2013 - Novak - Interventions For Children With CP

DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY REVIEW
A systematic review of interventions for children with cerebral

palsy: state of the evidence
IONA NOVAK 1,2
| SARAH MCINTYRE 1,2 | CATHERINE MORGAN 1,2 | LANIE CAMPBELL 2 | LEIGHA DARK 1 |
NATALIE MORTON 1 | ELISE STUMBLES 1 | SALLI-ANN WILSON 1 | SHONA GOLDSMITH 1,2
1 Cerebral Palsy Alliance, Sydney; 2 University of Notre Dame Australia, Sydney, Australia.
Correspondence to Associate Professor Iona Novak, Head of Research, Cerebral Palsy Alliance Research Institute, PO Box 560, Darlinghurst NSW 1300, Australia.
E-mail: inovak@cerebralpalsy.org.au
This article is commented on by Msall on pages 877–878 of this issue.
PUBLICATION DATA AIM The aim of this study was to describe systematically the best available intervention
Accepted for publication 5th June 2013. evidence for children with cerebral palsy (CP).
METHOD This study was a systematic review of systematic reviews. The following databases
ABBREVIATIONS were searched: CINAHL, Cochrane Library, DARE, EMBASE, Google Scholar MEDLINE,
COPM Canadian Occupational Perfor- OTSeeker, PEDro, PsycBITE, PsycINFO, and speechBITE. Two independent reviewers
mance Measure determined whether studies met the inclusion criteria. These were that (1) the study was a
GAS Goal Attainment Scaling systematic review or the next best available; (2) it was a medical/allied health intervention;
MACS Manual Ability Classification and (3) that more than 25% of participants were children with CP. Interventions were coded
System using the Oxford Levels of Evidence; GRADE; Evidence Alert Traffic Light; and the
NDT Neurodevelopmental therapy International Classification of Function, Disability and Health.
RESULTS Overall, 166 articles met the inclusion criteria (74% systematic reviews) across 64
discrete interventions seeking 131 outcomes. Of the outcomes assessed, 16% (21 out of 131)
were graded ‘do it’ (green go); 58% (76 out of 131) ‘probably do it’ (yellow measure); 20% (26
out of 131) ‘probably do not do it’ (yellow measure); and 6% (8 out of 131) ‘do not do it’ (red
stop). Green interventions included anticonvulsants, bimanual training, botulinum toxin,
bisphosphonates, casting, constraint-induced movement therapy, context-focused therapy,
diazepam, fitness training, goal-directed training, hip surveillance, home programmes,
occupational therapy after botulinum toxin, pressure care, and selective dorsal rhizotomy.
Most (70%) evidence for intervention was lower level (yellow) while 6% was ineffective (red).
INTERPRETATION Evidence supports 15 green light interventions. All yellow light
interventions should be accompanied by a sensitive outcome measure to monitor progress
and red light interventions should be discontinued since alternatives exist.
Thirty to 40% of interventions have no reported evidence- possibility of newer, safer, and more effective interventions.
based and, alarmingly, another 20% of interventions pro- Orthopaedic surgery and movement normalization were
vided are ineffectual, unnecessary, or harmful.1 The gap once the mainstays of intervention, but localized antispas-
between research and practice has been well documented ticity medications and motor learning interventions have
in systematic reviews1 across multiple diagnoses, special- gained increased popularity.4,5 Thus, the sheer volume of
ties, and countries. Surveys confirm that, unfortunately, the research published makes it hard for clinicians to keep up
research–practice gap occurs within the cerebral palsy (CP) to date.6 Systematic reviews seek to provide evidence sum-
field to the same degree.2,3 This gap exists despite numer- maries, but, in spite of this, clinicians find it difficult to
ous systematic reviews providing guidance about what does interpret review findings and stay abreast of these
and does not work for children with CP. When clinicians syntheses.7 Furthermore, the introduction of new and
want to help, families expect effective interventions, and sometimes competing effective interventions increases the
the health system depends upon cost-effective services, the complexity of clinical reasoning required by clinicians, who
provision of ineffectual interventions is illogical. In view are primarily motivated to improve outcomes for children.8
of this, why is there such variable uptake of best available In the last 10 years, the field has adopted the World
evidence within real clinical practice? Health Organization’s International Classification of Func-
In the last decade, the CP evidence base has rapidly tioning, Disability and Health (ICF),9 which has redefined
expanded, providing clinicians and families with the the way clinicians understand CP and think about inter-
© 2013 Mac Keith Press DOI: 10.1111/dmcn.12246 885

14698749, 2013, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/dmcn.12246 by Nat Prov Indonesia, Wiley Online Library on [27/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
vention options. From an ICF perspective, CP impacts on What this paper adds
a person’s ‘functioning’, (inclusive of body structures [e.g. • Of 64 discrete CP interventions, 24% are proven to be effective.
limbs], body functions [e.g. intellectual function], activities • 70% have uncertain effects and routine outcome measurement is necessary.
[e.g. walking], and participation [e.g. playing sport]), which • 6% are proven to be ineffective.
in turn may cause ‘disabilities’, such as impairments, activ- • Effective interventions reflect current neuroscience and pharmacological
knowledge.
ity limitations, and participation restrictions. Moreover, • All effective interventions worked at only one level of the ICF.
each person with CP lives within a personalized environ-
ment and thus their context also contributes to determin- Search strategy
ing their independence, comprising personal factors (e.g. Our review was carried out using a protocol based upon
motivation) and environmental factors (e.g. architectural recommendations from the Cochrane Collaboration and
accessibility).9,10 Thus, there are many potential problems PRISMA statements.21,22 Relevant articles were identified
a child with CP may face and seek intervention for. The by searching the CINAHL (1983–2012); Cochrane Data-
field has chosen a philosophical shift away from almost base of Systematic Reviews (1993–2013; www.cochra-
exclusively redressing physical impairments underlying ne.org); Database of Reviews of Effectiveness (DARE);
functional problems to adopting an additional focus on EMBASE (1980–2012); ERIC; Google Scholar; MED-
maximizing children’s environment, their independence in LINE (1956–2012); OTSeeker (www.otseeker.com); Phys-
daily activities, and their community participation.11 Fur- iotherapy Evidence Database (PEDro [www.pedro.fhs.usyd.
thermore, clinicians applying the recommended goal-based edu.au]); Psychological database for Brain Impairment
approach seek to choose interventions guided by what Treatment Efficacy (PsycBITE [www.psycbite.com]); Psy-
would best help the family achieve their goals.12–14 Couple cINFO (1935–2012); PubMED; and Speech Pathology
these philosophical preferences with widespread barriers to Database for Best Interventions and Treatment Efficacy
research implementation (such as limited time, insufficient (speechBITE [www.speechbite.com]). Searches were sup-
library access, limited research appraisal skills, attitudinal plemented by hand searching. The search of published
blocks to research, and differing patient preferences), and studies was performed in July and August 2011 and
there is no assurance that children with CP will receive updated in December 2012. Interventions and keywords
evidence-based interventions.1,15,16 for investigation were identified using (1) contributing
The aim of this paper was to describe systematically the authors’ knowledge of the field; (2) internationally recog-
best available evidence for CP interventions using the nized CP websites such as the American Academy of Cere-
GRADE17 system and to complement these findings with bral Palsy and Developmental Medicine (www.aacpdm.
the Evidence Alert Traffic Light System18 in order to pro- org), CanChild (www.canchild.ca), the Cerebral Palsy Alli-
vide knowledge translation guidance to clinicians about what ance (www.cerebralpalsy.org.au), Cincinnati Children’s
to do. The purpose of rating the whole CP intervention evi- Hospital (www.cincinnatichildrens.org), Karolinksa Insitu-
dence base within the one paper was to provide clinicians, tet (www.ki.se), NetChild (www.netchild.nl), NeuroDev-
managers, and policy-makers with a ‘helicopter’ view of best Net (www.neurodevnet.ca), and Reaching for the Stars
available intervention evidence that could be used to (1) (www.reachingforthestars.org); and (3) the top 20 hits in
inform decision-making by succinctly describing current Google using the search term ‘cerebral palsy’ as an indica-
evidence about CP interventions across the wide span of dis- tor of popular subject matter.
ciplines involved in care; (2) rapidly aid comparative clinical Electronic databases were searched with EBSCO host
decision-making about similar interventions; and (3) provide software using PICOs [patient/problem, intervention, com-
a comprehensive resource that could be used by knowledge parison, and outcome] search terms. The full search strat-
brokers to help prioritize the creation of knowledge transla- egy is available from the authors on request.
tion tools to promote evidence implementation.19
Inclusion criteria
METHOD Published studies about intervention for children with CP
Study design fulfilling criteria under the headings below were included.
A systematic review of systematic reviews (i.e. the highest
level of CP intervention research evidence available) was Type of study
conducted in order to provide an overview of the current First, studies of level 1 evidence (systematic reviews),
state of CP intervention evidence. Systematic reviews were rated using the Oxford 2011 Levels of Evidence were
preferentially sought since reviews provide a summary of preferentially sought.23 The Oxford 2011 Levels of
large bodies of evidence and reviews help to explain differ- Evidence for treatment benefits include level 1, a system-
ences among studies. Moreover, reviews limit bias which atic review of randomized trials or n-of-1 trials; level 2, a
assists clinicians, managers, and policy-makers with deci- randomized trial or observational study with dramatic
sion-making about current best available evidence.20 How- effect; level 3, a non-randomized controlled cohort/
ever, for interventions for which no systematic reviews follow-up study; level 4, a case series, case–control study,
existed, lower levels of evidence were included to illumi- or a historically controlled study; and level 5, mechanism-
nate the current state of the evidence. based reasoning.
886 Developmental Medicine & Child Neurology 2013, 55: 885–910

Evidence of Oxford levels 2 to 4 were included only if Where multiple systematic reviews existed and newer
(1) level 1 evidence did not exist on the topic and then the level 1 to 2 evidence superseded the findings of earlier
next best available highest level of evidence was included; level 1 evidence, the grades were assigned based on the
or if (2) level 2 randomized controlled trial(s) had been most recent high-quality evidence.
published since the latest systematic review, which substan-
tially changed knowledge about the topic. Types of intervention
Second, retrieved bodies of evidence were coded using Studies were included if they involved the provision of and
the GRADE17 system and Evidence Alert Traffic Light intervention by either a medical practitioner or allied
System18 using two independent raters, with 100% health professional.
agreement reached. The GRADE17 system was chosen
because it is a criterion standard evidence-grading tool Types of participants
and is endorsed by the World Health Organization. Def- Studies were included if they explicitly involved human
initions of the GRADE terms appear in the notes to participants and more than 25% of the participants were
Table I and a full description of panel rating processes children with CP.
are available from www.gradeworkinggroup.org/publica- Studies were excluded from the review if (1) they were
tions/JCE_series (retrieved 8 March 2013). Notably, the diagnostic studies, prognostic studies, or interventions
GRADE system rates both (1) the quality of the evi- aimed at preventing CP (e.g. magnesium sulphate186 and
dence (randomized trials, high; observational studies, low; hypothermia187); (2) they provided lower levels of evidence,
and other levels of evidence, very low, but it is worth unless no systematic review had been published; (3) partici-
mentioning that high-quality evidence is downgraded if pants were adults, although if a study predominantly
methodological flaws exist and low-quality evidence is (>75%) studied children but included a small proportion of
upgraded if high and certain effect sizes exist [e.g. popu- young adults (<25%) the paper was included; (4) they
lation-based CP register data])17 and (2) the strength of reviewed generic prophylaxis interventions (e.g. good par-
the recommendation for use, which weighs up trade-offs enting, standard neonatal care for all infants, i.e. not CP-
between the benefits and harms of using the interven- specific interventions); (5) they reviewed a whole discipline,
tion, whereby a panel considers (a) the methodological not individual interventions (e.g. physiotherapy, occupa-
quality of the evidence supporting estimates of likely tional therapy, speech pathology); (6) they were considered
benefit and likely risk; (b) inconvenience; (c) the impor- alternative and complementary interventions with no pub-
tance of the outcome that the treatment prevents; (d) lished evidence; (7) a second publication of the same study
the magnitude of the treatment effect; (e) the precision published the same results; and (8) they were unpublished
of the estimate of the treatment effect; (f) the risks asso- or not peer reviewed.
ciated with therapy; (g) the burdens of therapy; (h) the
costs; and (i) the varying values.17 The GRADE method- Data abstraction
ology means that sometimes bodies of evidence may be A data abstraction sheet based on the Cochrane’s recom-
assigned a strong recommendation even when the quality mendations21 was developed. Abstracts identified from
of the evidence is low. This is either because there is a searches were screened by two independent raters (CP
high likelihood of harm from no intervention (e.g. anti- research experts and knowledge brokers) to determine their
convulsants to prevent seizures or ulcer prevention pres- eligibility for further review. Abstracts were retained for
sure care) or because the treatment has a low effect size full review if they met the inclusion criteria or if more
and is expensive to provide, but a safe, more effective, information was required from the full text to confirm that
cost-comparable alternative exists (e.g. phenol vs botu- the study met all the eligibility criteria. Two independent
linum toxin A; or neurodevelopmental therapy [NDT] vs reviewers then reviewed full-text versions of all retained
motor learning). The Evidence Alert Traffic Light Sys- articles and all additional articles identified by hand search-
tem18 was chosen because it is a GRADE-complementary ing. Full-text articles were retained if they met inclusion
knowledge translation tool, designed to assist clinicians criteria. Agreement on inclusion and exclusion assignment
to obtain easily readable, clinically useful answers within of the full-text articles was unanimous. Data extracted from
minutes.6 The Evidence Alert also provides a simple, included studies comprised the authors and date of the
common language between clinicians, families, managers, study; the type and purpose of the intervention imple-
and funders, based upon three-level colour coding that mented; the study design; the original authors’ conclusions
recommends a course of action for implementation of about efficacy across study outcomes; and the original
the evidence within clinical practice. The Evidence Alert authors’ conclusions on strength of evidence (based on
System18 has been shown to increase by threefold clini- their assessment of whether there was no evidence of bene-
cians’ reading habits about CP research.24 Figure 1 fit, qualified support, or strong support). For lower level
describes the GRADE system and the Evidence Alert evidence, risk of bias was assessed using the Cochrane
System and their relationship to each other. Table I criteria.
shows the included studies, best evidence levels grades The data extracted from each included study were sum-
and traffic light classification.25–185 marized, tabulated, and assigned a level of evidence rating
Review 887
Table I: Included studies, best available evidence levels, grades and traffic lights
GRADE
Oxford Quality
evidence of Strength of rec- Traffic light
Intervention Intervention outcome (ICF level) Citations Panel comments level evidence ommendation action
1 Acupuncture: electro-stimulation to scalp and Improved gross motor function Zhang25 Insufficient evidence 1 Low Weak + Yellow
body via needles and manual pressure (A) MEASURE
2 Alcohol: muscular injections to induce chemical Reduce muscle spasticity Delgado26 Insufficient evidence to support, but BoNT-A exists 1 N/A Weak Yellow
denervation for treating local spasticity locally via injections (BS) as a highly effective alternative – therefore MEASURE
probably do not use alcohol unless BoNT-A total
dose limitations in play
3 Alternative and augmentative communication: Improved general Pennington27 Lower-quality supporting evidence 1 Very low Weak + Yellow |
technology alternatives to verbal speech, e.g. communication skills (A) MEASURE
communication boards, speech generating Improved communication skills Branson28 Lower-quality supporting evidence 1 Very low Weak + Yellow
devices of pre-school children (A) MEASURE
Improved communication skills Pennington29 Lower-quality supporting evidence 1 Very low Weak + Yellow
of conversational partners (P) MEASURE
Enhanced supplementation of Hanson30 Lower-quality supporting evidence 1 Very low Weak + Yellow
verbal speech (A) Millar31 1 MEASURE
4 Animal-assisted therapy: service animals to Improved socialization and Mun~ oz Lasa32 Lower-quality supporting evidence 1 Very low Weak + Yellow
provide companionship and assist with mood; reduced stress, anxiety MEASURE
independence, e.g. seizure first aid, door and loneliness; and improved

