusion, the state of IPS patients deteriorates over six
month, but a short period of multidisciplinary rehabilitation may improve mobility.
Reuter et al. applied an intensive exercise training [10].
They found signicant improvements of UPDRS sum
score, Columbia University Rating Score and basic motor
test. These effects had everyday life relevance and
improved quality of life as the improvements in the
adjective mood questionnaire of Zeersen and the sickness
impact prole suggested. A long-term effect was observed
for six weeks after training. The training was carried out
twice weekly over 14 weeks with single sessions of 60 min.
Timing of assessment and medication was not specied.
We rated the study as medium clinically relevant. The
study was designed as an open long-term pilot trial; the
study design for this reason was level III.
The idea that follow up treatments may be useful was
introduced by Dam et al. [11]. They designed a study in
which three phases of therapy (each consisting of 12 h of
therapy, ve days a week, for four weeks) were interrupted
by two to three therapy-free months. They found
signicant improvements in the Northwestern University
Disability Scale [12] but no signicant changes in Becks
depression inventory [13]. Patients were divided into one of
two therapy groups; one group received conventional
physiotherapy, the other Parkinson specic therapy
using visual and auditory cues. Whereas both groups
improved after each therapy phase, only the group with
Parkinson specic therapy had a long lasting benet.
These results suggest a benecial effect by a repeated
therapy but also the need to apply a Parkinson specic
therapy. Unfortunately, Dam et al. [11] failed to perform
the assessments at the same time of day. Therefore, the
data have to be interpreted with caution.
Marchese et al. [14] also proved the importance of IPS
specic rehabilitation techniques. They were able to show
that only the rehabilitation program including cueing
strategies leads to a long-term improvement.
A recently published study by Pellecchia et al. [15]
stressed the existence of a three months long-term effect.
The ADL section of the UPDRS, the self-assessment
Parkinsons disease Disability Scale, and the 10 m walk test
maintained their clinical improvement.
The other studies mentioned in Tables 1 and 2 support
these results [1623]. They bear more or less serious
inaccuracies concerning statistics, study design or clinical
relevance. Nevertheless, they point towards the same
direction as Comella, Trend and Dam.
Other research synthesis focuses on methodological
aspects. De Goede et al. [24] assessed the methodology of
12 studies by the evaluation of nine items (randomization
procedures, matching procedures, blinding procedures,

No positive effects

Improvements in step
length, average walking
speed, number of standups and sit-downs

Barthel Index remained

the same in therapy
group and worsened in
controls

Signicant improvement
for therapy group in
recent memory,
diminution of nauseas,
less urinary retention

No improvement in
torque measurements,
EMG analysis, gait.
Signicant improvement
for transfer and dressing

Northwestern University
Disability Scale and gait
velocity improved
signicantly in the
therapy group

Gibberd et al. (1981)

Szekely et al.(1982)

Gauthier et al.(987)

Hurwitz (1989)

Pedersen et al. (1990)

Formisano et al.
(1992)

Cedarbaum et
al.(1992)

Effectiveness

Author and year of

public-cation

No long-term effect
4 months after
therapy
Not documented

Not Documented

No positive effect in
patient
questionnaire
UPDRS, subscore
ADL signicantly
improved

Northwestern
University Disability
Scale reects
activities of daily
living

Home supervised
exercise regimen,
once weekly,
nursing-student
supervised therapy
for 8 months
Group training 2
times a week for 12
weeks
Physiotherapy
(single and group),
occupational and
speech and language
therapy (single) for
29 days (mean)
Physiotherapy,
occupational
therapy and speech
and language

Long-term effect for

patients after 6 and
12 months,
worsening for
controls
Not documented

Barthel Index
reects everyday life
relevance

The improvements
concern everyday life

Small groups with 2

therapists in each
group, 5 weeks, 2
therapies/week

Not documented

List of 17 items of
everyday life
activities showed
little change, but not
statistically analyzed

Therapy
frequency+setting
8 active and 8
passive therapies in 4
weeks
1 session per week
for 13 weeks, group
therapy

Long-term effect

Not documented

Not documented

Every-day life
relevance

1h

Not documented

No positive effect in
patient
questionnaire
Not documented

60 min

2 h of therapy/ day

Not documented

Slight change in
Becks depression
inventory, no change
in Bradbury Affect
balance test, no PD
specic quality of
live tests
Bradburn-Index
improving in
patients

Not documented

Quality of live

30 min

2h

2 h sessions, during
the rst hour
physical therapy
training, second
hour support and
teaching

