Outcomes of importance to children and young adults with cerebral palsy, their parents and health professionals following lower limb orthopaedic surgery: A qualitative study to inform a Core Outcome Set

Abstract Introduction Although several outcomes are commonly measured to assess the effect of surgery for young people with cerebral palsy (CP), these are selected mainly by health professionals and researchers. Including the perspectives of a broader range of stakeholders is an essential step towards determining important outcomes for assessment. This qualitative study involves the development of a core outcome set (COS) for lower limb orthopaedic surgery for ambulant children with CP. Objective This study aimed to identify outcomes that matter to children and young people with CP, their parents and healthcare professionals following lower limb orthopaedic surgery. Methods Semi‐structured interviews were conducted with 10 healthcare professionals, 10 children and young people with CP and 8 parents. Interview data were analysed by content analysis supported by the International Classification of Functioning, Disability and Health (ICF‐CY) supplemented by thematic analysis. Findings Thirty‐one outcomes were identified in total, which were linked to eleven second‐level ICF‐CY categories. There were differences between stakeholder groups in preferences and expectations from surgical outcomes. Healthcare professionals and children with their parents identified 31 and 25 outcomes, respectively. Health outcomes valued by participants were lower limb alignment and symmetry, flexibility and muscle strength, mental health, fatigue, pain, function in life, mobility, participation, being independent, quality of life and adverse events. Compared to previous published trials, 10 new outcomes were revealed by this study. Conclusion The researchers identified outcomes that are important to all stakeholders following lower limb orthopaedic surgery for ambulant CP. Including these outcomes in future studies would promote patient‐centred care for children and young adults with CP. Findings will be used to inform an international Delphi survey and develop a COS in this field. Patient and Public Contribution This study was informed by an advisory group including a young adult with CP and a parent of a child with CP. This group engaged in the design of the study and the information material to support the interview (information sheet and interview topic guide).

and parents' priorities in measuring outcomes after orthopaedic surgery. To our knowledge, no qualitative study has been undertaken to inform a COS in the field of lower limb orthopaedic surgery for CP. The role of qualitative research in the development of COS has been previously explored, 12,13 and has been advocated by groups such as the Core Outcome Measures in Effectiveness Trials initiative to ensure that the outcomes being considered for prioritisation are comprehensive. 9 This study forms part of a COS development project, of which the protocol has been previously described and published. 14 The first stage in the project involved identifying a 'long-list' of potentially important outcomes. These were prioritized during the second stage, which included a Delphi survey and a final consensus meeting. In the study reported here, we explored the experiences, perceptions and priorities of stakeholders (children and young people with CP, their parents and healthcare professionals) who had been involved with lower limb orthopaedic surgery. The aim is to identify outcomes that have not been previously reported in the literature but matter to the relevant stakeholders. The findings informed the development of a COS in lower limb orthopaedic surgery for ambulant CP, 14 and was conducted and reported using COnsolidated criteria for REported Qualitative research (COREQ). 15 2 | METHODS

| Qualitative descriptive study
The researchers used a qualitative descriptive approach using content analysis of semi-structured interviews with relevant stakeholders (health professionals, children with CP and their parents). A semi-structured interview, guided by a list of essential topics, was chosen to facilitate open discussion and to avoid a closed question and answer format. 16 The participants' experiences of relevant outcomes were explored.
Previous research with children with CP found that the concept of health outcomes was difficult for children to grasp quickly; 17 therefore, a child-friendly technique using Talking Mats (https:// www.talkingmats.com/) was used with the children (8-15 years) to enhance participation and interest in answering the questions.
This study was approved by the Research Ethics Committee (19/SC/0357) and received R&D approval from the relevant NHS Trust.

