Defining usual physiotherapy care in ambulant children with cerebral palsy in the United Kingdom: A mixed methods consensus study

Abstract Background Ambulant children with cerebral palsy (CP) undertake physiotherapy to improve balance and walking. However, there are no relevant clinical guidelines to standardize usual physiotherapy care in the United Kingdom. A consensus process can be used to define usual physiotherapy care for children with CP. The resulting usual care checklist can support the development of clinical guidelines and be used to measure fidelity to usual care in the control groups of trials for children with CP. Methods Twelve expert physiotherapists were recruited. In Phase 1, statements on usual care were developed using a survey and two nominal groups. Phase 2 included a literature review to support usual physiotherapy interventions. Phase 3 used a confirmatory survey, which also captured changes to provision during the COVID‐19 pandemic. Consensus was calculated by deriving the mean of the deviations from the median score (MDM). High consensus was deemed to be where MDM < 0.42. Results Physiotherapists reached high consensus on five outcome measures (MDM range 0–0.375) and nine areas of assessment (MDM range 0–0.25). Physiotherapists reached moderate consensus on task‐specific training (MDM = 0.75), delivered at weekly intensity for 4–6 weeks (MDM = 0.43). There was high consensus (MDM = 0) that children should participate in modified sport and fitness activities and that children with Gross Motor Function Classification System Level III should be monitored on long‐term pathways (MDM = 0.29). Conclusions Physiotherapists reached consensus on two usual care interventions, and a checklist was developed to inform the control groups of future randomized controlled trials. Further consensus work is required to establish clinical guidelines to standardize usual physiotherapy care in the United Kingdom.


| INTRODUCTION
Cerebral palsy (CP) is an umbrella term describing a group of permanent disorders affecting the development of posture and movement affecting 2.1 per 1000 children (Oskoui et al., 2013). Motor impairments associated with CP make walking more effortful and significantly limit children's participation at school and in the community (Kamp et al., 2014). Children with CP can experience primary movement impairments such as spasticity, weakness or reduced selective movement control (Rosenbaum et al., 2007). The severity of the movement disorder can be described using the Gross Motor Function Classification System (GMFCS) (Palisano et al., 1997).
Children with GMFCS Levels I-III are able to walk with varying levels of support or orthoses and tend to achieve their peak motor performance by age 9 (Hanna et al., 2008;Palisano et al., 2007).
However, secondary musculoskeletal impairments can develop during periods of rapid growth, presenting further challenges to walking and balance skills.
Physiotherapists provide advice and therapeutic interventions aimed at addressing primary impairments and preventing secondary complications of CP. Young people with CP and their families want to know which physiotherapy interventions are the most effective and the frequency and intensity required to achieve optimum mobility (Morris et al., 2015). Physiotherapy service provision may vary depending on resources and how emerging evidence (Franki et al., 2012;Hägglund et al., 2014;Novak et al., 2013) and national guidance is implemented (Mugglestone et al., 2012). Currently, there is no standardization of physiotherapy care for ambulant children with CP in the United Kingdom.
The highest level of evidence for the effectiveness of an intervention is through meta-analysis of randomized controlled trials (RCTs) (Centre for Evidence-Based Medicine, OCEBM Levels of Evidence Working Group, 2011). In many physiotherapy studies, the control group undertakes 'usual care', but this is often unspecified. Usual care across studies is likely to vary in the frequency and intensity of physiotherapy, and participants in a control group could be undertaking activities similar to the experimental intervention. It is essential to define usual care within the research setting to ensure the effect size of an intervention within a trial is correctly measured. Therefore, a definition of usual care is crucial to ensure robust research findings and to inform the development of evidence-based clinical pathways (Royal College of Physicians, 2016).