Not documented

Duration therapy
units

Not documented

16

45

10

Tested at the same time

of the day
Not documented
(medication was not kept
stable throughout the
study)

30

16

Not documented

Not documented

24

Number of
participants

Measures in the
optimally effective time
of medication

Not documented

Timing of
assessment+medication

198

Table 1
Clinical relevance

ARTICLE IN PRESS

M. Jobges et al. / Parkinsonism and Related Disorders 13 (2007) 195202

Trend et al. (2002)

Marchese et
al.(2000)

Reuter et al. (1999)

Sitzia et al. (1998)

Signicant
improvements after
therapy in depression
(Hospital Anxiety and
Depression Scale and
Euroquol-5d), mobility
(timed walk over 10 m),
voice, articulation, and
speech (Emerson and
Enderby measures of
voice and articulation).

Signicant
improvements for
Northwestern University
Disability Scale. No
signicant change in
Becks depression
inventory
Two studies in one: In
study one signicant
improvements in
UPDRS, Functional
independence measure
(FIM), Barthel index
(BI) and gait velocity. In
study two signicant
improvements in
UPDRS, Webster rating
scale (WRS),
Northwestern University
disability scale (NUDS),
FIM and step length
Three items of the
Nottingham health
prole (NHP) changed
signicantly (pain,
emotional reaction, and
mobility).
Signicant
improvements of
UPDRS sum score,
Columbia University
Rating Score and basic
motor test.
Signicant
improvements in the
UPDRS in both groups

Dam et al. (1996)

Patti et al. (1996)

Signicant
improvements for
therapy group in
UPDRS subtest ADL
and motor function

Comella et.al. (1994)

The improvements
affect everyday life.
UPDRS part ADL
was tested, but
results were not
documented

Signicant
improvements in
ADL section of
UPDRS in both
groups

Signicant
improvements in
ADL parts of
UPDRS

6 weeks after
rehabilitation only in
the cued group,
UPDRS ADL and
motor section were
still improved
See Wade et
al.(2003)

Multidisciplinary
team approach, 6
days of therapy in 6
weeks. In the
morning
individualised single
therapy, in the
afternoon group
therapy

A cued or a noncued rehabilitation

program 3 times a
week for 6 weeks

2 h of individualised
treatment, duration
of group therapy not
documented

60 min

60 min

Intensive exercise
training in water and
in gymnasium, twice
weekly for 14 weeks

A long-term effect
could be observed
for 6 weeks after
training

Not documented

Not documented

In both studies 4
weeks of therapy
(physiotherapy,
occupational
therapy and speech
and language
therapy).

Signicant
improvement in
Euroquol-5d and a
visual analogue scale
of quality of life

NHP total score

changed
signicantly,
indicating an
improvement of
quality of live
Improvements in the
Adjective Mood
Questionnaire of
Zeersen and the
Sickness Impact
prole
Not documented

Not documented

Not documented

12 h per unit

Interdisciplinary
rehabilitation
program

In study one not

tested. In study two
UPDRS after six
months without
training still
signicantly higher
than baseline.

BI and NUDS are

reecting everyday
life relevance

Only geriatric
depression scale was
tested, this stayed
unchanged

1 h per unit

Follow up 4 weeks
after therapy showed
the 3 signicantly
changed items

The improvements
in the therapy group
lasted up to 12
months after the
training

The improvements
concern everyday life

Not documented

Was tested 6 months

after therapy, no
long-term effect

Signicant
improvements for
therapy group in
UPDRS subtest
ADL

therapy, 1therapy 3
days a week
69 exercises
concerning physioand occupational
therapy for 3 units a
week and for 4
weeks
5 days a week for 4
weeks, repeated 3
times with 3 months
of therapy free
intervals
interspersed

Assessment at the same

time of the day

Patients were always

examined at the peakdose effect of their
medication

Not mentioned

Katamnestic study
design refers to
subjective assessment of
the whole day

Not documented

Not documented

Measurement during
on

118

20

16

58

28

40

18

ARTICLE IN PRESS

M. Jobges et al. / Parkinsonism and Related Disorders 13 (2007) 195202

199

Effectiveness

24 weeks after treatment

faster stand-walk-sit test,
other parameters
worsened or did not
change signicantly

After therapy part IV of

UPDRS was signicantly
improved, other parts of
UPDRS remained
unchanged

Signicant
Improvements in
UPDRS motor score, 10meter-walking test after
training: proprioceptive
neuromuscular
stimulation, Mezieres
technique, and water
training for dexterity

