The aim of this section is to provide an overview of the study design and the methods of data collection. Each of the four phases is described in turn, followed by a section on data analysis for all phases and our patient and public involvement (PPI) activities. The original study protocol has been published1 and parts of this chapter have been reproduced from that publication [published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://www.bmj.com/company/products-services/rights-and-licensing/. This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: http://creativecommons.org/licenses/by/4.0/. The text below includes minor additions and formatting changes to the original text].
A list of protocol amendments is provided in Report Supplementary Material 2.
Research design
A transformative, mixed-methods case study design67 was used, integrating quantitative and qualitative methods across four phases over 3 years (Figure 2). Acknowledged for giving ‘a voice to the powerless and voiceless’,68 the case study approach enabled the views of children with LD and their parents to be prioritised and explored ‘in depth and within its real-life context’.69 In this study, a single hospital site represented each case, and seven cases (phase 2) were included: four specialist children’s hospitals and three non-specialist hospitals. For every two children with LD recruited one child without LD was also recruited, allowing the experiences of the two groups of patients to be compared. Data from diverse sources were synthesised to enable an understanding of whether or not children with LD and their families receive equal access to high-quality hospital care and services. The production of thick, rich descriptions of the phenomena, using in-depth interviews and creative research methods, meant that the complexities of the situation and the factors that can contribute to those complexities emerged.70
FIGURE 2 Four-phase transformative mixed-methods study design. The numbers included in the figure are recruitment targets and not the actual recruitment numbers. (Figure not reproduced here; see the original article.)
Phase 1 involved mapping the organisational context for the delivery of hospital care to children with LD, as well as comparing hospital staff members’ perceptions of their ability to identify the needs of children with and children without LD and their families and to provide high-quality care to effectively meet those needs.
Phase 2 compared how hospital care and services are experienced by children with and children without LD and their families, and staff members’ views of caring for them. A sample of staff from NHS community trusts were also surveyed.
Phase 3 used child and parent questionnaires to compare levels of satisfaction with inpatient hospital care.
Phase 4 synthesised the study findings to develop the outline content for a staff training DVD (digital versatile disc).
Approvals
Full ethics and health research authority approval for this study was obtained prior to the study commencing (London–Stanmore Research Ethics Committee: reference 16/LO/0645). Local research and development approval was also obtained from each of the 24 participating hospital sites.
Phase 1 research methods and analysis
Sample and setting
Children in England are treated either in specialist (tertiary) children’s hospitals (which may stand alone or be part of a wider NHS trust) or in general hospital settings (secondary care) that have one or more paediatric wards. All 15 specialist children’s hospitals in England agreed to participate following an invitation e-mail sent to the trust’s chief nurse through the Association of Chief Children’s Nurses (ACCN).71 A sampling strategy informed the recruitment of a selection of non-children’s hospitals, based on their proximity and referral patterns to the specialist children’s hospitals and their throughput of children with LD. Senior clinical or managerial staff with a specific responsibility for LD or those working in a dedicated LD role were eligible to participate in the interviews. All clinical and non-clinical hospital staff who had contact with children were eligible to complete the survey. A local collaborator for each participating site was identified.
Recruitment and consent
Staff who were eligible for interview were identified and approached by the local collaborator and given an information sheet and a consent form. With the staff member’s permission, their contact details were given to the research team, who telephoned the staff member to answer any questions and agree a date for the face-to-face or telephone interview. Staff who were eligible to be surveyed were e-mailed a link to an online survey by the local collaborator, with paper copies also available. The return of a completed survey was taken as consent to participate.
Data collection
Staff interviews
Interviews were semistructured, followed an interview guide (see Report Supplementary Material 3), and focused on the delivery of services to children with LD at the organisational level, including policies, systems and practices. Interviews lasted 30–45 minutes and were audio-recorded and transcribed verbatim. A minimum of two interviews per site were planned to ensure completeness of data.
Staff survey
A review of the literature informed the survey development in consultation with experts in the field and parent members of the Study Steering Committee (SSC). The survey, which was anonymous, had two parts; the first focused on children with LD and long-term conditions and the second focused on children without LD with long-term conditions. The survey focused on five elements of care (capability, capacity, confidence, safety and values) for those with and those without LD, with additional questions regarding access to care and processes used to identify and track those with LD. Likert scales were used for each question, with the majority of questions rated on a five-point scale of ‘strongly agree’ to ‘strongly disagree’. Definitions of LD and long-term conditions were provided for clarification (see Report Supplementary Material 4).
Hospital policies
Sites were asked to provide policies related to the following areas of practice: (1) LD policy (or equivalent), (2) patient (children and young people) experience, (3) child protection, (4) complaints, (5) safeguarding, (6) communication and (7) restraint and holding.