opening, crossing roads leisure (BS and P)
Improved independence via Winkle33 Lower-quality supporting evidence 1 Very low Weak + Yellow
service dogs (P) MEASURE
5 Anticonvulsants: medications to prevent Improved seizure control (BS) – No evidence in CP. Since high quality evidence – N/A Strong + Green GO
seizures exists in non-CP populations and there are high
risks of adverse events from uncontrolled
seizures therefore – do use anticonvulsants
6 Assistive technology: equipment or devices to Improved independence in Wilson34 Lower-quality supporting evidence 2 Low Weak + Yellow
improve independence e.g. walking frames, activities of daily living (A and MEASURE
wheelchairs, adapted computer access P)
Improved computer access via Davies35 Lower-quality supporting evidence 1 Very low Weak + Yellow
a switch or key guard (A) MEASURE
36
Improved independence in Jones Lower-quality supporting evidence 2 Low Weak + Yellow
early mobility via powered Livingstone37 1 MEASURE
wheelchairs (A and P)
Improved participation in Chantry38 Lower-quality supporting evidence 1 Very low Weak + Yellow
education, communication Sandlund39 1 MEASURE
and play via alternative
computer access (P)
Improved function via robotic Laufer40 Lower-quality supporting evidence 1 Very low Weak + Yellow
training or virtual reality (A) Parsons41 1 MEASURE
Sandlund39 1
Snider42 1
Wang43 1
Improved transfers via a hoist Jung44 Insufficient evidence 1 Very low Weak + Yellow
(A) MEASURE
45
Improved weight bearing and Pin Insufficient evidence 1 Low Weak + Yellow
bone mineral density via a MEASURE
standing frame (BS)
Improved sleep positioning via Wynn46 Lower-quality supporting evidence 1 Very low Weak + Yellow
a sleep system (BS) MEASURE
47
Reduced carer burden (E) Nicolson Lower-quality supporting evidence 1 Very low Weak + Yellow
MEASURE
26
7 Baclofen (oral): antispasticity medication Reduced spasticity (BS) Delagado Lower-quality supporting evidence 1 Low Weak + Yellow
MEASURE
Table I: Continued
GRADE
Oxford Quality
8 Behaviour therapy: positive behaviour support, Improved child behaviour (from Roberts48 Effective, but low CP numbers were included in the 2 Low Weak + Yellow
behavioural interventions, and positive the Stepping Stones Triple P Sanders49 study samples and publication bias existed 2 MEASURE
parenting Programme) (A)
Improved parenting skills (E) Whittingham50 Insufficient evidence 1 Very low Weak + Yellow
MEASURE
9 Bimanual training: repetitive task training in the Improved hand function, i.e. Gordon51 Effective. Equal effectiveness to constraint-induced 2 High Strong + Green GO
use of two hands together bilateral hand use for children Sakzewski4 movement therapy 1
with hemiplegia (A) Sakzewski52 2
10 Biofeedback: electronic feedback about muscle Improved muscle activation Dursun53 Effective if combined with other treatments 2 Low Weak + Yellow
activity to teach voluntary control and active range of motion MEASURE
(BS)
Improved walking (A) Dursun53 Insufficient evidence 2 Low Weak + Yellow
MEASURE
Improved hand function (A) Bloom54 Lower-quality supporting evidence 4 Very low Weak + Yellow
MEASURE
11 Bisphosphonates: medication to suppress bone Improved bone mineral density Fehlings55 Effective. Small RCTs suggest a positive effect and 1 Moderate Strong + Green GO
reabsorption to treat osteoporosis (BS) Hough56 there are high risks of adverse events from no 1
treatment
12 Botulinum toxin (BoNT-A): medication injected Reduced lower limb muscle Ade-Hall57 Effective and safe 1 High Strong + Green GO
into overactive spastic muscles to locally block spasticity (BS) Albavera- 1
spasticity Hernandez58
Boyd59 1
Heinen60 1
Koog61 1
Lukban62 1
Love63 1
Mulligan64 1
Reduced upper limb muscle Fehlings65 Insufficient evidence. Note: function was 1 Moderate Strong + Green GO
spasticity (BS) Reeuwijk66 preferentially measured over spasticity reduction 1
Wasiak67 in high quality studies. Since the drug is highly 1
effective in lower limb muscles, we expect
comparable results – therefore do use BoNT-A
Reduced hypertonia of the neck Novak68 Insufficient evidence. Since high-quality evidence 1 N/A Weak + Yellow
muscles (BS) supports tone reduction in primary dystonia (non- MEASURE
CP populations), we expect similar results –
therefore probably do use BoNT-A
Improved walking function (A) Koog61 Probably effective in combination with 1 Moderate Strong + Green GO
Love63 physiotherapy therefore do use 1
Ryll69 1
Improved hand function and Boyd59 Effective in combination with occupational therapy 1 High Strong + Green GO
performance of functional Fehlings65 1
hand activities (A) Hoare70 1
Hoare71 1
Reduced pain (BS) Rawicki72 Insufficient evidence 1 Very low Weak + Yellow
MEASURE
73
Reduced drooling (BS) Lim Effective short term and given the adverse social 1 Moderate Strong + Green GO
Reddihough74 outcomes from no treatment – do use 1
Walshe75 1
Review
889
Table I: Continued
GRADE
Oxford Quality
13 Casting: Plaster casts applied to limbs to (a) Improved passive range of Autti-Ramo76 Effective. Gains in ankle range of motion are very 1 Low Strong + Green GO
stretch muscles for muscle lengthening, i.e. motion of the lower limbs Blackmore77 small but are potentially clinically meaningful for 1
contracture reduction casts changed regularly; (BS) Effgen78 children that need more dorsiflexion to walk, 1
or (b) reduce spasticity Katalinic79 therefore – do use 1
Improved passive range of Autti-Ramo76 Insufficient evidence 1 Low Weak + Yellow
motion of the upper limbs Lannin80 1 MEASURE
(BS) Teplicky81 1
Improved function (A) Autti-Ramo76 Insufficient evidence 1 Low Weak Yellow
Blackmore77 1 MEASURE
Effgen78 1
Katalinic79 1
Augmented effects of BoNT Blackmore77 Effective but gains are small 1 Low Weak + Green GO
(BS)
79
Reduced muscle spasticity (BS) Katalinic Insufficient evidence. Newer understandings of 1 Low Weak Yellow
Teplicky82 spasticity indicate a ‘local’ intervention will not 1 MEASURE

improve a ‘central’ condition – therefore probably
do not use casting for spasticity reduction
14 Coaching parents: emotional support, Improved parenting skills and Graham82 Insufficient evidence. More research needed with 4 Very low Weak + Yellow
information exchange and a structured coping (E) stronger designs MEASURE
process of tutoring parenting behaviours
15 Cognitive behaviour therapy (CBT): identifying Improved depression, anxiety, – No evidence in CP. Since high-quality evidence – N/A Weak + Yellow
unhelpful thoughts and behaviours and sleep, attention, behaviour supports CBT in non-CP populations – therefore MEASURE
teaching cognitive restructuring and self- and enuresis (BS) probably do use CBT
management of constructive thinking and
actions
16 Communication training: training Improved interaction between Pennington29 Insufficient evidence 1 Very low Weak + Yellow
communication partners to effectively children and their parents (P) Pennington83 1 MEASURE
communicate, e.g. Interaction Training; Hanen;
It Takes Two to Talk
17 Conductive education (CE): a Hungarian Improved ‘orthofunction’ Darrah84 Conflicting evidence. Majority of studies show no 1 Low Weak Yellow
educational classroom-based approach to (response to biological and Tuersley- difference to no treatment 1 MEASURE
teaching movement using rhythmic intention, social demands) (BS) Dixon85
routines and groups Improved performance of Darrah84 Conflicting evidence. Majority of studies show no 1 Low Weak Yellow
functional activities (A) Tuersley- difference to no treatment 1 MEASURE
Dixon85
Improved cognition (BS) Tuersley- Conflicting evidence. Majority of studies show no 1 Low Weak Yellow
Dixon85 difference to no treatment MEASURE
18 Constraint-induced movement therapy (CIMT): Improved hand function of the Boyd59 Effective. Even more RCTs have been published 1 Moderate Strong + Green GO
constraining the dominant hand in a mitt or affected hand for children Hoare86 after the included reviews confirming 1
cast, to enable intensive training of the with hemiplegia (A) Huang87 effectiveness 1
hemiplegic hand Nascimento88 1
Sakzewski4 1
19 Context-focused therapy: changing the task or Improved function (A) Law89 Effective. Note: a single rigorous RCT shows equal 2 High Strong + Green GO
the environment (but not the child) to effectiveness to child-focused therapy
promote successful task performance
20 Counselling (parents): fostering understanding Improved parental coping and – No evidence in CP. No published research – N/A Weak + Yellow
of how life problems lead to distress, mental health (E) evidence, opinion papers existed MEASURE
relationship breakdown and mental health Improved parental coping via Palit90 Insufficient evidence 4 Very low Weak + Yellow
issues, to improve communication and parent to parent support (E) MEASURE
interpersonal skills
Table I: Continued
GRADE
Oxford Quality
21 Cranial osteopathy: palpation using small Improved mobility, quality of Wyatt91 Ineffective. Note: a single rigorous RCT shows no 2 High Strong Red STOP
movements to ease musculoskeletal strain life and general health (A and benefit when compared to no treatment
and treat the central nervous system P)
22 Dantrolene: antispasticity medication Reduce spasticity (generalized) Delgado26 Insufficient evidence 1 Low Weak Yellow
(BS) MEASURE
26
23 Diazepam: antispasticity medication Reduce spasticity (generalized) Delgado Effective short term, therefore – do use 1 Moderate Strong + Green GO
(BS)
92
24 Dysphagia management: promoting safe Improved safety of swallow via Snider Lower-quality supporting evidence 1 Low Weak + Yellow
swallowing by changing food textures, sitting thickened fluids i.e. less MEASURE
position, oral motor skills and using oral aspiration (BS)
appliances and equipment Improved safety of swallow via Snider92 Conflicting evidence 1 Low Weak + Yellow
upright positioning, i.e. less MEASURE
aspiration (BS)
25 Early intervention (EI): therapy and early Improved motor outcomes (BS Blauw- Evidence supports general stimulation, 1 Moderate Weak + Yellow
education to promote acquisition of and A) Hospers93 developmental approaches and parent coaching MEASURE
milestones, via group or individual stimulus Blauw programmes. Gains are superior to NDT or 1
Hospers94 traditional physiotherapy
Turnbull95 1
Ziviani96 1
Improved cognitive outcomes Blauw- High quality evidence supports EI in non-CP 1 Low Weak + Yellow
(BS) Hospers93 populations. Moderate evidence supports EI MEASURE
Blauw program memes for at risk pre-term infants, 1
Hospers94 aimed at mimicking the intrauterine environment
Turnbull95 1
Ziviani96 1
26 Electrical stimulation (ES, NMES, FES): Improved gait parameters (BS) Cauraugh97 Insufficient evidence. Effective in laboratory, 1 Low Weak + Yellow
electrical stimulation of a muscle through a Wright97 unknown effectiveness in the community 1 MEASURE
skin electrode to induce passive muscle Improved muscle strength (BS) Kerr98 Lower-quality supporting evidence 1 Moderate Weak + Yellow
contractions for strengthening or motor Scianni99 1 MEASURE
activation Wright97 1
Augmented effects of Lannin100 Conflicting evidence. More evidence needed 1 Low Weak + Yellow
Botulinum toxin (BS) Wright101 1 MEASURE
27 Fitness training: planned structured activities Improved aerobic fitness (BS) Butler102 Effective short term and only in those that have 1 Moderate Strong + Green GO
involving repeated movement of skeletal Rogers103 sufficient motor skills to undertake aerobic 1
muscles that result in energy expenditure to Verschuren102 training. No carryover when training stops. 1
improve or maintain levels of physical fitness Therefore do use but only in the right patient and
plan to continue the programme long term
Improved function and Butler102 Insufficient evidence. Aerobic fitness does not 1 Moderate Weak Yellow
participation (A and P) Rogers103 appear to translate to activity and participation 1 MEASURE
Verschure n102 gains 1
28 Fundoplication (including Nissen and Reduction of gastro- Vernon- No CP-specific evidence 1 N/A Weak + Yellow
laparoscopic; gastric plication): surgical oesophageal reflux (BS) Roberts103 MEASURE
procedure to strengthen the barrier to acid
reflux, e.g. by wrapping the fundus around the
oesophagus
Review
891
Table I: Continued
GRADE
Oxford Quality
29 Gastrostomy: surgical placement of a non-oral Improved growth and weight Arrowsmith104 Adverse events occur 3 Very low Weak + Yellow
feeding tube to prevent or reverse growth (BS) Kong105 3 MEASURE
failure, or prevent aspiration pneumonia, e.g. Samson- 1
percutaneous endoscopic gastrostomy (PEG), Fang106 1
jejunostomy Sleigh107,108 1
Sullivan109 3
Sullivan110 3
Vernon- 4
Roberts111
30 Goal-directed training/functional training: task Improved gross motor function Ketelaar112 Effective. Some probability of bias within included 2 Low Weak + Yellow
specific practice of child-set goal-based (A) Lowing113 studies 3 MEASURE
activities using a motor learning approach Improved hand function (A) Novak13 Effective. Can be delivered via a home programme 2 High Strong + Green GO
Sakzewski52 or used in combination with CIMT and bimanual 2