Author and year of

public-cation

Wade et al. (2003)

Brefel-Courbon et
al.(2003)

Pellecchia et al.
(2004)

Table 1 (continued )

After 20 week
follow-up no
signicant effect was
found

Improvements in
UPDRS ADL score,
ten-meter walking
test and selfassessment, PD
disability maintained
after 3 months

ADL part of the

UPDRS declined
signicantly

Signicant
improvements in
PDQ-39 and SF-36

Self-ratingdepression score of
Zung was
signicantly
reduced, signicant
improvements in
self-assessment PD
disability score

Not documented

60 min individual
training

31

20

Not documented

Time of day of testing

remained constant
throughout the study

137

Not documented

PDQ-39 is not
signicantly
changed, indicating
no inuence on
quality of live 24
weeks after the
training

2 hours of
individualised
treatment, duration
of group therapy not
documented

Multidisciplinary
team approach, 6
days of therapy in 6
weeks. In the
morning
individualised single
therapy, in the
afternoon group
therapy
21-day period spa
therapy with thermal
baths, drinking
mineral waters,
various types of
showers, and
underwater massage,
3 afternoons
physiotherapy,
speech and
relaxation therapy 2
afternoons a week
Proprioceptive
neuromuscular
stimulation,
Mezieres technique,
water training of
dexterity 3 times a
week for 20 weeks

The improvement
directly after therapy
reported by Trend et
al. 2002 did not last
for 24 weeks after
therapy

The improvement as
well as the
deteriorations affect
everyday life (PDQ39, SF-36, Euroqol5d)

200

ADL part of
UPDRS unchanged

Number of
participants

Timing of
assessment+medication

Quality of live

Duration therapy
units

Therapy
frequency+setting

Long-term effect

Every-day life
relevance

ARTICLE IN PRESS

M. Jobges et al. / Parkinsonism and Related Disorders 13 (2007) 195202

ARTICLE IN PRESS
M. Jobges et al. / Parkinsonism and Related Disorders 13 (2007) 195202
Table 2
Summarized rating of all studies pertaining their clinical relevance and
study design
Author and year of publication

Clinical relevance

Study design

Gibberd et al. (1981)

Szekely et al. (1982)
Gauthier et al. (1987)
Hurwitz A. (1989)
Pedersen et al. (1990)
Cedarbaum et al. (1992)
Formisano et al. (1992)
Comella et.al. (1994)
Dam et al. (1996)
Patti et al. (1996)
Sitzia et al. (1998)
Reuter et al. (1999)
Marchese et al. (2000)
Trend et al. (2002)
Wade et al. (2003)
Brefel-Courbon et al. (2003)
Pellecchia et al. (2004)

Low
Medium
Medium
Low
Low
Medium
Medium
High
High
Medium
Low
Medium
Medium
High
Medium
Medium
High

Level
Level
Level
Level
Level
Level
Level
Level
Level
Level
Level
Level
Level
Level
Level
Level
Level

II
III
II
II
III
III
II
I
I
II+III
III
III
I
III
III
I
III

analysis of dropouts, reliability and validity of assessment

instruments, control of co-interventions, comparability of
baseline characteristics and applied statistics). They established an index with a maximum of 19 points and rated the
studies applying physical therapy with eight to 14 points.
This was interpreted as low methodological quality. In a
second step, they performed a quantitative or metaanalysis using the model of Hedges. This meta-analysis
was carried out regarding four parameters, i.e. activities of
daily living, walking speed, stride length, and neurological
signs. For activities of daily living, they found two of seven
effect sizes being signicant, for walking speed three of six,
for stride length two of four and for neurological signs
none of three. As a result of De Goedes evaluation more
tribute will be paid to methodological aspects in the future.
However, the analysis bears some limitations: The analyzed
studies are heterogeneous; only ve studies apply a therapy
regimen inuencing more than a single symptom. Of the
remaining seven studies, two use methods of behavioural
therapy, one uses rhythmic acoustic stimulation, one uses
karate arm training, and three use skill training. In our
opinion merging all these different approaches is inappropriate to answer the question whether physical therapy
is effective in IPS.
In 2002 a task force commissioned by the Movement
Disorder Society reported on Management of Parkinsons
disease. They employed similar methods of analysis. The
study quality score was derived from a list of key
methodological topics in accordance with a published
checklist relevant for determining the methodological
soundness of a trial [25]. A percentage score was calculated
for each study and used as an indicator of the overall
quality of a study. In addition each study was assigned one
of three levels of evidence (Level-I studies are randomized,
controlled trials; Level-II studies are controlled clinical
trials or observational controlled studies such as cohort or