Phase 2 research methods
Phase 2 compared how hospital care and services are experienced by children with and children without LD and their families, and staff’s views of caring for them. A sample of staff from NHS community trusts were also surveyed.
Setting and sample (hospital sites)
A subset of phase 1 hospitals participated in phase 2 data collection, based on the:
- strength of the organisational context for delivery of care to children with LD
- staff’s perceived ability to identify and meet the needs of children with LD
- appointment of a learning disability liaison nurse who had the remit to improve care for children with LD.
To enable an objective selection of sites, scoring criteria were developed by the SSC, which was provided with anonymised information relating to four domains: (1) phase 1 recruitment, (2) hospital facilities and geography, (3) model of care and (4) phase 1 survey results. The executive team applied the scoring criteria, a shortlist was compiled and seven hospitals were included: four specialist children’s hospitals and three non-specialist hospitals (see Appendix 1, Figure 34). All sites had a principal investigator who was responsible for the study and a research nurse or equivalent who screened and approached eligible patients and provided day-to-day contact for the research team.
Sampling (participants)
Phase 2 involved data collection with a range of parents and children and staff working in both the hospital and the community (Table 1). A range of inclusion and exclusion criteria were applied (see Appendix 2).
| Participants | Sampling strategy | Sample size |
|---|---|---|
| Children and parents | A purposive sampling strategy using a sampling matrix to ensure diversity according to severity of LD, age, admission type, length of stay | 56–64 families |
| Hospital staff | A purposive sampling strategy using a sampling matrix to ensure diversity in terms of professional role. Includes subset of staff identified by families as potential participants | 112–28 |
| Ward managers | One from each study ward | 12 |
| Community staff | A convenience sample of staff from NHS community trusts providing health-care services to children in geographical proximity to the phase 2 hospital sites | 280–320 |
We did not recruit children aged ≥ 16 years because we anticipated that they would be in the process of transitioning to adult services, which would make it challenging to draw out similarities and differences in care and for children to be able to accurately reflect on only children’s services, and also because of the increasing body of work related to transition generally.
Operational definition of learning disability
The theoretical definition of LD is not always easily operationalised in practice. Among very young children, only severe LD is likely to be apparent,72 and some children never receive a formal diagnosis of LD, with medical records often stating ‘global developmental delay’ or a ‘syndrome without a name’. For the purposes of the study, a child was classified as having LD if any one of the following was documented in their medical notes:
- a diagnosis of LD
- a condition always accompanied by a degree of LD (e.g. Down syndrome)
- global developmental delay
- attendance at a special needs school accompanied by the parent verbally confirming that the child had LD.
Matching children and young people without learning disability
Children without LD were screened using four matching criteria: (1) age (5–7 years, 8–11 years or 12–15 years and 364 days), (2) expected length of stay (short, 1–2 nights; medium, 3–7 nights; long, ≥ 8 nights) and (3) reason for admission (surgical, medical or investigations/tests).
Recruitment and consent
Families
There were three pathways to recruiting and consenting children and their families (Figure 3). All children were invited to participate irrespective of their abilities. The researchers ensured that all participants understood what was being asked of them and that they could have a break or withdraw from the study at any time. Four versions of patient information leaflets were available so that children of different ages and abilities could be included: two were symbol based (see Report Supplementary Material 5) and two were word based (see Report Supplementary Material 6). Children who took part provided verbal or written assent, and their agreement was confirmed just prior to data collection. When possible, the consent/assent process took place in the family home to enable researchers to build rapport with families, ascertain the child’s abilities and interests to tailor data collection activities and ensure that the study questions were relevant and sensitive.
FIGURE 3 Pathway to recruitment: families. PI, principal investigator. (Figure not reproduced here; see the original article.)
Hospital staff
During the parent interview, participants were asked to identify hospital staff involved in the care of their child whom they were happy to be invited for interview. These staff names were given to the research nurse, who established the willingness of these staff members to participate. The research nurse also identified and approached additional staff, and provided them with an information sheet and a consent form. With permission, the contact details of staff interested in participating were given to the research team, who contacted them to answer any questions and arrange a mutually convenient time for data collection.
Community staff
Each local collaborator distributed the online survey by e-mail. The front page of the survey provided a synopsis of the study, guidance for completing the survey and study contact details. Submission of the survey was taken as consent to participate.
Methods of data collection with parents
Parents were invited to share their experiences through four data collection methods: hospital diary, photographs, safety review form and an interview following the child’s discharge. The diary offered flexibility in how parents shared their experiences and could be completed at any time of the day/night. Completing parent interviews post discharge enabled researchers to prioritise data collection with children during the hospital admission.
Hospital diary
Paper diaries and a pen were provided in an envelope, and parents were asked to leave these on the ward prior to discharge so that the research nurse could collect them. Guidance on completing the diary was provided, including prompts about who they had interacted with, what information they had been given, decisions that had been made and how they felt about what had happened. Space was provided for further reflections.