| Study sample and sample size
A purposive approach to sampling was used to recruit stakeholders with a range of demographic characteristics such as children's age, Gross Motor Function Classification System (GMFCS) level, 18 surgery type and age at surgery, health professional's background and years of experience. The sampling strategy aimed to capture a range of views concerning the outcomes of orthopaedic surgery treatment.
The sample included children with CP and their parents, health professionals specializing in CP lower limb orthopaedic interventions and health researchers with an interest in CP or children with disability. Table 1 presents the sample criteria. The sample was drawn from a leading hospital in the UK specializing in orthopaedic surgery for children with CP.
There are no agreed criteria for determining sample size in qualitative research. 19 Participants were included in the study until data saturation was reached. Data saturation was defined in this study as the point at which three consecutive interviews generated no additional data. 20,21

| Recruitment
Health professionals and researchers based at the tertiary hospital where recruitment was conducted were invited to participate in the study if they fulfilled the criteria (Table 1). An email invitation was sent to potential participants by the research team. Participants were offered the opportunity to have their interview in private, either at the hospital or at the university department. Informed consent was obtained before the interview and confidentiality was maintained throughout the study.
Children and their parents, who attended the paediatric orthopaedic service at the hospital between October 2019 and June 2020 and fulfilled the criteria (Table 1), were approached by their direct care team during the child's consultation appointment. The invitation included an information sheet indicating that participants could withdraw from the study at any time. Children were given the choice to be interviewed with or without a parent present. Written consent was obtained from all participants (children and parents) on the day of the interview. Participants' demographic information was collected during the interview to ensure that the sample fulfilled the purposive sampling criteria.

| Interviews
A semi-structured interview guide was developed with a range of stakeholders, including specialists in CP orthopaedic surgery, a social scientist, researchers, children and young adults with CP and parents of a child with CP to ensure that all questions were open, relevant and sensitively worded. The interview guide included a series of open-ended questions based on a topic guide to ensure that key areas were covered (File S1).
Interviews initially focused on the participants' current day-today life experiences and surgical interventions related to lower limbs.
Participants were asked to reflect on their priorities after surgery, how these priorities changed following their surgical intervention and how their priorities might differ compared to other stakeholders' groups.
To facilitate the discussion and communication during the children's interviews, the 'Talking Mat' was used. 22 This is an interactive resource using picture symbols to facilitate communication and  to facilitate the development of ideas during ongoing interviews.

| Data analysis
A qualitative content analysis was performed 25 using a deductive approach, with the ICF-CY as a framework to categorize the data. 5 The interviews were read through to obtain an overview of the data and to identify meaningful concepts from the interviews. In this context, 'Meaningful concepts' refer to those that reflect the essence of what participants are saying. These meaningful concepts were linked to the most relevant ICF-CY categories following a set of formal rules and procedures as determined by Cieza et al. 26 In addition, inductive thematic analysis 27 was performed to develop new categories from the meaningful concepts that did not fit the chosen framework.

| Linking to the ICF-CY framework
According to established linking rules, 26 each line of text was coded according to its meaningful concepts and was linked to an ICF-CY category. Some data could be linked to more than one ICF category.
For example, for the statement 'for the ambulatory kids, they can have discomfort because of their spasticity or because of joint problems. So that would always be my prime goal', the meaningful concepts would be 'muscle spasticity' and 'joint problem', and the ICF categories linked to them would be 'b735 muscle tone function' and 'd710 mobility of joint function'. If the meaningful concept was not explicitly named in the ICF-CY category, the 'nc; not covered' category was applied. If there was insufficient information to decide about which ICF category should be linked, the concept was assigned as 'nd; not defined' and this included concepts referring to the quality of life (nd-qol).

| Thematic analysis
Meaningful concepts that could not be linked to ICF-CY categories (nc, nd or nd-qol) were pooled together and analysed thematically. 27 Thematic analysis is a process where meaningful concepts are compared and organized into categories of 'themes', which share a common essence.

| Accuracy and rigour of the analysis
To ensure consistency of linking results, identification of meaningful concepts and linking of ICF-CY categories, the linking was conducted independently by two qualitative researchers trained in using the ICF-CY classification system. For data that did not fit the ICF-CY framework, two researchers placed meaningful concepts into categories. Differences arising in interpretation between the researchers were resolved through discussion.