201

case-control studies; Level-III studies are non-controlled

studies like case series). Eight studies were included for
analysis [26].
Five Studies were rated as Level I concerning study
design and statistics. In this group, only two studies
evaluated a complete rehabilitation program. The other
three evaluated karate training, facial muscle exercise, and
exercises focusing on spinal exibility. Three other studies
were graded as Level II [4,11,17]. In conclusion, the
authors found insufcient evidence for the efcacy of
physical/occupational therapy as adjunct treatment.
It may be wise to consider these reports on methodological quality with caution. As emphasized before, these
analyses were not evaluating homogenous groups. De
Goede et al. [24] mixed studies on behavioural therapy and
acoustic stimulation and karate training. From a pure
methodological point of view this may be possible nevertheless it seems unreasonable. The procedure of the
Movement Disorder Society task force [3,26] is comparable to De Goede et al. [24]. They analyzed training of only
the facial muscles and complete rehabilitation programs
targeting many symptoms in one section. Again, this may
be useful to stress methodological aspects, but not
sufcient to draw meaningful conclusions about the
different therapies evaluated in these studies or even apply
them to physio-/occupational therapy in general.
To avoid these problems we investigated only studies
applying a non-symptom-specic approach. Non symptom-specic studies evaluate the effect of a rehabilitation
program on three or more target symptoms. Therefore,
interpretation of the results is more coherent. Furthermore,
focusing on items of clinical relevance has not been done
before. In summary, there is clinically relevant evidence
that physical disability in moderately advanced IPS
benets from non-symptom specic rehabilitation programs.
In particular this paper is not entering into competition
with meta-analysis, it is a supplement to meta-analysis. It is
not the subject of this approach to contest the careful
realization of the principles of evidence-based medicine in
future studies. On the contrary many former studies would
have proted from the documentation of diagnostic criteria
and the stage of the disease of the patients. On the other
hand, we hope to open a discussion if this recently
developed approach of rating clinical relevance is worth
further elaboration and if this method may help to glean
valuable data from papers that may lack strong methodology, and may help to direct further research.
Even though these results are encouraging, many
questions remain unanswered: How can a long-term effect
be reached? How often should the rehabilitation program
be repeated? Is rehabilitative therapy useful in patients with
moderate to severe IPS? Obviously, further studies are
required. Hopefully, these further studies will employ study
designs considering criteria of evidence-based medicine.
Finally, as Dam et al. [11] and Marchese et al. [14]
briey mentioned, only the application of IPS specic

ARTICLE IN PRESS
202

M. Jobges et al. / Parkinsonism and Related Disorders 13 (2007) 195202

rehabilitation can achieve long-term therapeutic effects. In

our opinion an IPS specic rehabilitation program should
be individually composed of established symptom-specic
therapies. As a consequence symptom-specic IPS therapies have to be analysed concerning their efcacy, clinical
relevance, study design and statistics.
References
[1] Braak H, Del Tredici K, Bratzke H, Hamm-Clement J, SandmannKeil D, Rub U. Staging of the intracerebral inclusion body pathology
associated with idiopathic Parkinsons disease (preclinical and clinical
stages). J Neurol 2002;249(Suppl 3) III/15.
[2] Deane KH, Ellis-Hill C, Jones D, Whurr R, Ben-Shlomo Y, Playford
ED, et al. Systematic review of paramedical therapies for Parkinsons
disease. Mov Disord 2002;17:98491.
[3] The Movement Disorder Society Task Force. Management of
Parkinsons Disease: An evidencebased review. Introduction.
Mov Disord 2002;17(Suppl 4):S16.
[4] Gibberd FB, Page NGR, Spencer KM, Kinnear E, Hawksworth JB.
Controlled trial of physiotherapy and occupational therapy for
Parkinsons disease. British Med J 1981;282:1196.
[5] Fahn S, Elton RL. Members of the UPDRS development committee.
Unied Parkinsons disease rating scale. In: Fahn S, Marsden CD,
Calne DB, Goldstein M, editors. Recent Developments in Parkinsons
disease, vol. II. Florham Park: Macmillan Healthcare Information;
1987. p. 15363 p. 293304.
[6] Pedersen SW, Oberg B, Insulander A, Vretman M. Group training in
Parkinsonism: quantitative measurements of treatment. Scand J
Rehabil Med 1990;22:20711.
[7] Comella CL, Stebbins GT, Brown-Toms N, Goetz CG. Physical
therapy and Parkinsons disease: A controlled clinical trial. Neurology 1994;44:3768.
[8] Trend P, Kaye J, Gage H, Owen C, Wade D. Short-term effectiveness
of intensive multidisciplinary rehabilitation for people with Parkinsons disease and their carers. Clin Rehabil 2002;16:71725.
[9] Wade DT, Gage H, Owen C, Trend P, Grossmith C, Kaye J.
Multidisciplinary rehabilitation for people with Parkinsons disease: a
randomized controlled study. J Neurol Neurosurg Psychiatry 2003;
74:15862.
[10] Reuter I, Engelhardt M, Stecker K, Baas H. Therapeutic value of
exercise training in Parkinsons disease. Med Sci Sports Exerc 1999;
31:15449.
[11] Dam M, Tonin P, Casson S, Bracco F, Piron L, Pizzolato G, et al.
Effects of conventional and sensory-enhanced physiotherapy
on disability of Parkinsons disease patients. Adv Neurol 1996;69:
5515.