Wallen14 training. Low probability of bias within included 2
studies
Improved self-care (A) Novak13 Effective. Low probability of bias within included 2 High Strong + Green GO
Wallen14 studies 2
31 Hand surgery: surgery to improve hand Improved thumb-in-palm Smeulders114 Lower-quality supporting evidence 1 Very low Weak + Yellow
function and alignment posture (BS) MEASURE
115
32 Hip surgery: orthopaedic surgery to improve Reduced hip subluxation via Stott Most studies were uncontrolled 1 Very low Weak + Yellow
musculoskeletal alignment of the hip soft tissue surgery (adductor MEASURE
release) (BS)
Reduced hip subluxation via Brunner116 Studies were retrospective and uncontrolled 4 Very low Weak + Yellow
bony surgery (BS) Huh117 4 MEASURE
33 Hip surveillance: active surveillance and Reduced hip dislocation and Gordon118 Hip surveillance is a regular assessment process 1 Moderate Strong + Green GO
treatment for hip joint integrity to prevent hip need for orthopaedic surgery so as the right treatments can be provided in a
dislocation (BS) timely manner, as such the studies were
appropriately designed as observational studies
not RCTs. Do use as there are substantive
adverse events from no surveillance
34 Hippotherapy: therapeutic horse riding to Improved hip and trunk Snider119 Effective 1 Low Weak + Yellow
practice balance and symmetry symmetry and stability (BS) Sterba120 1 MEASURE
Zadnikar121 1
Improved gross motor function Whalen122 Effective. Larger studies needed 1 Low Weak + Yellow
(A) MEASURE
123
Improved participation (P) Davis Insufficient evidence. Sensitive measures required 2 Moderate Weak Yellow
in future studies MEASURE
124
35 Home programmes: therapeutic practice of Improved performance of Novak Effective. Note: a single rigorous RCT shows 1 Moderate Strong + Green GO
goal-based tasks by the child, led by the functional activities (A) Novak13 effectiveness, with a low probability of bias 2
parent and supported by the therapist, in the Improved participation (P) Novak13 Insufficient evidence. Sensitive measures required 2 Moderate Weak Yellow
home environment in future studies MEASURE
125
36 Hydrotherapy: aquatic-based exercises Improved vitals and gross Chrysagis Lower-quality supporting evidence 2 Low Weak + Yellow
motor function (BS and A) Getz126 1 MEASURE
Gorter127 1
37 Hyperbaric oxygen (HBO): inhaled 100% oxygen Improved performance of Collet128 Ineffective. Adverse events can also occur 2 High Strong Red STOP
inside a pressurized hyperbaric chamber functional activities (A) McDonagh129 1
Table I: Continued
GRADE
Oxford Quality
38 Intrathecal baclofen (ITB): antispasticity Reduced lower limb spasticity Butler130 Predominantly low-quality supporting evidence. 1 Low Weak + Yellow
medication delivered directly to the spinal (BS) Creedon131 The size of the gains varies between studies 1 MEASURE
cord via a pump surgically implanted within Dan132 1
the abdomen Delgado26 1
Kolaski133 1
Reduced upper limb spasticity Butler130 Insufficient evidence. The effect on upper limb is 1 Low Weak Yellow
(BS) Creedon131 less than for the lower limb and some authors 1 MEASURE
Dan132 question whether ITB is clinically worthwhile for 1
Delgado26 the purposes of reducing upper limb spasticity 1
Kolaski133 1
Reduced dystonia (BS) Albanese134 Lower-quality supporting evidence 1 Very low Weak + Yellow
Butler130 1 MEASURE
Improved function and health Hoving135 Lower-quality supporting evidence 2 Low Weak + Yellow
related quality of life (A, P and Hoving136 2 MEASURE
PF) Kolaski133 1
Improved walking ability in Pin137 Insufficient evidence. Some children with CP 1 Very low Weak Yellow
ambulant children (A) improve but many experience adverse events MEASURE
including inability to walk
39 Massage: therapeutic stroking and circular Reduced pain (BS) Hernandez- Conflicting evidence 2 Low Weak + Yellow
motions applied by a massage therapist to Reif138 MEASURE
muscles to relieve pain and tension Nilsson139 2
Reduced spasticity (BS) Alizad140 Conflicting evidence 2 Low Weak + Yellow
Hernandez- 2 MEASURE
Reif138
Improved function (A) Hernandez- Conflicting evidence 2 Low Weak + Yellow
Reif138 MEASURE
40 Neurodevelopmental therapy (NDT, Bobath): Normalized movement (BS) Brown141 Ineffective. No gains superior to other treatments 1 Low Strong Red STOP
direct, passive handling and guidance to Butler142 1
optimise function Prevent contracture Brown141 Ineffective because immediate gains in range of 1 Low Strong Red STOP
development (BS) Butler142 motion observed within the session do not carry 1
over
Improved function (A) Brown141 Conflicting systematic review evidence. Early 1 Low Weak Yellow
Butler142 reviews suggested no benefits. The more recent 1 MEASURE
Martin143 review included one new trial suggesting 1
possible benefit of higher doses of NDT
compared with lower doses of NDT; however,
this is not a conventional method for establishing
treatment efficacy and should be interpreted with
caution. Other evidence shows that motor
learning produces superior functional gains to
NDT
Enhanced social emotional and Brown141 Ineffective. No evidence to support claim 1 Low Strong Red STOP
cognitive skills (BS and PF) Butler142 1
41 Occupational therapy after BoNT: improved Improved goal achievement of Boyd59 Effective 1 High Strong + Green GO
hand use via CIMT, goal-directed training, upper limb activities (A) Fehlings65 1
strength training and functional hand splints. Hoare70 1
improved symptom management via casting Hoare71 1
and immobilisation splints Lannin99 1
Review
893
Table I: Continued
GRADE
Oxford Quality
42 Oral motor treatment: sensory stimulation to Improved verbal speech as a – No evidence in CP. Insufficient evidence to support – N/A Weak Yellow
lips, jaw, tongue, soft palate, larynx, and result of non-speech oral or refute in non-CP200 populations MEASURE
respiratory muscles to influence the motor exercises (BS)
oropharyngeal mechanism Improved safety of swallowing Snider92 Insufficient evidence 1 Very low Weak Yellow
and reduced drooling (BS) Wilcox144 1 MEASURE
43 Orthopaedic surgery: surgical prevention or Correct equinus foot deformity Shore145 Lower-quality supporting evidence with no 1 Low Weak + Yellow
correction of musculoskeletal disorders and (BS) superior surgical technique evident. Studies MEASURE
associated muscles, joints, and ligaments, e.g. indicated that early surgery was a major risk
muscle lengthening factor for recurrent equinus deformity
44 Orthotics (splints): removable external devices Improved stride length and Autti-Ramo76 Positive effects on ankle range of motion, gait 1 Very low Weak + Yellow
designed to support weak or ineffective joints range of motion via AFOs (BS) Blackmor e77 kinetics and kinematics, but the quality of the 1 MEASURE
or muscles Effgen78 evidence is low 1
Figueiredo146 1
Harris147 1
Morris148 1
Teplicky81 1

Improved lower limb function Autti-Ramo76 Insufficient evidence 1 Very low Weak Yellow
(A) Blackmore77 1 MEASURE
Effgen78 1
Figueiredo146 1
Harris147 1
Morris148 1
Teplicky81 1
Improved upper limb function Teplicky81 Insufficient evidence 1 Very low Weak Yellow
(A) MEASURE
Prevention of contracture (BS) Teplicky81 High-quality evidence shows ineffective in non-CP 1 Very low Weak Yellow
populations, but insufficient CP studies to be MEASURE
certain
Prevention of hip dislocation Graham149 High-quality evidence shows may slow hip 2 High Strong Red STOP
via hip orthoses and dislocation rate slightly but essentially ineffective
botulinum toxin (BS) for preventing hip dislocation
45 Parent training: educating and coaching parents Improved parenting skills to Whittingham50 Lower-quality supporting evidence 1 Very low Weak + Yellow
to change their child’s behaviour or skills, plus facilitate child development MEASURE
improve parenting (E)
46 Phenol: muscular injections to induce chemical Reduce spasticity locally (BS) Delgado26 No CP studies appraised. Since high-quality 1 N/A Weak Yellow
denervation for treating local spasticity evidence supports BoNT-A. However, in clinical MEASURE
care, phenol is sometimes used positively in
combination with BoNT-A to enable injection of
more muscles groups to remain within safe total
dose restrictions
47 Play therapy: play and creative arts to enhance Improved play skills (A) Redditi Insufficient evidence 2 Low Weak + Yellow
emotional wellbeing and advance play skills Hanzlik150 MEASURE
Improved child coping and – No evidence in CP – N/A Weak + Yellow
reduced stress (BS and PF) MEASURE
48 Pressure care: prevention of pressure ulcers via Reduced ulcer development via McInnes151 Effective. Alternating pressure mattresses more 1 Low Strong + Green GO
good positioning, repositioning, and suitable high-specification foam cost-effective than alternating pressure overlays
support surfaces mattresses, alternating
pressure mattresses, and
medical grade sheepskins (BS)
Reduced ulcer development McInnes151 Insufficient evidence 1 Low Weak + Yellow
from wheelchair seat cushions MEASURE
(BS)
Table I: Continued
GRADE
Oxford Quality
49 Respite: temporary caregiving break for parents Improve family functioning and Strunk152 Lower-quality supporting evidence 1 Very low Weak + Yellow
where the child is usually accommodated reduce parental stress (E) MEASURE
outside the home
50 Seating and positioning: assistive technology Improved pulmonary function Farley153 Lower-quality supporting evidence 1 Very low Weak + Yellow
that enables a person to sit upright with (BS) Ryan154 1 MEASURE
functional, symmetrical or comfortable Improved posture and postural Chung155 Insufficient evidence 1 Very low Weak + Yellow
posture, to enable function control (BS) Farley153 1 MEASURE
Roxborough156 1
Ryan154 1
Improved hand function (A) Farley153 Lower-quality supporting evidence 1 Very low Weak + Yellow
McNamara157 1 MEASURE
Ryan154 1
Reduced pressure via tilt (BS) Michael158 Lower-quality supporting evidence 1 Very low Weak + Yellow
MEASURE
159
51 Selective dorsal rhizotomy (SDR): neurosurgical Reduced spasticity (BS) Grunt Effective 1 Moderate Strong + Green GO
procedure that selectively severs nerve roots McLaughlin160 1
in the spinal cord, to relieve spasticity Steinbok161 1
Improved gait kinematics (BS) Grunt159 Effective 1 Low Strong + Green GO
McLaughlin160 1
Steinbok161 1
Improved function and Grunt159 Evidence for improved gross motor function, but 1 Very low Weak + Yellow
participation (A and P) McLaughlin160 no supporting evidence for general activities and 1 MEASURE
Steinbok161 participation and this should not be the primary 1
goal for using SDR
52 Sensory integration (SI): therapeutic activities Improved sensory organization Vargas162 Ineffective. Since meta-analyses of SI compared 1 Low Strong Red STOP
to organize sensation from the body and (BS) with no treatment had average effect sizes of 0.03
environment, to facilitate adaptive responses, (for most recent studies)
e.g. hammock swinging Improved motor skills (A) Vargas162 Ineffective. Since Goal directed training, CIMT or 1 Low Strong Red STOP
bimanual therapy exist as effective alternatives.
Meta-analyses of SI compared to no treatment
had average effect sizes of 0.03 (for most recent
studies), and 0.09 for SI compared to alternative
treatments. Note reviews for non-CP populations
excluded
53 Sensory processing: therapeutic activities to Improved function (A) – No evidence in CP. Since performance-based – N/A Weak Yellow
organize more appropriate responsiveness approaches (e.g. CO-OP) are more favourable MEASURE
(i.e. not hyper-responsive and not than impairment-based approaches, e.g. sensory
hyporesponsive) to task and environmental processing (in non-CP populations)
demands, including self-regulation
54 Single event multilevel surgery with therapy: Improved long-term functional McGinley163 Lower-quality supporting evidence 1 Very low Weak + Yellow
multiple simultaneous surgical procedures at mobility (A) MEASURE
different levels of the lower limb to either
improve gait or prevent deterioration
55 Social stories: an individualized book describing Improved communication and Test164 Insufficient evidence 1 Very low Weak + Yellow
a situation, skill, or concept and the relevant management of emotions and MEASURE
social cues, perspectives, and common behaviours (A)
responses to prepare a child for a social
situation
56 Solution-focused brief therapy: resource Reduced parental depression, – No evidence in CP. Since low-quality evidence – N/A Weak + Yellow
orientated and goal focused approach to improved coping and shows emergent effectiveness in non-CP MEASURE
Review
generating solutions to life challenges improved parenting skills (E) populations198
895
Table I: Continued
GRADE
Oxford Quality
57 Strength training (resistance): use of Improved lower limb strength Dodd165 Effective short term for improving muscle strength. 1 Low Weak + Yellow
progressively more challenging resistance to via progressive resistance Effgen78 Improved muscle strength does not carry over to 1 MEASURE
muscular contraction to build muscle strength training (BS) Jeglinsky166 function, other treatment approaches will be 1
and anaerobic endurance Martin143 needed for functional gains 1
Mockford167 1
Scianni168 1
Taylor169 1
Improved upper limb strength Kim170 Effective short term for improving muscle strength 2 Low Weak + Yellow
via progressive resistance MEASURE
training (BS)
Improved function via Scianni168 Insufficient evidence 1 Low Weak Yellow
progressive resistance training MEASURE
(A)
Improved function via Martin143 Lower-quality supporting evidence 1 Low Weak + Yellow
functional training using MEASURE
resistance within functional

tasks (A)
58 Stretching: use of an external passive force Contracture prevention via Katalinic79 Ineffective. Comprehensive and robust meta- 1 Moderate Weak Yellow
(e.g. parent) exerted upon the limb to move it manual stretching (BS) Wiart171 analysis showed no immediate, or short– to 1 MEASURE
into a new and lengthened position medium-term benefits (<7mo), but, since only a
small number of CP studies were included within
the review, it is not possible to be certain about
this recommendation for CP
Contracture prevention via Autti-Ramo76 Insufficient evidence 1 Low Weak + Yellow
splinting or positioning (BS) Pin45 1 MEASURE
Teplicky81 1
59 Therasuits: a breathable soft dynamic orthotic Improved gross motor function Alagesan172 Conflicting evidence. One trial suggests positive 2 Low Weak Yellow
full body suit, designed to improve (A) Bailes173 effect the other suggest no benefits 2 MEASURE
proprioception, reduce reflexes, restore
synergies and provide resistance
60 Tizanidine: antispasticity medication Reduce spasticity (generalized) Delgado26 Insufficient evidence 1 Low Weak + Yellow
(BS) MEASURE
61 Treadmill training: walking practice on a Improved weight bearing (BS) Zwicker174 Lower-quality supporting evidence 1 Low Weak + Yellow
treadmill, which includes partial body support MEASURE
Improved functional walking Damiano175 Lower-level supporting evidence. However, 1 Low Weak + Yellow
(A) Mutlu176 overground walking more effective than partial 1 MEASURE
Willoughby177 body weight-supported treadmill training 1
Zwicker174 1
62 Vitamin D (with our without calcium or growth Improved bone mineral density Fehlings55 Insufficient evidence 1 Low Weak + Yellow
hormones): dietary vitamin supplement for (BS) Hough56 1 MEASURE
bone density
63 Vojta: therapist applied pressure to defined Improve strength and Brandt178 Conflicting evidence. Studies claim to ‘cure’ early 2 Very low Weak Yellow
zones on the body whilst positioned in prone, movement, plus lessen d’Avignon179 CP, which is not consistent with any of the other 2 MEASURE
supine or side lying, where the stimulus leads severity of CP (BS) Kanda180 literature about CP having no known cure. Also 3
to automatically and involuntarily complex Liu181 the studies reported high dropout rates due to 2
movement Wu182 child distress. Studies have a high probability of 2
Zhang183 bias, e.g. lack of: random sequence generation; 2
Zhao188 concealed allocation, study blinding, 2
psychometrically sound instruments; plus
incomplete outcome data collection and selective
reporting
using the Oxford Levels of Evidence; a categorization
MEASURE
MEASURE
Traffic light
would not; ‘Strong ’ means ‘do not do it’, indicating a judgement that most well-informed people would make.17 A, activities; BS, body structures and function; P, participation; RCT, ran-
make but a substantial minority would not; ‘Weak ’ means ‘probably do not do it’, indicating a judgement that a majority of well-informed people would make but a substantial minority
means ‘do it’, indicating a judgement that most well-informed people would make; ‘Weak +’ means ‘probably do it’, indicating a judgement that a majority of well-informed people would
action
using GRADE; a colour coding scheme using the Evidence
Under ‘Quality of evidence’, ‘High’ means that further research is very unlikely to change our confidence in the estimate of effect; ‘Moderate’ means that further research is likely to have
an important impact on our confidence in the estimate of effect and may change the estimate; ‘Low’ means that further research is very likely to have an important impact on our confi-
Yellow
Yellow
dence in the estimate of effect and is likely to change the estimate; and ‘Very low’ means that any estimate of effect is very uncertain.17 Under ‘Strength of recommendation’, ‘Strong +’
Alert Traffic Light system, and an ICF domain (Table I).
More specifically, each intervention outcome sought by
Strength of rec-
included study authors was assigned an ICF domain based
ommendation
upon published literature.176 It has been acknowledged in
the literature that ICF coding is notoriously complex to
Weak
Weak
GRADE
apply since CP is a disability not a disease, and thus direct

interventions do not ultimately alter underlying disease
processes.10 To overcome this challenge, we applied ICF
evidence
Very low
Very low
Quality
codes using CP literature precedents, where the outcome

of
measure within the included trials had been ICF coded by

other authoritative researchers.10 Of note, ICF linking
evidence
Oxford
level
rules typically cluster together (1) body structure and func-

tions; and (2) activities and participation. To prevent loss
1
of findings obscured within aggregated data, we separated

population, but no effect in CP. Small numbers of
activities from participation because we wanted to illumi-