| METHODS
The aim of this study was to reach consensus on current This study adopted a three-phase design (Figure 1). Phase 1 used idea generation and nominal group technique (NGT) to establish consensus statements on usual physiotherapy care aimed at improving balance and walking in children with CP, GMFCS I-III.
Phase 2 was a literature review to establish the evidence base underpinning the interventions identified in the consensus statements. Phase 3 used a survey to confirm consensus on the usual care checklist.
Two NGs were established in Phase 1. The first NG consisted of six paediatric community physiotherapists from National Health Service (NHS) providers in South West UK. The physiotherapy managers of five child development centres recruited participants. They gave information packs to interested clinicians. The manager was asked to nominate one or two staff volunteers to participate during work time. A national NG was formed with six community physiotherapists from the rest of the United Kingdom. Adverts were placed in the Association of Paediatric Chartered Physiotherapists e-bulletin.
Interested physiotherapists were invited to respond directly to the chief investigator, who sent them an information pack. Participants were eligible if they had over 2 years of experience in paediatric physiotherapy and held a current community paediatric caseload in the United Kingdom, with an NHS provider. In Phase 3, all 12 participants from Phase 1 were invited to complete a confirmatory survey.

Key messages
• A checklist of usual physiotherapy care in the United Kingdom has been developed for ambulant children with cerebral palsy to inform the control groups in randomized controlled trials.
• Usual physiotherapy care should include task-focused therapy, facilitation of modified sport and participation in community activity.
• Physiotherapy tools were identified for the assessment of balance and mobility and measurement of treatment outcomes.
• Children with Gross Motor Function Classification System Level III should remain on long-term monitoring pathways.
• The usual intensity of physiotherapy treatment in the United Kingdom is weekly for 4-6 weeks and is lower than that which is reported to be effective in research literature.

| Phase 1: Development of consensus statements
Phase 1 employed the NGT, a consensus process that encourages individual participation and a non-hierarchical exchange of ideas (Ven & Delbecq, 1974). It has previously been used within physiotherapy to reach consensus on interventions that influence motor development in children with CP (Bartlett & Palisano, 2002). NGT involves a three-stage process of decision-making during a structured group meeting led by a skilled, neutral facilitator (Bartlett & Palisano, 2002;Delbecq et al., 1975

| NGs
The lead author, an experienced paediatric physiotherapist and researcher, facilitated the NGs. Her position at the group was of a neutral facilitator, and other members of the research team supported the process: HH documented notes, and JM administered the scoring. Participants were asked to consider the minimal physiotherapy care usually undertaken by a physiotherapist, regardless of NHS setting. Careful consideration was given to the scope of the physiotherapy role. Participants excluded the provision of orthotics, as orthotists are autonomous practitioners responsible for the assessment and prescription of orthotics.
The statements on usual care were presented to participants at the beginning of the SW NG. Participants scored their level of agreement with each statement using a 5-point Likert-type scale (1 = strongly disagree, 2 = disagree, 3 = undecided, 4 = agree and 5 = strongly agree). The mean group score was calculated for each statement at the end of each scoring round. Participants were presented with the group median score alongside their individual scores for each statement. The facilitator encouraged a round-robin feedback from the participants for each statement. Participants explored the relative merits of each statement and were able to evaluate their ideas compared with those held by the group. Participants discussed and then revised the statements. The group revisited any ideas previously set aside for further discussion to see if they wished to include them.
For example, hydrotherapy was a subject initially set aside, and was revisited by both groups, but remained excluded. Participants rescored all the statements where consensus was not reached in the previous scoring round.

F I G U R E 1 Flow diagram showing the three phases of the consensus study
The statements on usual care developed during the South West (SW) NG were presented at the beginning of the national NG, in an iterative process. The national NG decided to include an idea that had been excluded by the SW NG. This was related to the importance of advocating wheelchair mobility for children assessed as GMFCS Level III. This was taken forward into Phase 2.
At the end of Phase 1, the levels of consensus for the 10 statements on usual physiotherapy care were calculated for each NG. Six physiotherapy interventions were proposed by the NGs as usual care.