Safety review form
The SHINE tool73 was adapted for parents to document positive and negative aspects of their child’s care in relation to six areas of safety: medication, communication and information, equipment, unexpected complications of care, hygiene/cleanliness and ‘other’ (see Report Supplementary Material 7). Parents could also indicate whether they had seen something that was unsafe, and whether or not this had been dealt with. The form was completed just prior to the child’s discharge, or, if it was not, the researcher completed the form with parents during their home interview.
Photography
Parents were asked to capture images of three things that they thought worked well and three things that could be improved about their hospital experience using a camera provided.
Semistructured interview
Interviews were planned for as soon after the child’s discharge as possible. Flexibility was offered in terms of location and timing and also the format (either face to face or over the telephone). Interviews focused on the child’s recent hospital admission, and how the experience differed or not from previous admissions to the same or different hospitals. The semistructured interview guide (see Report Supplementary Material 8) included questions about the child’s needs and whether or not these were identified and met, and the role parents played. Interviews were audio-recorded and transcribed verbatim. All families were given a short study evaluation form to complete (see Report Supplementary Material 9).
Methods of data collection with children
A multimodal approach was used based on the premise that children are experts about their own lives and should be enabled to share their experience in accordance with their abilities and preferences. Researchers spoke to parents on the telephone in advance of data collection to ascertain how the data collection activities needed to be tailored for the child. Three activities were available to elicit data: Modified Talking MatsTM (Talking Mats Centre, Stirling, UK),74 a sticker exercise and a hospital tour. For children unable to participate themselves, parents were invited to participate as a proxy, providing answers from the perspective of their child. Data collection primarily took place at the bedside. With permission, data collection sessions were recorded and transcribed verbatim. A retrospective review of children’s medical notes was also undertaken.
Modified Talking Mats
Talking Mats74 is a communication symbol tool designed by speech and language therapists that uses picture symbols to assist people with a range of communication difficulties in expressing themselves. It has been used with people with LD. A professional artist was commissioned to produce the additional symbols needed for data collection (see Appendix 3), which were then checked with children with and children without LD to ensure that each illustration was a clear and accurate representation of the person/place/object/concept we were wanting to portray. Talking Mats has a ‘top line’ showing a character with a thumbs-up, a thumbs-down or a shrug of the shoulders, which were used in the study to denote ‘like’, ‘don’t like’ and ‘not sure’. For children who required concrete ‘like’ or ‘dislike’, the middle option of ‘not sure’ was removed. Children were presented with a range of symbol cards relevant to their inpatient experience, and they were asked, depending on their ability, to indicate or to physically place these under the appropriate top line card. Depending on the ability, engagement and concentration of the child and their specific hospital experience, the number and type (abstract vs. concrete) of symbol cards varied. Children were encouraged to elaborate, where possible, their feelings about the placement of each card. Children could indicate their preferences verbally, with eye gaze, using communication software or with their parents as communication partners.
Sticker exercise
Children were given a short survey about their interactions with hospital staff and a set of red (frown) and green (smiley) face stickers to indicate a yes or no response to each question. Three versions of the questionnaire were produced: a shorter and longer version using symbols and words (see Report Supplementary Material 10) and a word-only version (see Report Supplementary Material 11), with the addition of frequency responses to indicate how often something did or did not happen. This tended to be used with children who had higher cognition. The researcher read each question aloud, where necessary, and supported those with physical impairments to place their sticker.
Hospital tour with photography
To help children think about their hospital experience, we invited them to take the researcher on a hospital ‘tour’, identifying areas they had accessed during their admission and taking photographs using a digital camera. In practice, tours led by the child focused on the ward or bed space. If a child could not physically take a photograph, their parent or the researcher would take it on their behalf, from the child’s perspective. With the agreement of the child, the ‘tour’ was audio-recorded to capture data relating to what was being photographed. Children were free to take photographs of anything they chose, with the exception of other patients, visitors or identifiable information. Photographs of a member of hospital staff required the approval and written consent of that person. Immediately following the tour, the photographs were printed or viewed on the researcher’s laptop, depending on the child’s preference, and discussed. All children (or their parents) were offered copies of the photographs to keep.
Retrospective mapping of hospital appointments
A retrospective mapping of all inpatient stays and outpatient appointments for the preceding 2 years at the participating hospital was conducted using paper and electronic hospital records for each child and the electronic appointment system. Data from inpatient (e.g. age, diagnosis, treating team, date of discharge, hospital passport, reasonable adjustments) and outpatient (e.g. date and time of appointment, if they did not attend) records were collated in a Microsoft Excel® (Microsoft Corporation, Redmond, WA, USA) spreadsheet.