Lower-quality supporting evidence in non-CP
nate whether or not participation outcomes were being

CP studies we cannot be certain about this
achieved. All the data required to answer the study ques-

tions were published within the papers, so no contact with
Panel comments
authors was necessary.
Ethics and registration

recommendation for CP
The study did not involve contact with people, so the need
Insufficient evidence
for ethical approval was waived by the Cerebral Palsy Alli-

ance’s Human Research ethics committee. This systematic
review was not registered.
RESULTS
Using the search strategy, 33 485 citations were identified,
of which 166 articles met the inclusion criteria for review
Citations
del Pozo-
del Pozo-
(Fig. 2).
Cruz185
Cruz185
Participants
Intervention outcome (ICF level)
For the purpose of this study, participants had CP, which is

a complex and heterogeneous condition. We included stud-
Improved gait (BS and A)
ies about children with CP of any motor subtype (spastic,

Improved strength (BS)
dyskinetic, or ataxic), any topography (hemiplegic/unilat-

domized controlled trial; E, environmental; PF, personal factors.
eral, diplegic/bilateral, or quadriplegic/bilateral), and any

functional ability level (Gross Motor Function Classifica-
tion System [GMFCS]188 levels I to V and Manual Ability
Classification System [MACS]189 levels I to V). There was
substantial emphasis in the medical literature on interven-
Whole-body vibration: assistive technology that
tions to reduce spasticity, the most prevalent motor impair-

transmits low-frequency vibration to the body
through a broad contact area of a vibrating
ment.190 There was also a heavy emphasis in the therapy

surface, e.g. feet in standing, buttocks in
literature on interventions designed to improve motor out-

comes consistent with CP being a physical disability. The
higher-quality studies defined the child’s motor function
abilities using the GMFCS and MACS to enable better
interpretation of treatment effects taking into account the
sitting, or whole body
severity of the disability. However, there was insufficient

homogeneity of reporting across studies to enable reporting
Table I: Continued
by GMFCS level, which was our original intended strategy.

Intervention
Levels of evidence and ICF

High levels of evidence existed in the literature summariz-
ing interventions for children with CP (Table I). Of the
64
Review 897
Grade of evidence Traffic alert action
Quality Recommendation
High Strong + Green: Go
Moderate
Favourable
Low
Weak +
Very low
Yellow: Measure
Very low
Unfavourable
Weak –
Low
Moderate
High Strong – Red: Stop

High = Further research is very unlikely Strong + Do it Green = Go: Effective, therefore do it
to change our confidence in the Weak + Probably do it
estimate of effect Yellow = Measure: Uncertain effect,
Moderate = Further research is likely to have Weak – Probably don’t do it therefore measure
an important impact on our confidence Strong – Don’t do it outcomes to determine
in the estimate of effect and may if progress is made
change the estimate
Red = Stop: Ineffective, therefore don’t do it
Low = Further research is very likely to
have an important impact on our
confidence in the estimate of effect
and is likely to change the estimate
Very low = Any estimate of effect is very uncertain
Figure 1: Relationship between the GRADE and Traffic Light System.
166 included studies, the breakdown by level of evidence

as rated on the Oxford Levels of Evidence was level 1
(n=124), 74%; level 2 (n=30), 18%; level 3 (n=6), 4%; and
n = 148
level 4 (n=6), 4%.
Total no. of CINAHL
Cochrane n = 77 When the included articles were tallied in 5-year inter-
articles DARE n = 118
(n = 33 485) ERIC n = 16 vals by publication date, it was clear that the number of
Google scholar n = 308 000 systematic reviews published about CP intervention had
Medline n = 154
OT seeker n = 32 exponentially increased in recent years (Fig. 3).
PEDro n = 72 Almost none (2 of 166) of the systematic reviews
No. of potential PsychBITE n=1
articles after PsychINFO n = 66 retrieved graded the body of evidence summarized using
PubMED n = 1996
deletion by
SpeechBITE n=5 the GRADE system. We therefore carried out assignment
title (n = 231) of GRADEs using the recommended expert panel method-
Reasons for deletion included: ology. Using the GRADE system, of the 64 different CP
Not systematic reviews; not interventions reviewed across 131 intervention outcomes
No. of potential
>25% sample had cerebral
articles after 16% of outcomes assessed (n=21) were graded ‘do it’ (i.e.
palsy; duplicates; protocols
deletion by green light, go interventions); 58% (n=76) were graded
only; and practice guidelines
abstract ‘probably do it’ (i.e. yellow light, measure outcomes); 20%
(n = 171)
(n=26) were graded ‘probably do not do it’ (i.e. yellow
light, measure outcomes; see Fig. 1); and 6% (n=8) were
No. of potential graded ‘do not do it’ (i.e. red light, stop interventions; see
articles after Fig. 1). In line with the appraisal criteria for this review,
deletion by full- occupational therapy, physiotherapy, and medicine were
text (n = 166)
the disciplines that encompassed the highest number of
proven effective interventions for CP within their evidence
Systematic reviews n = 124
Total no. of base, which is not surprising given the long historical
RCTs n = 25
articles included Non-randomised n=7 research emphasis on redressing the physical aspects of
(n = 166) Case series n=4
CP. In the fields of psychology, speech pathology, social
work, and education, the evidence base for all interventions
reviewed was lower level or inconclusive (yellow), but, in
Figure 2: Flow diagram of included articles. keeping with interdisciplinary care, psychologists and social

80 ing ankle range of motion; (3) hip surveillance for main-
Number of published cerebral palsy n = 72
intervention systematic reviews
taining hip joint integrity; (4) constraint-induced
70
movement therapy, bimanual training, context-focused
60 therapy, goal-directed/functional training, occupational
50 therapy following BoNT, and home programmes for
improving motor activity performance and/or self-care;
40 n = 35 (5) fitness training for improving fitness; (6) bisphospho-
30 nates for improving bone density; (7) pressure care for
20 reducing the risk of pressure ulcers; and (8) anticonvulsants
n = 14
for managing seizures (despite no CP-specific anticonvul-
10
n=3 sant evidence existing, the panel rated the strength of the
0 recommendation as strong plus (do it) because good-qual-
1993–1997 1998–2002 2003–2007 2008–2012
ity evidence supports anticonvulsants in non-CP popula-
Years published tions,191 and serious harm, even death, can arise from no
treatment).
Green light effective interventions were mapped against
Figure 3: Number of published cerebral palsy intervention systematic the ICF by the outcomes that had been measured in the
reviews. literature and the corresponding traffic light code was
applied (Table II). First, Table II shows that green-light
workers applied high-level evidence from other diagnostic effective interventions were all aimed at either the body
groups (e.g. bimanual, cognitive behaviour therapy, coun- structures and function level or the activities levels on the
selling, Triple P49). In the field of speech pathology, it is ICF. The conspicuous finding here was that there were no
worth noting that it is difficult to conduct studies of aug- proven effective interventions for addressing the participa-
mentative and alternative communication (AAC) using tion, environment, or personal factors levels of the ICF,
conventional rigorous methodologies because included par- even though these are philosophical priorities. Second,
ticipants often have different disability types and, accord- Table II shows that when effective body structures and
ingly, differing levels of expressive, receptive, and social functions interventions were measured for an effect at the
communication abilities. AAC interventions require multi- activities level (all of the time) evidence of effect was either
factorial measurement because effective device utilization lower level or inconclusive and, therefore, was coded yel-
relies on changes in all of these domains from best-practice low light. In other words, the positive effects of body
speech, language, and teaching strategies and from chang- structure interventions did not translate ‘upstream’ to the
ing the mode of communication. Thus, adequately measur- activities level. This finding seems to suggest that you ‘get
ing and attributing interventions effects to each component what you give’. This finding has, however, an alternative
of these integrated treatment approaches remains challeng- interpretation – we do not yet know if body structures and
ing. Amongst the alternative and complementary medicine functions intervention improves outcomes at the activities
interventions offered by some clinicians, the findings were level because of the measurement artefact created by ran-
of even poorer quality, because an even greater proportion domized trials only being powered to detect change in one
of the interventions were proven ineffective. However, the primary end-point. Third, Table II shows that green light
real rate of ineffective alternative and complementary inter- activity-level interventions were effective at the activities
ventions may be even higher as so many had to be level of the ICF, but minimal measurement had been
excluded from this review as a result of the lack of any undertaken to illuminate whether or not there was also any
published peer-reviewed literature about the approaches translation of impact ‘downstream’ to the body structures
(e.g. advanced biomechanical rehabilitation). and functions level.
Each intervention was coded using the ICF by the inter-
vention’s desired outcome. Out of the 131 intervention Yellow light measure outcomes interventions
outcomes for children with CP identified in this study, A high proportion (70%) of the CP interventions within
n=66 (51%) were aimed at the body structures and func- clinical care had either lower-level evidence supporting
tion level; n=39 (30%) were aimed at the activity level; n=7 their effectiveness or inconclusive evidence, including acu-
(5%) were aimed at the participation level; n=8 (6%) were puncture; alcohol (intramuscular injections for spasticity
aimed at the environment level; and the remaining n=11 reduction); AAC; animal-assisted therapy; assistive technol-
(8%) were aimed at combinations of ICF levels. ogy; baclofen (oral); behaviour therapy and coaching;
cognitive behaviour therapy; communication training;
Green light go interventions conductive education; counselling; oral dantrolene; dyspha-
In the papers retrieved, the following CP interventions gia management; early intervention (for motor out-
were shown to be effective: (1) botulinum toxin (BoNT), comes); electrical stimulation; fundoplication; gastrostomy;
diazepam, and selective dorsal rhizotomy for reducing hand surgery; hip surgery; hippotherapy; hydrotherapy;
muscle spasticity; (2) casting for improving and maintain- intrathecal baclofen; massage; orthoses; oral–motor
Review 899
Table II: Green light interventions (and their other indications) by level of ICF
ICF level
Body
structures
and Personal
Intervention function Activity Participation Environment factors
Body structures and function interventions

1. Anticonvulsants G
2. Botulinum toxin G
3. Bisphosphonates G
4. Casting (ankle) G Y
5. Diazepam G
6. Fitness training G Y Y
7. Hip surveillance G
8. Pressure care G
9. Selective dorsal rhizotomy G Y Y
Activities interventions
10. Bimanual training G
11. Constraint-induced movement therapy G
12. Context-focused therapy G
13. Goal-directed training/functional training G
14. Home programmes G Y
15. Occupational therapy post botulinum toxin G
(upper limb)
G=green intervention when aimed at this level of the International Classification of Functioning, Disability and Health (ICF); Y=yellow
intervention when aimed at this level of the ICF.
therapy; orthopaedic surgery; parent training; phenol designs used, serious methodological flaws, the relevance
(intramuscular injections); play therapy; respite; seating and and sensitivity of the outcomes measures adopted, the diffi-
positioning; sensory processing; single-event multilevel sur- culty in assembling large homogeneous samples for niche
gery; social stories; solution-focused brief therapy; strength interventions, and most authors concluded that more rigor-
training; stretching; therasuits; oral tizanidine; treadmill ous research was needed.
training; oral vitamin D; Vojta; and whole-body vibration.
It is important to note that cognitive–behavioural ther- Red light stop interventions
apy,192–196 early intervention,196–198 parent training,49,50 Craniosacral therapy, hip bracing, hyperbaric oxygen,
and solution-focused brief therapy199 all have good-quality NDT, and sensory integration have all been shown to be
supporting evidence in non-CP populations. It is also ineffective in children with CP, and are therefore not rec-
important to note that oral–motor therapy200 and sensory ommended for standard care. Appropriately, effective alter-
processing201 have equivocal evidence in non-CP popula- natives exist that seek to provide the same clinical outcome
tions for which they were designed, and so there is no of interest.
strong or compelling reason to think either intervention To assist with comparative clinical decision-making
would work better in CP. Of note, there was great variabil- amongst intervention options for the same desired out-
ity in the volume and quality of the evidence available at come, we mapped the interventions that seek to provide
the yellow-light level. For example, some intervention evi- analogous outcomes using bubble charts. In the bubble
dence bases were downgraded to low quality, as per the charts, the size of the circle correlated to the volume of
GRADE guidelines for dealing with imperfect randomized published evidence. The circle size was calculated using
controlled trials (e.g. hippotherapy and biofeedback). How- (1) the number of published papers on the topic; and (2)
ever, for some interventions simply next to no evidence has the total score for the level of evidence (calculated by
been published and what has been published involves reverse coding of the Oxford Levels of Evidence, i.e. expert
very small numbers and is of low quality (e.g. whole-body opinion=1, randomized controlled trial [RCT]=5). The
vibration). location of the circle on the y-axis of the graph corre-
The yellow-light included reviews that could not dem- sponds to the GRADE system rating. The colour of the
onstrate robust evidence of effectiveness when strict sys- circle correlates to the Evidence Alert System (Fig. 4).
tematic review criteria about design quality, adequate
sample size, and independent replication were used to DISCUSSION
judge the evidence. Yellow-light reviews contained only High levels of evidence existed in the literature summarizing
marginal amounts of good-quality evidence when criteria intervention options for children with CP. Akin to other
were applied to reduce the possibility of biases explaining fields of medicine and allied health, there has been an expo-
the proposed treatment benefits. Most yellow-light system- nential increase in the number of systematic reviews pub-
atic review authors commented upon the low quality of the lished about CP intervention6 revealing the emergence of

a
Spasticity Contracture Improved
Effective management muscle
management
strength
S+ Do it
Botulinum toxin
SDR
Diazepam Casting
lower limb
Strength
W+ Probably do it Casting training
Tizan- Orthotics lower limb
ITB upper
AFOs
idine limb
Strength
training
Orthotic upper
SEMLS hand limb
Baclofen Dantr-
oral oline ES
Ortho- Hand
paedics surg- Whole
? Unknown in CP body
vibration
Alcohol Phenol
Worth it line
Stretching
manual Vojta
Casting
W– Probably don’t do it
S– Don’t do it NDT
Ineffective
b Improved motor activities Improved function

& self care
Goal-
directed Goal-
CIMT training Bimanual directed
training training
Home
programs
OT post
Context botulinum
focused Home
toxin programs
therapy
Seating
EI Treadmill
Hippo-
Bio- training Botulinum toxin
therapy
feedback
SEMLS Assistive
technology SDR
Hydro- &
therapy therapy
Orthotic
hand ITB
Sensory
Proc.
Conductive Vojta Massage

Education
Thera-
suits
NDT
NDT Hyper- SI
baric 02
Figure 4: State of the evidence for cerebral palsy intervention by outcomes.
Review 901
c Improved Mealtime Improved Improved Improved
communication management behaviour parent bone
& social skills coping density
Bisphos-
phonates
Behaviour Behaviour
therapy therapy
Communication Gastrost
Social Communication Assistive
training omy stories training technology
Coach- standing
Dysphagia ing frames
Play
management therapy
AAC Solut- Vitamin
ion D
Social Fundopli focus Whole
stories cation Coun-
CBT body
selling vibration
Oro- Oro-
motor motor
Figure 4: Contiuned.
highly effective prevention interventions.186,187 There is no or self-care function; (4) BoNT, diazepam, or selective
reason to think that this trend may decline. This finding has dorsal rhizotomy for spasticity management; (5) fitness
important implications for managers, knowledge brokers, training for aerobic fitness; (6) pressure care for reducing
and clinicians about finding effective and efficient ways for the risk of ulcers; (7) bisphosphonates for improving bone
health professionals to remain up to date with the latest mineral density; and (8) anticonvulsants for managing sei-
practice. Best available knowledge translation evidence sug- zures. When delivering interventions to children with CP,
gests that managers and senior clinical mentors can help it is paramount that clinicians choose evidence-based inter-
staff maintain up-to-date knowledge via interactive evi- ventions at the activities and participation level that hone
dence-based practice continuing education sessions and the child’s strengths and reflect their interests and motiva-
journal clubs, but multiple tailored strategies will be tions, and ultimately seek to help children live an inclusive
required to change their use of evidence.202 This systematic and contented life. However, when choosing interventions
review could form the basis of policy, educational, and at the body structure and functions level, the primary pur-
knowledge translation material because it is a comprehensive pose is to mitigate the natural history of CP (such as hip
summary of the evidence base. dislocation) and the probable physical decline from second-
ary impairments,118 rather than trying to fix the condition.
Recommendations for practice We must also remain mindful that conflicts can arise
Based upon the best available evidence, standard care for between what families hope for and what the evidence sug-
children with CP should include the following suite of gests will be helpful or is realistically possible.202 Part of
interventions options (where the interventions would being truly family centred is to act as an information
address the family’s goals): (1) casting for improving ankle resource to the family, which will include honest and open
range of motion for weight bearing and/or walking; (2) hip disclosure about prognosis using evidence-based tools to
surveillance for maintaining hip joint integrity; (3) biman- guide these difficult conversations.203 Similarly, designing
ual training, constraint-induced movement therapy, con- services based upon goals set by the family5,64 is best prac-
text-focused therapy, goal-directed/functional training, tice and can also help to set the scene for discussing what
and/or home programmes for improving motor activities is realistic and possible from intervention.