| Inclusion and exclusion criteria and study selection
Systematic reviews or studies in the English language were included where they reported physiotherapy interventions with outcomes related to walking and balance. Where no systematic review was found, RCTs and then experimental studies were included. Papers were excluded where the results are reported in a systematic review or were superseded by more recent studies. Protocol-only publications and papers that did not report an outcome relating to balance or walking were excluded. The results are presented in accordance with PRISMA guidelines (Moher et al., 2011). Figure 2 shows that of the 670 abstracts reviewed, 105 full papers were retrieved for abstract review; of these, there were 75 systematic reviews, 29 RCTs and one experimental design study. Only 15 papers met the criteria for full review and were assessed for bias using the CASP tool (Hannes et al., 2010). These comprised 12 systematic reviews (Anttila et al., 2008;Bania et al., 2019;Booth et al., 2018;Clutterbuck et al., 2019;Corsi et al., 2021;Dewar et al., 2015;Elnahhas et al., 2019;Franki et al., 2012;Galey et al., 2017;Moreau et al., 2016;Novak et al., 2020;Ritzmann et al., 2018;Tustin & Patel, 2017), two RCTs (Kimoto et al., 2019;Valentín-Gudiol et al., 2017) and one non-randomized crossover trial (Salem et al., 2010). The strength of evidence for interventions identified as usual care were rated as high, moderate, low or very low levels of evidence (Balshem et al., 2011) (see Table 1).

| Phase 3: Confirmatory survey
The final online survey allowed participants to score subsections of each statement of usual care in more detail. For example, participants were asked to rate individual assessment tools from the list identified in Phase 1 using the 5-point Likert-type scale. Interventions were presented alongside the evidence summary (Table 1), and participants were asked to indicate whether they thought the intervention should be included or excluded as usual care or if they were undecided.
Participants were asked to comment on why they decided to award each score in order to gain more insight into their views and experiences.

| RESULTS
Twelve physiotherapists participated across the two NGs in Phase 1. The median age of participants was 43 years (range 28-60) with a median level post qualification of 21.5 years (range 7-38) with 18.5 years (range 3-29) in paediatrics. Table 2 shows the similarity between both NGs. Eight of the 12 participants completed the Phase 3 confirmatory anonymous survey.
Participants developed 10 statements on usual care during the NGs. They described six areas of intervention to be included in the literature review: participation in physical activities, flexibility exercises, prolonged passive stretching; strength training; and task-specific or functional activity training. Participants identified a list of assessment tools and outcome measures to be included in the confirmatory survey. Both groups reached a high level of consensus (MDM < 0.42) for all 10 statements on usual care at the end of the Phase 1 process (Table 3). Participants in the SW group tended to award a higher median score for each topic.
The literature review appraised evidence for the six interventions identified as usual care during Phase 1. Evidence for each intervention was explored in relation to outcomes of walking, balance and gross motor function. The evidence summary (Table 1) shows moderate to low evidence to support fitness training and modified sport. There was low evidence supporting prolonged passive stretching (excluding orthotics) using serial casting or prolonged standing frame use. There was moderate to high evidence against the use of progressive strength training. Strength training did not improve gait characteristics or postural control and was associated with multiple adverse events. Task (Table 4).

| Outcome measurement
In Phase 1, participants developed a list of 17 outcome measures used to evaluate episodes of care.

| Physiotherapy interventions
In Phase 1, participants reached a high level of consensus on a list of interventions considered as usual care (Table 3). However, after consideration of the evidence summary (Table 1) presented alongside the survey, participants only reached consensus on including two of the six interventions into the usual care position statement (Table 6).

| Participation in sport and activity
There was high consensus (MDM = 0) that the physiotherapist's role is to encourage physical activities and facilitate children to access school and community resources to develop active lifestyles.
Physiotherapists considered that the level of daily activity makes an important difference to the outcomes of children. They recognized that the level of support from home and school is critical, for example:

| Prolonged passive stretching
In  All participants said that essential face-to-face visits were possible for some children at home or at COVID-secure premises.
Many respondents reported that they provided an assessment and management programme, but they were unable to offer routine monitoring or blocks of treatment at the height of the pandemic. The overall frequency and amount of contact per child has therefore reduced dramatically. All participants reported that children had reduced levels of activity in lockdown due to lack of access to sports facilities at school and in the community.

