A novel interactive training device to improve walking Ability and quality of life for Children with CErebral Palsy Trial: ACCEPT study An exploration of balance, its measurement and usual physiotherapy care in children with cerebral palsy, to inform a mixed methods feasibility RCT.

Cerebral palsy (CP) is a common disability with its onset in childhood, resulting from a non-progressive lesion to the brain. Motor impairments associated with CP are spasticity, weakness and reduced selective movement; these interfere with various aspects of balance. Children undertake physiotherapy to maintain or improve their mobility but find it hard to train at the correct frequency, intensity and in functional positions. A novel interactive trainer, the Happy Rehab™ (Innovaid, Denmark), was designed to address this. Its efficacy as an intervention to improve walking and balance is unknown. To evaluate this requires a Randomised Controlled Trial (RCT). The usual care to which it is being compared must first be defined. It is also essential to understand the mechanisms of balance in order to select an appropriate outcome measure. Current clinical tests for balance do not capture the selective deficits in postural control that occur during functional movements. Therefore, a new clinical test for measuring dynamic balance is required to determine the effectiveness of such interventions. The main aim of the thesis as to determine the feasibility of a RCT investigating the clinical effectiveness of training for 10 weeks using the Happy Rehab interactive trainer and the impact on the lived experience and adherence to therapeutic exercise programmes by children with CP and their families. Four objectives were identified to meet this aim. Firstly, to determine the reliability and validity of a measure of dynamic balance to use as an outcome measure in the RCT. Secondly, to define and describe usual physiotherapy care to inform the control group of the RCT. Thirdly to establish the feasibility of an RCT using the Happy Rehab interactive trainer. Finally, to explore the experiences of children, parents and clinicians of their participation in the feasibility study to inform the effective delivery of a full RCT. The thesis starts by exploring the mechanisms of balance used to inform the development of The Next Step test. It was developed through observations of the anticipatory postural adjustments (APAs) of children making a single step to medial and laterally placed targets. The validity and reliability of the Next Step test was established in (n=16) ambulant children with CP and (n=14) typically developing (TD) children. Children with CP had smaller medio-lateral APAs and greater stepping error. The grand average of the peak medio-lateral motion of the centre of pressure (ML COP) was identified as the Next Step summary score for use in the feasibility RCT. Usual physiotherapy care was explored through a consensus process involving twelve physiotherapists. It used surveys, nominal group technique and a literature review. The resultant usual care checklist informed the control group of the feasibility RCT. The feasibility RCT recruited fifteen children with CP, aged 8-18, with Gross Motor Function Classification System (GMFCS) levels I-III. They were randomised into two groups. The intervention group (n=8) undertook 10 weeks training using the Happy Rehab. The control group (n=7) carried out usual physiotherapy care. The Pediatric Balance Scale and the peak ML COP were assessed at baseline, 10 and 20 weeks. Families kept e-diaries to record adherence. Data were summarised using descriptive statistics. The PBS was more complete than the Next Step test at 10 weeks, PBS (100%) and Next Step (91%). Three children reached the ceiling of the PBS. Signals of efficacy of the intervention were detected in PBS, Next Step test as well as clinical measures of strength, spasticity and range of movement. Qualitative interviewing was conducted with three physiotherapists, nine parent-child dyads, from the control group (n=4) and the intervention group (n=5) to understand participants’ views on the feasibility of the trial and intervention. Data generation with the children was undertaken using semi-structured interviews incorporating a photo-elicitation method at the end of week 10. One decliner (n=1) interview was undertaken. Data were transcribed and analysed thematically using triangulation with the diary entries. Five main themes were identified from the interviews: ‘Fitting therapy into normal life’, ‘Motivation to exercise’, ‘The opportunity to try something new’, ‘Physios out of their comfort zone’, and ‘Altruism and the burden of participation’. Children found the gaming aspect motivating and enjoyable. Recruitment and the technical support for the intervention were impacted by the COVID pandemic. In conclusion, the Happy Rehab intervention showed signals of efficacy for children with CP who are able to use the device at home for 10 weeks. However, some families could not accommodate the device at home. Physiotherapists required ongoing technical support to set up the device. The proposed RCT requires further work to establish a suitable primary outcome measure. A full RCT is not feasible without technical support for devices in the UK