Going forward, systematic and disciplined use of out- recommendations for evaluating these eight domains in a
come measures within all specialties is required for way that was sensitive to change. The first of these eight
generating new evidence and confirming treatment effects domains is impairment, which can be subdivided into (1)
of commonly used interventions. Routine outcome spasticity, measured using the Modified Tardieu Scale5,64
measurement is especially important when yellow-light and (2) fine motor, measured using the Melbourne Assess-
interventions are being applied, and could circumnavigate ment of Unilateral Upper Limb Function11 and the Quality
some of the genuine research barriers including low avail- of Upper Extremity Skills Test.11 The second domain is
ability of research funds and difficulties in assembling large general health. Valid and reliable instruments exist regarding
homogenous samples. This recommendation is particularly general health in the literature, but less is understood about
vital for the fields of speech pathology, social work, and whether these measures are sensitive to change in CP, and
psychology that provide key services to children with CP, therefore no recommendations are made at this juncture.
without strong evidence, as of yet, to support their prac- Third is the gross motor skills domain, measured using the
tice. These professions have been overshadowed in the CP Gross Motor Function Measure.73,206,207 The fourth domain
research arena until recently, when the field stopped solely is self-care/fine motor skills, which can be subdivided into
redressing physical impairments and started to look further (a) self-care, measured using the Paediatric Evaluation of
afield to engendering outcomes in well-being and partici- Disability Inventory206 and the Activities Scale for
pation. In addition, systematic and disciplined use of out- Kids207,208 and (b) fine motor, measured using the Assisting
come measures is also needed when prescribing assistive Hand Assessment for activities performance measurement.11
technology and assistive devices (such as wheelchairs, walk- Fifth is the speech/communication domain, measured using
ing frames, and communication devices) for children with GAS.209 The sixth domain is integration/participation which
CP, because devices form a large part of standard care. To can be measured using the COPM or GAS204 (note that
date, specialized equipment and technology has been vastly other domain-specific measures exist such as the LIFE-H,
under-researched, probably because the benefits are easily but this does not have adequate sensitivity to detect change).
observable (such as independent mobility) and the studies Finally, regarding both the seventh domain, quality of life,
are expensive to conduct; however, in light of device aban- and the eighth domain, caregiver instruments, valid and reli-
donment issues and associated costs, extensive efficacy able instruments exist in the literature, but less is understood
research is warranted at both an individual and a popula- about whether these measures are sensitive to change, and
tion level. Moreover, prescribing assistive technology with therefore recommendations for use are not made at this
a specialized appearance (such as orthotics, suits, comput- juncture.
erized devices, robotics) may well elevate expectations of In line with the principles of evidence-based care and as
good outcomes and give rise to an overinflated perception a cost-saving measure, it is highly recommended that cra-
of high-quality expert care. Thus, it is essential to know if niosacral therapy, hip bracing, hyperbaric oxygen, neurode-
the interventions are working, so as to prevent device velopmental therapy, and sensory integration should all be
abandonment, false hopes, and unnecessary effort. discontinued from CP care. Interestingly, these ineffective
When yellow-light interventions are used, it is imperative interventions for the most part are founded upon out-dated
that clinicians utilize a sufficiently sensitive outcome mea- neurological theories about CP. For example, hyperbaric
sure to confirm whether or not the intervention is working oxygen as a treatment for CP was based on the now dis-
and if it is helping the child achieve their family’s goals. The proven assumption that all CP arises from a lack of oxygen
Canadian Occupational Performance Measure (COPM) and during birth (true for only 5–10% of cases190) and that
Goal Attainment Scaling (GAS)5,64,204 have been widely increased oxygenation ought to help repair brain function.
adopted in the literature for assessing goal achievement Neurodevelopmental therapy sought to reduce hyper-ref-
because they are valid, reliable, sensitive to change, and clin- lexia by repositioning the limb on stretch, providing a local
ically affordable. Moreover, both measures work well within pattern-breaking effect mimicking spasticity reduction, but
the family-centred approach because they encourage family- we now know (1) that local effects do not translate to a
led goal setting and facilitate individualization, which is reduction in centrally driven spasticity long term210; and
important for such a heterogeneous condition as CP. For (2) that no substantive evidence exists to support the idea
yellow-light interventions, in addition to measuring whether that inhibition of primitive reflex patterns promotes motor
goals are achieved, it may be desirable to measure if the development.12 Likewise, ‘bottom-up’ approaches, in which
intervention is actually achieving what it purports to do for children’s underlying motor deficits are treated with the
each individual. Systematic individual outcome measure- aim of preparing them for function (such as neurodevelop-
ment, conducted at a population level with data aggregation, mental therapy and sensory integration) were commend-
would introduce the possibility of rapidly expanding the able pursuits when originally invented but disappointingly
evidence base amongst this heterogeneous population. have little carryover into functional activities.12
Parents, young people, and doctors have identified eight Over a decade ago, CP research experts12 and systematic
consensus measurement domains, important for assessing review authors called for ‘concerted efforts to investigate
the impact of a CP intervention, that span the ICF levels.205 other therapy approaches that may prove more clearly
We identified systematic reviews that provided measurement beneficial’.142 These therapy experts were referring to
Review 903
performance-based or ‘top-down’ approaches based on including small sample sizes (n<16) and extremely low
motor learning theory, in which interventions focus methodological quality such as a lack of blinding, inten-
directly on specific task training in activities of interest and tion-to-treat analysis, concealed allocation, etc. In the three
are not concerned with underlying impairments in body NDT dosing RCTs, two studies (studying n=96 children)
structures and function.201 This visionary advice, in con- found no difference between intense or regular NDT,
cert with the researchers who rigorously tested their theo- whereas one more recent study, by Tsorlakis212 (n=34),
ries, has transformed CP rehabilitation in recent years. showed favourable outcomes from higher-intensity NDT
The majority of the ‘do it’ or green-light effective CP over lower-intensity NDT. The most recent NDT system-
therapy evidence generated in the last 10 years are in fact atic review143 cited the Tsorlakis212 RCT as the sole high-
top-down therapy approaches, aimed at improving activi- level evidence for NDT being favourable, excluding older
ties performance and inducing neuroplasticity, and include evidence and thus all the unfavourable NDT RCTs. Since
bimanual training, constraint-induced movement therapy, this is not a standard systematic review methodology for
context-focused therapy, goal-directed/functional training, providing proof of efficacy, the results of this systematic
occupational therapy after toxin, and home programmes. review143 should be interpreted with caution. The differ-
Consistent with the theoretical underpinnings, research has ence in inclusion criteria between the systematic reviews
not focused on whether these top-down approaches had a explains why the newer systematic review143 suggests a
positive effect at the body structures and function level of more favourable benefit from NDT than the earlier sys-
the ICF (Table II). tematic reviews that concluded ineffectiveness.141,142
Given the sudden increase in new effective treatment In order to determine the strength of recommendation,
options available, it is essential that the field widely the panel weighed up the balance of benefits and harms from
embraces and implements these interventions in order to NDT and concluded that there was strong evidence that
ensure that children with CP achieve the best possible out- NDT does not improve contracture and tone, along with
comes. Adoption of evidence-based practice also involves weak evidence that NDT does not improve function. This
the difficult task of getting clinicians to stop providing was because, first, when the methodological quality of the
ineffective treatments that they ‘love’.211 It has been sug- evidence base was considered, the highest quality evidence
gested that the field requires professionals ‘who want to do suggested NDT was ineffective, with only low-quality, high
the best they can for their patients, who are willing to con- risk of bias studies finding a favourable benefit from NDT.
tinually question their own managements, and who have Second, the importance of the outcome that NDT aims to
readily available sources of information about what does prevent was considered: (1) regarding contracture, which is
work’.211 Our present systematic review seeks to provide painful and can limit function, high-quality RCTs showed
the CP field with a comprehensive overview about what that casting was a superior treatment to NDT for contrac-
works for children with CP and what does not (Fig. 4). ture management and therefore the panel favoured casting;
Based on best available evidence, the challenge now is for (2) regarding tone reduction, the highest quality evidence
the field to stop permissive endorsement of proven ineffec- suggested that NDT was ineffective for this indication and
tive interventions on the basis of perceived low risk and other evidence shows BoNT exists as a highly effective alter-
clinical expertise. This recommendation includes ceasing native and therefore the panel favoured BoNT or other
provision of the ever-popular NDT. This is because NDT effective pharmacological agents. Third, the magnitude and
has been a mainstay physiotherapy and occupational ther- precision of treatment effect was considered: only 3 out of
apy treatment for many years, but for the most part, the 15 trials found any benefit of NDT, and in these studies the
evidence base is unfavourable. Of note, contemporary treatment effects were small with very low precision esti-
NDT therapists eclectically include additional evidence- mates as a result of methodological flaws. Fourth, the bur-
based treatment approaches under the NDT banner (e.g. dens and costs of the therapy were considered: NDT is
motor learning and the philosophy of family-centred prac- time-consuming and expensive for families, and, what is
tice), and it is difficult to distil which treatment approaches more, a high-quality RCT shows that substantially better
are being used with fidelity and what features of the treat- functional motor gains are achieved from motor learning
ment are actually working. than from NDT at equal doses.213 Therefore, despite the
Nevertheless, three systematic reviews have been con- evidence being less well understood for the likelihood of
ducted of traditional NDT,141–143 including 18 discrete NDT influencing functional motor gains (yellow light), the
RCTs: 15 measuring efficacy and three measuring optimal panel favoured motor learning since superior gains were
dose. Of the 15 RCTs measuring NDT efficacy, 12 trials possible from an equal dose. Furthermore, since no other
(studying 674 children) found no statistically favourable body structure and function intervention in this review
benefits from NDT; these trials were of varying quality showed gains beyond the body structure and function level
(high, moderate, and low), whereas three trials (studying up into the activity level, it is hard to imagine why NDT
38 children) showed improvements in body structures and would be the exception to this trend.
functions such as gait parameters, spirometry, and mile- In summary, high-quality evidence demonstrates that
stone acquisition. The three favourable trials were all at casting is superior to NDT for managing contracture;
high risk of bias when assessed using the Cochrane criteria, BoNT exists as a highly effective alternative to NDT for