| DISCUSSION
In this study, we explored ideas of what constitutes usual physiotherapy care to improve walking and balance for ambulant children with CP in the United Kingdom. The study used an NG consensus process.
We examined the evidence supporting the interventions usually employed and developed a checklist of usual physiotherapy care for use in a future RCT (Appendix A).
We found a high level of consensus among physiotherapists to support the long-term monitoring of children with CP at risk of musculoskeletal decline. This approach is backed by a growing evidence base that advocates routine surveillance of hip migration, joint range of motion and spinal posture for all children with CP (Cans, 2000).
Where services do not currently include all children with CP in surveillance programmes, they give 'red flag' indicators for enabling timely access back into services. Physiotherapists play an essential role in identifying the need for orthotic and postural management equipment to optimize posture and mobility for children with CP.
Physiotherapists use collaborative goal setting to inform the need for treatment blocks usually delivered at an intensity of once per week, for 4-6 weeks. This contrasts with the frequency and intensity of usual physiotherapy care reported in some RCTs as 1-3 sessions of 30-60 min per week (Scholtes et al., 2007(Scholtes et al., , 2012. Participants reached moderate consensus that task-specific functional activity training should be included in usual care to improve balance and mobility. This is supported by both the National Institute of Clinical Excellence guidance (Mugglestone et al., 2012) and the evidence summary produced from the literature review. However, the reported frequency and intensity falls short of the dose reported to be effective in the literature. Intensive programmes delivered daily for 2 weeks have been shown to achieve the greatest functional improvements (Bleyenheuft et al., 2015). This level of resourcing for physiotherapy treatment programmes was not found within our study, which brings into question the ecological validity of these studies. Physiotherapy services in the United Kingdom might consider the efficiency gains of deploying current resources in a more concentrated way.
Physiotherapists in our study applied the principles of research findings by integrating gait training in community, home and school activity programmes.
We found further divergence between the evidence and usual  Referral and discharge criteria a Children and young people with GMFCS III are managed on a long-term multidisciplinary care pathway from initial referral to transition into adult services. b Children with GMFCS Levels I and II are offered episodes of care related to their functional needs and are discharged where there are no identifiable needs or their musculoskeletal condition is stable. c Where children are discharged, information is given to them, and their carers to help them identify key triggers/red flags for timely re-referral into the service.
d Ambulant children have ongoing access to orthotics as required.
2 Location of physiotherapy a Children are offered virtual consultations and face-to-face appointments, as appropriate. b Appointments take place in a children's physiotherapy department or a child-friendly general outpatient clinic setting.
c Appointments are offered in school or at home when indicated due to environmental needs or comorbidities.

Assessment
The following core areas should be included in assessment:

5
Physiotherapy interventions a Functional, task-specific training is used to treat walking and balance difficulties. Adjuncts to task-specific training include using a treadmill with or without body-weight support.
b Vibration training is used to improve balance and posture (where equipment exists).
c Physiotherapists encourage and facilitate children to develop active lifestyles including aerobic exercise, fitness training and modified sport.
d Strength training using progressive resisted exercise is not employed as a treatment to improve walking and balance.
e Prolonged passive stretching is used to manage contractures using orthotics, serial casting and supported standing programmes.
f Postural management approaches are employed for children with GMFCS III, including mobility equipment and environmental adaptation.
g Exercise and functional activities that encourage full joint range are recommended where there is risk of contracture development.
APP E NDIX A: CHECKLIST OF USUAL PHYSIOTHERAPY CARE AIMED AT IMPROVING WALKING AND BALANCE FOR AMBULANT

CHILDREN WHO HAVE CEREBRAL PALSY
Please enter an X in each box to indicate criteria are met.