[12] Ramaker C, Marinus J, Stiggelbout AM, Van Hilten BJ. Systematic

evaluation of rating scales for impairment and disability in
Parkinsons disease. Mov Disord 2002;17:86776.
[13] Schwab J, Bialow M, Clemmons R, Martin P, Holzer C. The Beck
depression inventory with medical inpatients. Acta Psychiat Scand
1967;43:25566.
[14] Marchese R, Diverio M, Zucchi F, Lentino C, Abbruzzese G. The
role of sensory cues in the rehabilitation of Parkinsonian patients: a
comparison of two physical therapy protocols. Mov Disord 2000;
15:87983.
[15] Pellecchia MT, Grasso A, Biancardi LG, Squillante M, Bonavita V,
Barone P. Physical therapy in Parkinsons disease: an open long-term
rehabilitation trial. J Neurol 2004;251:5958.
[16] Cedarbaum JM, Toy L, Silvestri M, Green-Parson A, Harts A,
McDowell FH. Rehabilitation programs in the management
of patients with Parkinsons disease. J Neurol Rehabil 1992;6:
719.
[17] Formisano R, Pratesi L, Modarelli FT, Bonifati V, Meco G.
Rehabilitation and Parkinsons disease. Scand J Rehabil Med
1992;24:15760.
[18] Gauthier L, Dalziel S, Gauthier S. The benets of group occupational
therapy for patients with Parkinsons disease. Am J Occup Therapy
1987;41:3605.
[19] Hurrwitz A. The benet of a home exercise regimen for
ambulatory Parkinsons disease patients. J Neurosci Nurs 1989;21:
1804.
[20] Patti F, Reggio A, Nicoletti F, Sellaroli T, Deinite G, Nicoletti F.
Effects of rehabilitation therapy on Parkinsoniansdisability and
functional independence. J Neurol Rehabil 1996;10:22331.
[21] Sitzia J, Haddrell V, Rice-Oxley M. Evaluation of a nurse-led
multidisciplinary neurological rehabilitation programme using the
Nottingham health prole. Clin Rehabil 1998;12:38994.
[22] Szekely BC, Kosanovich NN, Sheppard W. Adjunctive treatment in
Parkinsons disease: physical therapy and comprehensive group
therapy. Rehabil Lit 1982;43:726.
[23] Brefel-Courbon C, Desboeuf K, Thalamas C, Galitzky M, Senard
JM, Rascol O, et al. Clinical and economic analysis of spa therapy in
Parkinsons disease. Mov Disord 2003;18:57884.
[24] De Goede CJT, Keus SHJ, Kwakkel G, Wagenaar RC. The effects of
physical therapy in Parkinsons disease: a research synthesis. Arch
Phys Med Rehabil 2001;82:50915.
[25] Dixon RA, Munro JF, Silocks RP. Checklists for critical appraisal.
The evidence based medicine workbook: Critical appraisal for clinical
problems solving. Appendix I. Butterworth-Heinemann Medical,
1997.
[26] The Movement Disorder Society Task Force. Management of
Parkinsons Disease: An evidencebased review. Physical and
occupational therapy in Parkinsons disease. Mov Disord 2002;
17(Suppl 4):S1569.