managing tone since NDT is ineffective for this indication; developed, researchers need to measure the effects of par-
and despite less being known about whether NDT ticipation intervention using GAS or the COPM.
improves function, high-quality evidence indicates that
motor leaning is superior to NDT for improving function. Study limitations
Consequently, there are no circumstances where any of the All systematic reviews are prone to publication bias from
aims of NDT could not be achieved by a more effective the included trial data; therefore, this systematic review of
treatment. Thus, on the grounds of wanting to do the best systematic reviews may incorporate this inherent bias.
for children with CP, it is hard to rationalize a continued There is also no guarantee that absolutely all relevant sys-
place for traditional NDT within clinical care. tematic reviews were retrieved, despite the thorough search
strategy. Publication bias, however, is unlikely to be more
Recommendations for research of a problem when identifying systematic reviews than
In future, systematic review authors should assign a GRADE when identifying clinical trials. Moreover, conducting a
to the body of evidence summarized, to enable clinicians to systematic review of systematic reviews is a study limitation
more quickly interpret the findings of the review for clinical in its own right because the method does not create any
practice. For the motor learning interventions that were information that was not already available. Furthermore,
‘green light’, researchers have repeatedly called for future using a high-level synthesis helicopter view means that spe-
investigations to determine optimal dosing, to better assess cific intervention details about how the intervention took
the widely held belief that ‘more is better’. Understanding place, who benefitted from the intervention, and for how
optimal intensity of therapy is important for maximizing long the intervention was carried out for were not
outcomes, accurately costing services, and offering family- reported; clinicians would need to turn to the included
friendly, achievable interventions. For all the green-light papers to obtain this information. In its place we hope that
interventions, additional studies that evaluate long-term out- the knowledge synthesis will help to bridge the gap
comes are necessary. First, because families of children with between research and practice by providing comparisons of
CP have life-long caregiving responsibilities, an understand- varying interventions to aid decision making.
ing the impact of these time-intensive and expensive inter-
ventions would help with expectation management and
planning for lifetime care. Second, it is unknown if some CONCLUSION
interventions continue to add an incremental benefit when In conclusion, we found compelling evidence from sys-
used repeatedly over years or whether the gains are one-off tematic reviews to suggest that the following interventions
and short term only. Long-term outcome data are essential are effective at the body structures and function level
for costing and optimizing the outcomes of children with alone: anticonvulsants, ankle casting, BoNT, bisphospho-
CP. nates, diazepam, fitness training, hip surveillance, pressure
For the yellow-light interventions with lower-quality evi- care, and selective dorsal rhizotomy. We also found com-
dence or a paucity of research to support effectiveness, rec- pelling evidence from systematic reviews to suggest that
ommendations for research include the use of individual the following interventions improve function at the activi-
patient meta-analyses to accelerate data aggregation; collab- ties level: bimanual training, constraint-induced movement
orations that strategize multicentre data collection to over- therapy, context-focused therapy, goal-directed/functional
come sample size barriers; and the use of CP registries and training, home programmes, and occupational therapy
single-system designs if RCTs are deemed impossible or after BoNT. No interventions were shown to work con-
ethically undesirable to conduct. Use of these research clusively at more than one level of the ICF. Therefore, if
methodologies is advisable and appropriate across all disci- a body structures and function outcome is desired, the
plines but would have particular value if applied to the disci- intervention must be selected from the suite of evidence-
plines of orthopaedic surgery, speech pathology,214–216 and based body structures and function interventions. Con-
social work, in order to better substantiate the important versely, if an activities-level outcome is sought, top-down
contributions these clinicians make to CP care. The CP field learning interventions, acting at the activities level, must
would also benefit from social workers and psychologists be applied.
confirming the assumed benefits of proven interventions The lack of certain efficacy evidence for large propor-
from non-CP populations amongst children with CP. tions of the interventions in use within standard care is a
When the whole evidence base was viewed from a global problem for people with CP, healthcare providers, purchas-
perspective, there was a startling lack of interventions ers of healthcare, and funders. More research using rigor-
available to improve children’s participation within their ous designs is urgently needed as CP is the most common
community. Given that this has been identified by many of physical disability of childhood with a life-long impact.190
the systematic review authors as a priority area for inter-
vention, more research designed to measure the effects SUPPORTING INFORMATION
of participation interventions and funds dedicated to this Additional Supporting Information may be found in the online
end is urgently needed. Furthermore, until participation- version of this article:
specific measures with sensitivity to change have been Table SI: Search strategy.
Review 905
REFERENCES
1. Flores-Mateo G, Argimon JM. Evidence based prac- 18. Novak I, McIntyre S. The effect of Education with and rehabilitation medicine: a review of the recent
tice in postgraduate healthcare education: a systematic workplace supports on practitioners’ evidence-based literature. Panminerva Med 2011; 53: 129–36.
review. BMC Health Serv Res 2007; 7: 119. practice knowledge and implementation behaviours. 33. Winkle M, Crowe TK, Hendrix I. Service dogs and
2. Rodger S, Brown GT, Brown A. Profile of paediatric Aust Occup Ther J 2010; 57: 386–93. people with physical disabilities partnerships: a
occupational therapy practice in Australia. Aust Occup 19. Haynes RB. What kind of evidence is it that Evi- systematic review. Occup Ther Int 2012; 19: 54–66.
Ther J 2005; 52: 311–25. dence-Based Medicine advocates want health care 34. Wilson D, Mitchell J, Kemp B, Adkins R, Mann W.
3. Saleh M, Korner-Bitensky N, Snider L, et al. Actual providers and consumers to pay attention to? BMC Effects of assistive technology on functional decline in
vs. best practices for young children with cerebral Health Serv Res 2002; 2: 3. people ageing with a disability. Assist Technol 2009;
palsy: a survey of paediatric occupational therapists 20. Cook DJ, Mulrow CD, Haynes RB. Systematic 21: 208–17.
and physical therapists in Quebec, Canada. Dev reviews: synthesis of best evidence for clinical deci- 35. Davies TC, Mudge S, Ameratunga S, Stott NS.
Neurorehabil 2008; 11: 60–80. sions. Ann Intern Med 1997; 126: 376–80. Enabling self-directed computer use for individuals
4. Sakzewski L, Ziviani J, Boyd R. Systematic review 21. Higgins JPT, Green S, Collaboration C. Cochrane with cerebral palsy: a systematic review of assistive
and meta-analysis of therapeutic management of Handbook for Systematic Reviews of Interventions. devices and technologies. Dev Med Child Neurol 2010;
upper-limb dysfunction in children with congenital Chichester: Wiley Online Library, 2008. 52: 510–6.
hemiplegia. Pediatrics 2009; 123: e1111–22. 22. Liberati A, Altman DG, Tetzlaff J, et al. The PRIS- 36. Jones MA, McEwen IR, Neas BR. Effects of power
5. Love SC, Novak I, Kentish M, et al. Botulinum toxin MA statement for reporting systematic reviews and wheelchairs on the development and function of
assessment, intervention and after-care for lower limb meta-analyses of studies that evaluate health care young children with severe motor impairments. Pedi-
spasticity in children with cerebral palsy: international interventions: explanation and elaboration. PLoS Med atr Phys Ther 2012; 24: 131–40 10.1097/PEP.
consensus statement. Eur J Neurol 2010; 17(Suppl. 2): 2009; 6: e1000100. 0b013e31824c5fdc.
9–37. 23. OCEBM Levels of Evidence Working Group. The 37. Livingstone R. A critical review of powered mobility
6. Straus S, Haynes RB. Managing evidence-based Oxford Levels of Evidence 2. Oxford Centre for Evi- assessment and training for children. Disabil Rehabil
knowledge: the need for reliable, relevant and read- dence Based Medicine. https://www.cebm.net/index. Assist Technol 2010; 5: 392–400.
able resources. CMAJ 2009; 180: 942–5. aspx?o=5653 (accessed 02 April 2012). 38. Chantry J, Dunford C. How do computer assistive
7. Guyatt GH, Meade MO, Jaeschke RZ, Cook DJ, 24. Campbell L, Novak I, McIntyre S. Patterns and rates technologies enhance participation in childhood occu-
Haynes RB. Practitioners of evidence based care. Not of use of an evidence-based practice intranet resources pations for children with multiple and complex dis-
all clinicians need to appraise evidence from scratch for allied health professionals: a randomized controlled abilities? A review of the current literature. Br J Occup
but all need some skills. BMJ 2000; 320: 954–5. trial. Dev Med Child Neurol 2010; 52(S2): 31. Ther 2010; 73: 351–65.
8. Grol R, Grimshaw J. From best evidence to best 25. Zhang Y, Liu J, Wang J, He Q. Traditional Chinese 39. Sandlund M, McDonough S, H€ager-Ross C. Interac-
practice: effective implementation of change in medicine for treatment of cerebral palsy in children: a tive computer play in rehabilitation of children with
patients’ care. Lancet 2003; 362: 1225–30. systematic review of randomized clinical trials. J sensorimotor disorders: a systematic review. Dev Med
9. World Health Organization (WHO). International Altern Complement Med 2010; 16: 375–95. Child Neurol 2009; 51: 173–9.
Classification of Functioning, Disability and Health. 26. Delgado MR, Hirtz D, Aisen M, et al. Practice 40. Laufer Y, Weiss PL. Virtual reality in the assessment
Geneva: WHO, 2001. parameter: pharmacologic treatment of spasticity in and treatment of children with motor impairment: a
10. Adams Vargus J. Understanding function and other children and adolescents with cerebral palsy (an evi- systematic review. J Phys Ther Educ 2011; 25: 59–71.
outcomes in cerebral palsy. Phys Med Rehabil Clin N dence-based review): report of the Quality Standards 41. Parsons TD, Rizzo AA, Rogers S, York P. Virtual
Am 2009; 20: 567–75. Subcommittee of the American Academy of Neurol- reality in paediatric rehabilitation: a review. Dev Neu-
11. Gilmore R, Sakzewski L, Boyd R. Upper limb activity ogy and the Practice Committee of the Child Neurol- rorehabil 2009; 12: 224–38.
measures for 5- to 16-year-old children with congeni- ogy Society. Neurology 2010; 74: 336–43. 42. Snider L, Majnemer A, Darsaklis V. Virtual reality as
tal hemiplegia: a systematic review. Dev Med Child 27. Pennington L, Goldbart J, Marshall J. Speech and a therapeutic modality for children with cerebral
Neurol 2010; 52: 14–21. language therapy to improve the communication skills palsy. Dev Neurorehabil 2010; 13: 120–8.
12. Law M, Darrah J, Pollock N, et al. Family-centred of children with cerebral palsy. Cochrane Database Syst 43. Wang M, Reid D. Virtual reality in paediatric neu-
functional therapy for children with cerebral palsy. Rev 2004a; 2: CD003466. rorehabilitation: attention deficit hyperactivity disor-
Phys Occup Ther Pediatr 1998; 18: 83–102. 28. Branson D, Demchak M. The use of augmentative der, autism and cerebral palsy. Neuroepidemiology
13. Novak I, Cusick A, Lannin N. Occupational therapy and alternative communication methods with infants 2011; 36: 2–18.
home programmes for cerebral palsy: double-blind, ran- and toddlers with disabilities: a research review. Aug- 44. Jung Y, Bridge C. Evidence Based Research: The
domized, controlled trial. Pediatrics 2009; 124: e606–14. ment Altern Commun 2009; 25: 274–86. Effectiveness of Ceiling Hoists in Transferring People
14. Wallen M, Ziviani J, Naylor O, et al. Modified con- 29. Pennington L, Goldbart J, Marshall J. Interaction with Disabilities. Sydney: Home Modification Infor-
straint-induced therapy for children with hemiplegic training for conversational partners of children with mation Clearinghouse, University of New South
cerebral palsy: a randomized trial. Dev Med Child cerebral palsy: a systematic review. Int J Lang Com- Wales, 2009. Available from: www.homemods.info
Neurol 2011; 53: 1091–9. mun Disord 2004; 39: 151–70. 45. Pin T, Dyke P, Chan M. The effectiveness of passive
15. Haines A, Kuruvilla S, Borchert M. Bridging the 30. Hanson E, Yorkston K, Beukelman D. Speech supple- stretching in children with cerebral palsy. Dev Med
implementation gap between knowledge and action mentation techniques for dysarthria: a systematic Child Neurol 2006; 48: 855–62.
for health. Bull World Health Organ 2004; 82: 724–31. review. J Med Speech Lang Pathol 2004; 12: IX–XXIX. 46. Wynn N, Wickham J. Night-time positioning for
16. Mitton C, Adair CE, McKenzie E, Patten SB, Perry 31. Millar DC, Light JC, Schlosser RW. The impact of children with postural needs: what is the evidence to
BW. Knowledge transfer and exchange: review and augmentative and alternative communication interven- inform best practice? Br J Occup Ther 2009; 72:
synthesis of the literature. Milbank Q 2007; 85: tion on the speech production of individuals with 543–50.
729–68. developmental disabilities: a research review. J Speech 47. Nicolson A, Moir L, Millsteed J. Impact of assistive
17. GRADE Working Group. Grading quality of evi- Lang Hear Res 2006; 49: 248–64. technology on family caregivers of children with
dence and strength of recommendations. BMJ 2004; 32. Mu~
noz LS, Ferriero G, Brigatti E, Valero R, Fran- physical disabilities: a systematic review. Disabil Reha-
328: 1–8. chignoni F. Animal-assisted interventions in internal bil Assist Technol 2012; 7: 345–9.