| Participants and interviews
A total of 20 interviews were conducted (10 health professionals, 8 children and parent dyads, 2 young adults with CP individually).
Participants' demographic data are shown in Tables 2 and 3. All professionals were specialists in treating children with CP and lower limb deformities, with their experience ranging from 3 to over 20 years. The age range of the children interviewed was 10-17 years.
The severity of the condition, based on the GMFCS level, varied: Level I (n = 1), Level II (n = 5) and Level III (n = 4). Talking mats were used in three interviews. An example of a talking mat is shown in

| Outcomes identified
Thirty-one outcomes were generated from the data, and these were organized into five categories. Two categories were linked to the ICF-CY framework: (1) body function and structure and (2) activity and participation. Three additional categories were generated from a body of data that did not fit the ICF-CY framework: (3) independence, (4) quality of life (QoL) and (5)   In contrast, some health professionals were doubtful that correction of joint alignment would lead to functional improvement, but rather, they thought that it was important because parents and children felt that symmetry was integral to a positive body image.
If one leg's turning in or something. It won't necessarily be because it impairs them functionally, but [the] key focus is to get mechanically their walking as good as possible, which mostly means making their pattern as close to normal as possible. HP04 Health professionals emphasized that they would not expect surgery to reduce the need for an assistive walking device and highlighted situations where children's and parents' expectations of walking capacity postoperatively were not aligned. They described differences in these views between children and their parents, mainly related to the use of assistive devices. In contrast, some participants felt that using an assistive advice would increase independence.
As an example of this finding, one young adult and her mother described striving towards walking independently as: I haven't had someone say, 'Well, you won't be able to walk independently', but I've also had the physiotherapist ALMOAJIL ET AL.

| Independence
Being independent in daily life activities, self-care and walking were considered as important postoperative outcomes. Some felt that it was important to achieve as much independence as possible during childhood and noted the efforts of parents to help children achieve independence.
Independence is a big thing because before my third Participants viewed outcomes as interrelated, with some of them affecting higher-level outcomes. For example, participants described the importance of improving joint flexibility and muscle strength to be more active, as this in turn would improve QoL. The inter-relationship of outcomes has previously been identified and described for children with neurodisability. 17 This highlights the need to consider and identify the degree to which higher-level outcomes reflect primary clinical outcomes. These primary clinical outcomes are the stepping-stones that lead to the achievement of higher-level outcomes and longer-term changes. Our findings indicate the need for a comprehensive assessment approach as highlighted by the ICF-CY, to illustrate the interaction between outcomes and their impact.
Our findings included ten additional outcomes, beyond those identified in a recent scoping literature review ( The range of identified outcomes indicates the importance of

| STRENGTHS AND LIMITATIONS
This qualitative study preceded the consensus phase of developing a COS to be used in the field of lower limb orthopaedic surgery. 14 The study identified additional outcomes that had not been reported by a recent literature review on CP surgical trials. 12,35 The coding and analysis were undertaken by a minimum of two experienced researchers. Our methods were established a priori in a study protocol that had been subjected to a robust peer-review process. 14 A potential limitation of this interview study is that the participants were recruited from a single orthopaedic hospital.
However, this hospital is a tertiary referral centre covering a wide geographic area of South England. There may be outcomes that are relevant to other national and international stakeholders that have not been identified. We limited the interviews to UK-only participants for pragmatic reasons. A future consensus study (Delphi survey) that will form part of the COS development will be available internationally.
Qualitative research does not report statistical findings and does not aim to represent a particular population: Rather, it aims to generate ideas from a purposive sample that will enhance our understanding of a particular experience. Sociodemographic data were not collected for this study. The strength of our sample was that it was drawn from a specialist orthopaedic centre that serves a wide geographical area: Our primary aim was to develop a list of important outcomes from a purposive sample of stakeholders with experience, or professional interest, in CP. Future studies to explore the impact of different demographic variables, including ethnicity, education and income, on preferences and experiences would be useful.
Most of the children and parents were interviewed together, and this may have had an impact on the responses of both parent and child, which may have been different if they had been interviewed alone. However, the team considered that the most ethical approach was to give the children the choice to undertake an interview with or without a parent. This decision was in line with guidance following ethical review.

| CONCLUSION
This study identified 32 outcomes that are important to all stakeholders involved with lower limb orthopaedic surgery in CP. Some of these have not been previously identified or used in trials within this study field. It is important that future studies in this field should report on outcomes that are considered important by children, young people and their parents.