48. Roberts C, Mazzucchelli T, Studman L, Sanders MR. of spasticity in ambulant children with cerebral palsy- 78. Katalinic OM, Harvey LA, Herbert RD, et al. Stretch
Behavioural family intervention for children with a structured review. NZ J Physiother 2001; 29: 18–31. for the treatment and prevention of contractures.
developmental disabilities and behavioural problems. 64. Fehlings D, Novak I, Berweck S, et al. Botulinum toxin Cochrane Database Syst Rev 2010; 9: CD007455.
J Clin Child Adolesc Psychol 2006; 35: 180–93. assessment, intervention and follow-up for paediatric 79. Lannin NA, Novak I, Cusick A. A systematic review
49. Sanders MR, Mazzucchelli TG, Studman LJ. Step- upper limb hypertonicity: international consensus state- of upper extremity casting for children and adults
ping Stones Triple P: the theoretical basis and devel- ment. Eur J Neurol 2010; 17(Suppl. 2): 38–56. with central nervous system motor disorders. Clin
opment of an evidence-based positive parenting 65. Reeuwijk A, van Schie PEM, Becher JG, Kwakkel G. Rehabil 2007; 21: 963–76.
programme for families with a child who has a Effects of botulinum toxin type A on upper limb 80. Teplicky R, Law M, Russell D. The effectiveness of
disability. J Intellect Dev Disabil 2004; 29: 265–83. function in children with cerebral palsy: a systematic casts, orthoses, and splints for children with neurolog-
50. Whittingham K, Wee D, Boyd R. Systematic review review. Clin Rehabil 2006; 20: 375–87. ical disorders. Infants Young Child 2002; 15: 42–50.
of the efficacy of parenting interventions for children 66. Wasiak J, Hoare B, Wallen M. Botulinum toxin A as 81. Graham F, Rodger S, Ziviani J. Enabling occupa-
with cerebral palsy. Child Care Health Dev 2011; 37: an adjunct to treatment in the management of the tional performance of children through coaching par-
475–83. upper limb in children with spastic cerebral palsy. ents: three case reports. Phys Occup Ther Pediatr 2010;
51. Gordon AM, Hung YC, Brandao M, et al. Bimanual Cochrane Database Syst Rev 2004; 3: CD003469. 30: 4–15.
training and constraint-induced movement therapy in 67. Novak I, Campbell L, Boyce M, Fung V. Botulinum 82. Pennington L, Miller N, Robson S. Speech therapy
children with hemiplegic cerebral palsy: a randomized toxin assessment, intervention and aftercare for cervi- for children with dysarthria acquired before three
trial. Neurorehabil Neural Repair 2011; 25: 692–702. cal dystonia and other causes of hypertonia of the years of age. Cochrane Database Syst Rev 2009; 4:
52. Sakzewski L, Ziviani J, Abbott DF, et al. Randomized neck: international consensus statement. Eur J Neurol CD006937.
trial of constraint-induced movement therapy and 2010; 17: 94–108. 83. Darrah J, Watkins B, Chen L, Bonin C. Conductive
bimanual training on activity outcomes for children 68. Ryll U, Bastiaenen C, De Bie R, Staal B. Effects of leg education intervention for children with cerebral
with congenital hemiplegia. Dev Med Child Neurol muscle botulinum toxin A injections on walking in chil- palsy: an AACPDM evidence report. Dev Med Child
2011; 53: 313–20. dren with spasticity-related cerebral palsy: a systematic Neurol 2004; 46: 187–203.
53. Dursun E, Dursun N, Alican D. Effects of biofeed- review. Dev Med Child Neurol 2011; 53: 210–6. 84. Tuersley-Dixon L, Frederickson N. Conductive edu-
back treatment on gait in children with cerebral palsy. 69. Hoare BJ, Imms C. Upper-limb injections of botu- cation: appraising the evidence. Educ Psychol Pract
Disabil Rehabil 2004; 26: 116–20. linum toxin-A in children with cerebral palsy: a criti- 2010; 26: 353–73.
54. Bloom R, Przekop A, Sanger TD. Prolonged electro- cal review of the literature and clinical implications 85. Hoare BJ, Wasiak J, Imms C, Carey L. Constraint-
myogram biofeedback improves upper extremity func- for occupational therapists. Am J Occup Ther 2004; induced movement therapy in the treatment of the
tion in children with cerebral palsy. J Child Neurol 58: 389–97. upper limb in children with hemiplegic cerebral palsy.
2010; 25: 1480–4. 70. Hoare BJ, Wallen MA, Imms C, et al. Botulinum Cochrane Database Syst Rev 2007; 2: CD004149.
55. Fehlings D, Switzer L, Agarwal P, et al. Informing toxin A as an adjunct to treatment in the management 86. Huang H, Fetters L, Hale J, McBride A. Bound for
evidence-based clinical practice guidelines for children of the upper limb in children with spastic cerebral success: a systematic review of constraint-induced
with cerebral palsy at risk of osteoporosis: a system- palsy (UPDATE). Cochrane Database Syst Rev 2010; 1: movement therapy in children with cerebral palsy
atic review. Dev Med Child Neurol 2012; 54: 106–16. CD003469. supports improved arm and hand use. Phys Ther 2009;
56. Hough JP, Boyd RN, Keating JL. Systematic review 71. Rawicki B, Sheean G, Fung V, et al. Botulinum toxin 89: 1126–41.
of interventions for low bone mineral density in assessment, intervention and aftercare for paediatric 87. Nascimento L, Gl
oria A, Habib E. Effects of con-
children with cerebral palsy. Pediatrics 2010; 125: and adult niche indications including pain: interna- straint-induced movement therapy as a rehabilitation
e670–8. tional consensus statement. Eur J Neurol 2010; 17: strategy for the affected upper limb of children with
57. Ade-Hall RA, Moore AP. Botulinum toxin type A in 122–34. hemiparesis: systematic review of the literature. Rev
the treatment of lower limb spasticity in cerebral 72. Lim M, Mace A, Reza Nouraei S, Sandhu G. Botu- Bras Fisioter 2009; 13: 97–102.
palsy. Cochrane Database Syst Rev 2000; 2: CD001408. linum toxin in the management of sialorrhoea: a sys- 88. Law M, Darrah J, Pollock N, et al. Focus on func-
58. Albavera-Hernandez C, Rodriguez JM, Idrovo AJ. tematic review. Clin Otolaryngol 2006; 31: 267–72. tion: a cluster, randomized controlled trial comparing
Safety of botulinum toxin type A among children with 73. Reddihough D, Erasmus C, Johnson H, McKellar G, child-versus context-focused intervention for young
spasticity secondary to cerebral palsy: a systematic Jongerius P. Botulinum toxin assessment, intervention children with cerebral palsy. Dev Med Child Neurol
review of randomized clinical trials. Clin Rehabil 2009; and aftercare for paediatric and adult drooling: inter- 2011; 53: 621–9.
23: 394–407. national consensus statement. Eur J Neurol 2010; 17: 89. Palit A, Chatterjee AK. Parent-to-parent counseling-a
59. Boyd RN, Hays RM. Current evidence for the use of 109–21. gateway for developing positive mental health for the
botulinum toxin type A in the management of chil- 74. Walshe M, Smith M, Pennington L. Interventions for parents of children that have cerebral palsy with mul-
dren with cerebral palsy: a systematic review. Eur J drooling in children with cerebral palsy. Cochrane tiple disabilities. Int J Rehabil Res 2006; 29: 281–8.
Neurol 2001; 8(Suppl. 5): 1–20. Database Syst Rev 2012; 2: CD008624. 90. Wyatt K, Edwards V, Franck L, et al. Cranial osteop-
60. Heinen F, Desloovere K, Schroeder AS, et al. The 75. Autti-Ramo I, Suoranta J, Anttila H, Malmivaara A, athy for children with cerebral palsy: a randomized
updated European Consensus 2009 on the use of Bot- Makela M. Effectiveness of upper and lower limb controlled trial. Arch Dis Child 2011; 96: 505–12.
ulinum toxin for children with cerebral palsy. Eur J casting and orthoses in children with cerebral palsy: 91. Snider L, Majnemer A, Darsaklis V. Feeding inter-
Paediatr Neurol 2010; 14: 45–66. an overview of review articles. Am J Phys Med Rehabil ventions for children with cerebral palsy: a review of
61. Koog YH, Min BII. Effects of botulinum toxin A on 2006; 85: 89–103. the evidence. Phys Occup Ther Pediatr 2011; 31: 58–77.
calf muscles in children with cerebral palsy: a system- 76. Blackmore AM, Boettcher-Hunt E, Jordan M, Chan 92. Blauw-Hospers CH, Hadders-Algra M. A systematic
atic review. Clin Rehabil 2010; 24: 685–700. MD. A systematic review of the effects of casting on review of the effects of early intervention on motor
62. Lukban MB, Rosales RL, Dressler D. Effectiveness of equinus in children with cerebral palsy: an evidence development. Dev Med Child Neurol 2005; 47: 421–32.
botulinum toxin A for upper and lower limb spasticity report of the AACPDM. Dev Med Child Neurol 2007; 93. Blauw-Hospers CH, de Graaf-Peters VB, Dirks T,
in children with cerebral palsy: a summary of evi- 49: 781–90. Bos AF, Hadders-Algra M. Does early intervention in
dence. J Neural Transm 2009; 116: 319–31. 77. Effgen S, McEwen I. Review of selected physical infants at high risk for a developmental motor disor-
63. Mulligan H, Borkin H, Chaplin K, Croft N, Scherp therapy interventions for school age children with dis- der improve motor and cognitive development? Neu-
A. The efficacy of botulinum toxin A in the treatment abilities. Phys Ther Rev 2008; 13: 297–312. rosci Biobehav Rev 2007; 31: 1201–12.
Review 907
94. Turnbull JD. Early intervention for children with or 110. Sullivan P, Morrice J, Vernon-Roberts A, et al. Does 126. Getz M, Hutzler Y, Vermeer A. Effects of aquatic
at risk of cerebral palsy. Am J Dis Child 1993; 147: gastrostomy tube feeding in children with cerebral interventions in children with neuromotor impair-
54–9. palsy increase the risk of respiratory morbidity? Arch ments: a systematic review of the literature. Clin Reha-
95. Ziviani J, Feeney R, Rodger S, Watter P. Systematic Dis Child 2006; 91: 478–82. bil 2006; 20: 927–36.
review of early intervention programmemes for chil- 111. Vernon-Roberts A, Wells J, Grant H, et al. Gastros- 127. Gorter JW, Currie SJ. Aquatic exercise programmes
dren from birth to nine years who have a physical dis- tomy feeding in cerebral palsy: enough and no more. for children and adolescents with cerebral palsy: what
ability. Aust Occup Ther J 2010; 57: 210–23. Dev Med Child Neurol 2010; 52: 1099–105. do we know and where do we go? Int J Pediatr 2011;
96. Cauraugh JH, Naik SK, Hsu WH, Coombes SA, 112. Ketelaar M, Vermeer A, Hart H, van Petegem-van 2011: 712165.
Holt KG. Children with cerebral palsy: a systematic Beek E, Helders PJM. Effects of a functional therapy 128. Collet JP, Vanasse M, Marois P, et al. Hyperbaric
review and meta-analysis on gait and electrical stimu- programme on motor abilities of children with cere- oxygen for children with cerebral palsy: a randomized
lation. Clin Rehabil 2010; 24: 963–78. bral palsy. Phys Ther 2001; 81: 1534–45. multicentre trial. Lancet 2001; 357: 582–6.
97. Kerr C, McDowell B, McDonough S. Electrical stim- 113. L€
owing K, Bexelius A, Brogren Carlberg E. Activity 129. McDonagh MS, Morgan D, Carson S, Russman BS.
ulation in cerebral palsy: a review of effects on focused and goal directed therapy for children with Systematic review of hyperbaric oxygen therapy for
strength and motor function. Dev Med Child Neurol cerebral palsy-Do goals make a difference? Disabil cerebral palsy: the state of the evidence. Dev Med
2004; 46: 205–13. Rehabil 2009; 31: 1808–16. Child Neurol 2007; 49: 942–7.
98. Lannin N, Scheinberg A, Clark K. AACPDM system- 114. Smeulders M, Coester A, Kreulen M. Surgical treat- 130. Butler C, Campbell S. Evidence of the effects of
atic review of the effectiveness of therapy for children ment for the thumb-in-palm deformity in patients intrathecal baclofen for spastic and dystonic cerebral
with cerebral palsy after botulinum toxin A injections. with cerebral palsy. Cochrane Database Syst Rev 2005; palsy. AACPDM Treatment Outcomes Committee
Dev Med Child Neurol 2006; 48: 533–9. 4: CD004093. Review Panel. Dev Med Child Neurol 2000; 42:
99. Rogers A, Furler BL, Brinks S, Darrah J. A systematic 115. Stott NS, Piedrahita L. Effects of surgical adductor 634–45.
review of the effectiveness of aerobic exercise inter- releases for hip subluxation in cerebral palsy: an 131. Creedon SD, Dijkers MPJM, Hinderer SR. Intrathe-
ventions for children with cerebral palsy: an AACPDM evidence report. Dev Med Child Neurol cal baclofen for severe spasticity: a meta-analysis. Int J
AACPDM evidence report. Dev Med Child Neurol 2004; 46: 628–45. Rehabil Health 1997; 3: 171–85.
2008; 50: 808–14. 116. Brunner R, Baumann JU. Long-term effects of inter- 132. Dan B, Motta F, Vles JS, et al. Consensus on the
100. Wright PA, Durham S, Ewins DJ, Swain ID. Neuro- trochanteric varus-derotation osteotomy on femur and appropriate use of intrathecal baclofen (ITB) therapy
muscular electrical stimulation for children with acetabulum in spastic cerebral palsy: an 11- to 18-year in paediatric spasticity. Eur J Paediatr Neurol 2010;
cerebral palsy: a review. Arch Dis Child 2012; 97: follow-up study. J Pediatr Orthop 1997; 17: 585–91. 14: 19–28.
364–71. 117. Huh K, Rethlefsen SA, Wren TAL, Kay RM. Surgical 133. Kolaski K, Logan LR. Intrathecal baclofen in cerebral
101. Butler JM, Scianni A, Ada L. Effect of cardiorespira- management of hip subluxation and dislocation in palsy: a decade of treatment outcomes. J Pediatr Reha-
tory training on aerobic fitness and carryover to activ- children with cerebral palsy: isolated VDRO or com- bil Med 2008; 1: 3–32.
ity in children with cerebral palsy: a systematic bined surgery? J Pediatr Orthop 2011; 31: 858–63. 134. Albanese A, Barnes MP, Bhatia KP, et al. A systematic
review. Int J Rehabil Res 2010; 33: 97–103. 118. Gordon GS, Simkiss DE. A systematic review of the review on the diagnosis and treatment of primary (idi-
102. Verschuren O, Ketelaar M, Takken T, Helders P, evidence for hip surveillance in children with cerebral opathic) dystonia and dystonia plus syndromes: report
Gorter J. Exercise programmes for children with cere- palsy. J Bone Joint Surg Br 2006; 88: 1492–6. of an EFNS/MDS-ES Task Force. Eur J Neurol 2006;
bral palsy: a systematic review of the literature. Am J 119. Snider L, Korner-Bitensky N, Kammann C, Warner 13: 433–44.
Phys Med Rehabil 2008; 87: 404–17. S, Saleh M. Horseback riding as therapy for children 135. Hoving MA, van Raak EPM, Spincemaille GHJJ,
103. Vernon-Roberts A, Sullivan PB. Fundoplication versus with cerebral palsy: is there evidence of its effective- et al. Efficacy of intrathecal baclofen therapy in chil-
post-operative medication for gastro-oesophageal ness? Phys Occup Ther Pediatr 2007; 27: 5–23. dren with intractable spastic cerebral palsy: a random-
reflux in children with neurological impairment 120. Sterba J. Does horseback riding therapy or therapist- ized controlled trial. Eur J Paediatr Neurol 2009; 13:
undergoing gastrostomy. Cochrane Database Syst Rev directed hippotherapy rehabilitate children with cere- 240–6.
2007; 1: CD006151. bral palsy? Dev Med Child Neurol 2007; 49: 68–73. 136. Hoving MA, van Raak EPM, Spincemaille GHJJ,
104. Arrowsmith F, Allen J, Gaskin K, et al. The effect of 121. Zadnikar M, Kastrin A. Effects of hippotherapy and et al. Safety and one-year efficacy of intrathecal baclo-
gastrostomy tube feeding on body protein and bone therapeutic horseback riding on postural control or fen therapy in children with intractable spastic cere-
mineralization in children with quadriplegic cerebral balance in children with cerebral palsy: a meta-analy- bral palsy. Eur J Paediatr Neurol 2009; 13: 247–56.
palsy. Dev Med Child Neurol 2010; 52: 1043–7. sis. Dev Med Child Neurol 2011; 53: 684–91. 137. Pin T, McCartney L, Lewis J, Waugh M. Use of
105. Kong CK, Wong HSS. Weight-for-height values and 122. Whalen CN, Case-Smith J. Therapeutic effects of intrathecal baclofen therapy in ambulant children and
limb anthropometric composition of tube-fed children horseback riding therapy on gross motor function in adolescents with spasticity and dystonia of cerebral
with quadriplegic cerebral palsy. Pediatrics 2005; 116: children with cerebral palsy: a systematic review. Phys origin: a systematic review. Dev Med Child Neurol
e839–45. Occup Ther Pediatr 2011; 32: 229–42. 2011; 53: 885–95.
106. Samson-Fang L, Butler C, O’Donnell M. Effects of 123. Davis E, Davies B, Wolfe R, et al. A randomized con- 138. Hernandez-Reif M, Field T, Largie S, et al. Cerebral
gastrostomy feeding in children with cerebral palsy: trolled trial of the impact of therapeutic horse riding palsy symptoms in children decreased following mas-
an AACPDM evidence report. Dev Med Child Neurol on the quality of life, health, and function of children sage therapy. Early Child Dev Care 2005; 175: 445–56.
2003; 45: 415–26. with cerebral palsy. Dev Med Child Neurol 2009; 51: 139. Nilsson S, Johansson G, Ensk€ar K, Himmelmann K.
107. Sleigh G, Brocklehurst P. Gastrostomy feeding in 111–9. Massage therapy in post-operative rehabilitation of
cerebral palsy: a systematic review. Arch Dis Child 124. Novak I, Cusick A. Home programmemes in paediat- children and adolescents with cerebral palsy– a pilot
2004; 89: 534–9. ric occupational therapy for children with cerebral study. Complement Ther Clin Pract 2011; 17: 127–31.
108. Sleigh G, Sullivan PB, Thomas AG. Gastrostomy feed- palsy: where to start? Aust Occup Ther J 2006; 53: 140. Alizad V, Sajedi F, Vameghi R. Muscle tonicity of
ing versus oral feeding alone for children with cerebral 251–64. children with spastic cerebral palsy: how effective is
palsy. Cochrane Database Syst Rev 2004; 2: CD003943. 125. Chrysagis N, Douka A, Nikopoulos M, Apostolopou- Swedish massage? Iran J Child Neurol 2009; 3: 25–9.
109. Sullivan P, Alder N, Bachlet A, et al. Gastrostomy lou F, Koutsouki D. Effects of an aquatic programme 141. Brown GT, Burns SA. The efficacy of neurodevelop-
feeding in cerebral palsy: too much of a good thing? on gross motor function of children with spastic cere- mental treatment in paediatrics: a systematic review.
Dev Med Child Neurol 2006; 48: 877–82. bral palsy. Biol Exerc 2009; 5: 13–25. Br J Occup Ther 2001; 64: 235–44.

142. Butler C, Darrah J. Effects of neurodevelopmental 158. Michael S, Porter D, Pountney T. Tilted seat position view of systematic reviews. Pediatr Phys Ther 2010; 22:
treatment (NDT) for cerebral palsy: an AACPDM evi- for non-ambulant individuals with neurological and 361–77.
dence report. Dev Med Child Neurol 2001; 43: 778–90. neuromuscular impairment: a systematic review. Clin 175. Damiano DL, DeJong SL. A systematic review of the
143. Martin L, Baker R, Harvey A. A systematic review of Rehabil 2007; 21: 1063–74. effectiveness of treadmill training and body weight
common physiotherapy interventions in school-aged 159. Grunt S, Becher JG, Vermeulen RJ. Long-term out- support in paediatric rehabilitation. J Neurol Phys Ther
children with cerebral palsy. Phys Occup Ther Pediatr come and adverse effects of selective dorsal rhizotomy 2009; 33: 27–44.
2010; 30: 294–312. in children with cerebral palsy: a systematic review. 176. Mutlu A, Krosschell K, Gaebler Spira D. Treadmill
144. Wilcox DD, Potvin MC, Prelock PA. Oral motor Dev Med Child Neurol 2011; 53: 490–8. training with partial body-weight support in children
interventions and cerebral palsy: using evidence to 160. McLaughlin J, Bjornson K, Temkin N, et al. Selective with cerebral palsy: a systematic review. Dev Med
inform practice. Early Interv School Special Interest Sect dorsal rhizotomy: meta-analysis of three randomized Child Neurol 2009; 51: 268–75.
Q 2009; 16: 1–4. controlled trials. Dev Med Child Neurol 2002; 44: 177. Willoughby K, Dodd K, Shields N. A systematic
145. Shore BJ, White N, Kerr Graham H. Surgical correc- 17–25. review of the effectiveness of treadmill training for
tion of equinus deformity in children with cerebral 161. Steinbok P. Outcomes after selective dorsal rhizotomy children with cerebral palsy. Disabil Rehabil 2009; 31:
palsy: a systematic review. J Child Orthop 2010; 4: for spastic cerebral palsy. Childs Nerv Syst 2001; 17: 1–18. 1971–9.
277–90. 162. Vargas S, Camilli G. A meta-analysis of research on 178. Brandt S, Lonstrup HV, Marner T, et al. Prevention
146. Figueiredo EM, Ferreira GB, Maia Moreira RC, sensory integration treatment. Am J Occup Ther 1999; of cerebral palsy in motor risk infants by treatment ad
Kirkwood RN, Fetters L. Efficacy of ankle–foot 53: 189–98. modum Vojta. A controlled study. Acta Paediatr Scand
orthoses on gait of children with cerebral palsy: sys- 163. McGinley J, Dobson F, Ganeshalingam R, et al. Sin- 1980; 69: 283–6.
tematic review of literature. Pediatr Phys Ther 2008; gle-event multilevel surgery for children with cerebral 179. d’Avignon M, Noren L, Arman T. Early physiother-
20: 207–23. palsy: a systematic review. Dev Med Child Neurol 2012; apy ad modum Vojta or Bobath in infants with sus-
147. Harris SR, Roxborough L. Efficacy and effectiveness 54: 117–28. pected neuromotor disturbance. Neuropediatrics 1981;
of physical therapy in enhancing postural control in 164. Test DW, Richter S, Knight V, Spooner F. A com- 12: 232–41.
children with cerebral palsy. Neural Plast 2005; 12: prehensive review and meta-analysis of the social sto- 180. Kanda T, Pidcock FS, Hayakawa K, Yamori Y, Shik-
229–43. ries literature. Focus Autism Other Dev Disabl 2011; 26: ata Y. Motor outcome differences between two groups
148. Morris C. A review of the efficacy of lower limb 49–62. of children with spastic diplegia who received differ-
orthoses used for cerebral palsy. Dev Med Child Neurol 165. Dodd KJ, Taylor NF, Damiano DL. A systematic ent intensities of early onset physiotherapy followed
2002; 44: 205–11. review of the effectiveness of strength-training pro- for 5 years. Brain Dev 2004; 26: 118–26.
149. Graham HK, Boyd R, Carlin JB, et al. Does botu- grammes for people with cerebral palsy. Arch Phys 181. Liu ZH, Pan PG, Ma MM. Effects of acupuncture on
linum toxin a combined with bracing prevent hip dis- Med Rehabil 2002; 83: 1157–64. quality of life in children with spastic cerebral palsy.
placement in children with cerebral palsy and ‘hips at 166. Jeglinsky I, Surakka J, Carlberg EB, Autti-R€am€
o I. Zhongguo Zhong Xi Yi Jie He Za Zhi 2007; 27: 214–6.
risk’? A randomized, controlled trial. J Bone Joint Surg Evidence on physiotherapeutic interventions for adults 182. Wu C, Peng X, Li X, et al. Vojta and Bobath com-
Am 2008; 90: 23–33. with cerebral palsy is sparse. A systematic review. Clin bined treatment for high risk infants with brain dam-
150. Redditi Hanzlik JS. The effect of intervention on the Rehabil 2010; 24: 771–88. age at early period. Neural Regen Res 2007; 2: 121–5.
free-play experience for mothers and their infants with 167. Mockford M, Caulton JM. Systematic review of pro- 183. Zhang QH, Zheng D, Liu SQ, et al. Therapeutic
developmental delay and cerebral palsy. Phys Occup gressive strength training in children and adolescents effect of Peto method on the recovery of the motor
Ther Pediatr 1989; 9: 33–51. with cerebral palsy who are ambulatory. Pediatr Phys function in children with cerebral palsy. Zhongguo
151. McInnes E, Bell-Syer SE, Dumville JC, Legood R, Ther 2008; 20: 318–33. Linchuang Kangfu 2004; 8: 2902–3.
Cullum NA. Support surfaces for pressure ulcer pre- 168. Scianni A, Butler JM, Ada L, Teixeira-Salmela LF. 184. Zhao Y, Dong JP, Wang XJ, Wang JT, Liu ZM.
vention. Cochrane Database Syst Rev 2008; 4: Muscle strengthening is not effective in children and Application of sensory integration in central coordina-
CD001735. adolescents with cerebral palsy: a systematic review. tion disorder. Zhongguo Linchuang Kangfu 2005; 9:
152. Strunk JA. Respite care for families of special needs Aust J Physiother 2009; 55: 81–7. 110–1.
children: a systematic review. J Dev Phys Disabil 2010; 169. Taylor N, Dodd K, Damiano D. Progressive resis- 185. del Pozo-Cruz B, Adsuar JC, Parraca JA, et al. Using
22: 615–30. tance exercise in physical therapy: a summary of sys- whole-body vibration training in patients affected with
153. Farley R, Clark J, Davidson C. What is the evidence tematic reviews. Phys Ther 2005; 85: 1208–23. common neurological diseases: a systematic literature
for the effectiveness of postural management? Int J 170. Kim D-A, Lee J-A, Hwang P-W, et al. The effect of review. J Altern Complement Med 2012; 18: 29–41.
Ther Rehabil 2003; 10: 449–55. comprehensive hand repetitive intensive strength 186. Crowther CA, Hiller JE, Doyle LW. Magnesium sul-
154. Ryan SE. An overview of systematic reviews of adap- training (CHRIST) using motion analysis in children phate for preventing preterm birth in threatened pre-
tive seating interventions for children with cerebral with cerebral palsy. Ann Rehabil Med 2012; 36: 39–46. term labour. Cochrane Database Syst Rev 2002; 4:
palsy: where do we go from here? Disabil Rehabil Assist 171. Wiart L, Darrah J, Kembhavi G. Stretching with chil- CD001060.
Technol 2012; 7: 104–11. dren with cerebral palsy: what do we know and where 187. Shah PS. Hypothermia: a systematic review and meta-
155. Chung J, Evans J, Lee C, et al. Effectiveness of adap- are we going? Pediatr Phys Ther 2008; 20: 173–8. analysis of clinical trials. Semin Fetal Neonatal Med
tive seating on sitting posture and postural control in 172. Alagesan J, Shetty A. Effect of modified suit therapy 2010; 15: 238–46.
children with cerebral palsy. Pediatr Phys Ther 2008; in spastic diplegic cerebral palsy-a single blinded ran- 188. Palisano R, Rosenbaum P, Walter S, et al. Develop-
20: 303–17. domized controlled trial. Online J Health Allied Sci ment and reliability of a system to classify gross
156. Roxborough L. Review of the efficacy and effective- 2011; 9: 14. motor function in children with cerebral palsy. Dev
ness of adaptive seating for children with cerebral 173. Bailes AF, Greve K, Burch CK, et al. The effect of Med Child Neurol 1997; 39: 214–23.
palsy. Assist Technol 1995; 7: 17–25. suit wear during an intensive therapy programme in 189. Eliasson AC, Krumlinde-Sundholm L, Rosblad B,
157. McNamara L, Casey J. Seat inclinations affect the children with cerebral palsy. Pediatr Phys Ther 2011; et al. The Manual Ability Classification System
function of children with cerebral palsy: a review of 23: 136–42. (MACS) for children with cerebral palsy: scale devel-
the effect of different seat inclines. Disabil Rehabil 174. Zwicker JG, Mayson TA. Effectiveness of treadmill opment and evidence of validity and reliability.
Assist Technol 2007; 2: 309–18. training in children with motor impairments: an over- Dev Med Child Neurol 2006; 48: 549–54.
Review 909
190. ACPR Group. Report of the Australian Cerebral Palsy 199. Gingerich WJ, Eisengart S. Solution-focused brief tion for children with cerebral palsy. Dev Med Child
Register, Birth Years 1993–2003, 2009. therapy: a review of the outcome research. Fam Process Neurol 2008; 50: 190–8.
191. NICE. The Epilepsies: Clinical Practice Guideline. 2000; 39: 477–98. 208. Capio CM, Sit CH, Abernethy B, Rotor ER. Physical
The diagnosis and management of the epilepsies in 200. McCauley RJ, Strand E, Lof GL, Schooling T, Fry- activity measurement instruments for children with
adults and children in primary and secondary care. mark T. Evidence-based systematic review: effects of cerebral palsy: a systematic review. Dev Med Child
Clinical Guidelines 137. 2012; Available from: www. nonspeech oral motor exercises on speech. Am J Neurol 2010; 52: 908–16.
nice.org.uk/nicemedia/live/13635/57779/57779.pdf. Speech Lang Pathol 2009; 18: 343–60. 209. Schlosser RW. Goal attainment scaling as a clinical
192. David-Ferdon C, Kaslow NJ. Evidence-based psycho- 201. Polatajko HJ, Cantin N. Exploring the effectiveness of measurement technique in communication disorders:
social treatments for child and adolescent depression. occupational therapy interventions, other than the sen- a critical review. J Commun Disord 2004; 37: 217–39.
J Clin Child Adolesc Psychol 2008; 37: 62–104. sory integration approach, with children and adolescents 210. Lannin NA, Ada L. Neurorehabilitation splinting:
193. Eyberg SM, Nelson MM, Boggs SR. Evidence-based experiencing difficulty processing and integrating sen- theory and principles of clinical use. Neurorehabilita-
psychosocial treatments for children and adolescents sory information. Am J Occup Ther 2010; 64: 415–29. tion 2011; 28: 21–8.
with disruptive behaviour. J Clin Child Adolesc Psychol 202. Novak I, Russell D, Ketelaar M. Knowledge transla- 211. Doust J, Del Mar C. Why do doctors use treatments
2008; 37: 215–37. tion: can translation of research information improve that do not work? BMJ 2004; 328: 474–5.
194. Fabiano GA, Pelham WE Jr, Coles EK, et al. A meta- outcomes? In: Ronen GM, Rosenbaum PL, editors. 212. Tsorlakis N, Christina E, George G, Charalambos T.
analysis of behavioural treatments for attention-deficit/ Life Quality Outcomes in Young People with Neuro- Effect of intensive neurodevelopmental treatment in
hyperactivity disorder. Clin Psychol Rev 2009; 29: 129–40. logical and Developmental Conditions. London: Mac gross motor function of children with cerebral palsy.
195. Silverman WK, Pina AA, Viswesvaran C. Evidence- Keith Press, 2013: 265–81. Dev Med Child Neurol 2004; 46: 740–5.
based psychosocial treatments for phobic and anxiety 203. Novak I, Hines M, Goldsmith S, Barclay R. Clinical 213. Bar-Haim S, Harries N, Nammourah I, et al. Effec-
disorders in children and adolescents. J Clin Child prognostic messages from a systematic review on cere- tiveness of motor learning coaching in children with
Adolesc Psychol 2008; 37: 105–30. bral palsy. Pediatrics 2012; 130: e1285–312. cerebral palsy: a randomized controlled trial. Clin
196. Spittle AJ, Orton J, Doyle LW, Boyd R. Early devel- 204. Sakzewski L, Boyd R, Ziviani J. Clinimetric properties Rehabil 2010; 24: 1009–20.
opmental intervention programmes post hospital dis- of participation measures for 5-to 13-year-old chil- 214. Schlosser R. The role of single-subject experimental
charge to prevent motor and cognitive impairments in dren with cerebral palsy: a systematic review. Dev Med designs in evidence-based practice times. FOCUS
preterm infants. Cochrane Database Syst Rev 2007; 2: Child Neurol 2007; 49: 232–40. 2009; 22: 1–8.
CD005495. 205. Vargus-Adams J, Martin L. Measuring what matters 215. Hahs-Vaughn DL, Nye C. Understanding high qual-
197. Blok H, Fukkink RG, Gebhardt EC, Leseman PPM. in cerebral palsy: a breadth of important domains and ity research designs for speech language pathology.
The relevance of delivery mode and other programme outcome measures. Arch Phys Med Rehabil 2009; 90: Evid Based Commun Assess Interv 2008; 2: 218–24.
characteristics for the effectiveness of early childhood 2089–95. 216. Schlosser RW, Raghavendra P. Evidence-based prac-
intervention. Int J Behav Dev 2005; 29: 35–47. 206. Debuse D, Brace H. Outcome measures of activity for tice in augmentative and alternative communication.
198. Orton J, Spittle A, Doyle L, Anderson P, Boyd R. Do children with cerebral palsy: a systematic review. Pedi- Augment Altern Commun 2004; 20: 1–21.
early intervention programmemes improve cognitive atr Phys Ther 2011; 23: 221–31.
and motor outcomes for preterm infants after dis- 207. Harvey A, Robin J, Morris ME, Graham HK, Baker
charge? A systematic review. Dev Med Child Neurol R. A systematic review of measures of activity limita-
2009; 51: 851–9.

2013 - Novak - Interventions For Children With CP

Uploaded by

Copyright:

Available Formats

2013 - Novak - Interventions For Children With CP

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2013 - Novak - Interventions For Children With CP

Uploaded by

Copyright:

Available Formats

DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY REVIEW

A systematic review of interventions for children with cerebral

This article is commented on by Msall on pages 877–878 of this issue.

© 2013 Mac Keith Press DOI: 10.1111/dmcn.12246 885

886 Developmental Medicine & Child Neurology 2013, 55: 885–910

888 Developmental Medicine & Child Neurology 2013, 55: 885–910

890 Developmental Medicine & Child Neurology 2013, 55: 885–910

892 Developmental Medicine & Child Neurology 2013, 55: 885–910

894 Developmental Medicine & Child Neurology 2013, 55: 885–910

896 Developmental Medicine & Child Neurology 2013, 55: 885–910

apply since CP is a disability not a disease, and thus direct

codes using CP literature precedents, where the outcome

measure within the included trials had been ICF coded by

rules typically cluster together (1) body structure and func-

of findings obscured within aggregated data, we separated

activities from participation because we wanted to illumi-

nate whether or not participation outcomes were being

achieved. All the data required to answer the study ques-

authors was necessary.

Ethics and registration

for ethical approval was waived by the Cerebral Palsy Alli-

For the purpose of this study, participants had CP, which is

ies about children with CP of any motor subtype (spastic,

dyskinetic, or ataxic), any topography (hemiplegic/unilat-

eral, diplegic/bilateral, or quadriplegic/bilateral), and any

tions to reduce spasticity, the most prevalent motor impair-

ment.190 There was also a heavy emphasis in the therapy

literature on interventions designed to improve motor out-

severity of the disability. However, there was insufficient

by GMFCS level, which was our original intended strategy.

Levels of evidence and ICF

High Strong + Green: Go

High Strong – Red: Stop

Very low = Any estimate of effect is very uncertain

Figure 1: Relationship between the GRADE and Traffic Light System.

166 included studies, the breakdown by level of evidence

898 Developmental Medicine & Child Neurology 2013, 55: 885–910

Body structures and function interventions

900 Developmental Medicine & Child Neurology 2013, 55: 885–910

b Improved motor activities Improved function

Conductive Vojta Massage

Figure 4: State of the evidence for cerebral palsy intervention by outcomes.

902 Developmental Medicine & Child Neurology 2013, 55: 885–910

904 Developmental Medicine & Child Neurology 2013, 55: 885–910

906 Developmental Medicine & Child Neurology 2013, 55: 885–910

908 Developmental Medicine & Child Neurology 2013, 55: 885–910

910 Developmental Medicine & Child Neurology 2013, 55: 885–910

You might also like