Methods of data collection with staff
Hospital staff interviews
Interviews were conducted over the telephone or in person in a quiet room in the hospital, depending on the participant’s preference, using a semistructured interview guide (see Report Supplementary Material 12). Interviews lasted 30–60 minutes and were audio-recorded and transcribed verbatim.
Community staff survey
The phase 1 staff survey was modified, with questions added relating to access to secondary and tertiary care for children with and children without LD (see Report Supplementary Material 13). This was administered as per the phase 1 survey and stayed open for 3–5 weeks.
Phase 3 research methods
The aim of phase 3 was to compare the levels of satisfaction with hospital care between (1) children with and children without LD, and (2) parents of children with and parents of children without LD.
Recruitment and sampling
All seven phase 2 hospitals participated in phase 3, with data collection taking place on up to four wards identified by the principal investigator. All children and their parents who were admitted during the data collection period were eligible to participate. To facilitate the distribution and collection of surveys, no exclusion criteria were applied, and parents were asked to indicate if their child had a long-term condition, LD, neither or both. Participants were advised that returning a completed survey was taken as their consent to participate.
Data collection
An ‘easy-read’ survey (see Report Supplementary Material 14), based on a patient-reported experience measure developed by children and young people for children and young people,75 was developed for all children, irrespective of their age or perceived ability. An artist was commissioned to develop images to sit alongside each question, with a corresponding ‘thumbs-up’ or ‘thumbs-down’ and ‘smiley’ or ‘sad face’ tick box that children could use to indicate their response. The survey, available in English only, was piloted with inpatients at Great Ormond Street Hospital (GOSH) school, with and without LD, and minor revisions were made. Children could complete the survey independently or assisted by their parent, or the parent could complete it on the child’s behalf if the child was unable to. The parent survey (see Report Supplementary Material 15) contained a range of questions, the majority using a five-point Likert scale response. Non-English versions were available in seven languages. Wards were supplied with postboxes for completed surveys, which were collected by the research nurse and couriered to the research team.
Phase 2 and 3 data analysis
Qualitative and quantitative data were analysed in each phase before being merged and connected using data synthesis; congruence and incongruence between data sets was sought. Each data set was analysed by at least two members of the research team. Barriers to and facilitators of high-quality hospital care were identified for each data set, and these were then brought together, allowing a comparison of the factors identified by staff, parents and children. Specific examples of successful and effective measures that promote equal access were also identified. The analytical framework was compared with our initial theoretical framework in order to generate a final empirical framework of factors that affect the promoting of equal access to high-quality hospital care for children with and children without LD and their families.
Qualitative data analysis
The approach to analysis was relevant to each method of data collection; this is detailed in Table 2.
| Phase | Data | Method of analysis | Summary of analysis |
|---|---|---|---|
| 1 | Staff interviews | Framework analysis52,53 | Staff interviews were entered into NVivo 11; analysis was within and between data, applying the stages of framework analysis76 |
| 2 | Hospital documents | Summative content analysis77 | Predefined terminology (LD, special needs, intellectual disability, developmental delay, reasonable adjustments) was used to ascertain references to children with LD, creating a thematic framework based on content. The first set of documents was examined in detail and a coding frame was developed for use with subsequent documents |
| 2 | Parent interviews | Thematic analysis78 | Parent interviews were entered into NVivo 11 and read by at least two members of the research team; schematic maps were produced, checked and discussed further at team analysis meetings |
| 2 | Parent diaries | Thematic analysis78 | A subset of 20 diaries was read independently by two members of the research team, who then met to discuss the content and broad themes identified. An initial framework was developed and used when coding the diary content in NVivo 11, broadly identifying positive and negative experiences |
| 2 | Safety review form | Conventional content analysis77 | Data were extracted from the forms and summarised in a Microsoft Excel spreadsheet using the six domains. Cells were colour coded, and highlighted by one researcher as green indicating a positive issue, red indicating a negative issue and orange indicating in between. The spreadsheet was checked by a second researcher and minor revisions were made. Content analysis aided the counting and reporting of themes |
| 2 | Children’s interviews | Counting to produce visual maps Thematic analysis78 | Completed Talking Mats interviews and sticker charts were analysed for the frequency of responses, VILD charts produced to present data, interview data were analysed thematically |
Quantitative data analysis
Staff survey data (phases 1 and 2)
Descriptive statistics were used to characterise the sample. Composite variables were computed to represent capability, capacity, confidence, safety and values separately for questions related to LD and no LD (see Appendix 4, Table 15). All composite variables had acceptable internal reliability, with Cronbach’s alpha values of > 0.7. Composite variables and individual questions about involvement in service delivery and planning services, safety, values and meeting needs were analysed using Wilcoxon signed-rank tests, comparing responses about children with LD with those about children without LD for the total sample and, for hospital staff (phase 1), separately for respondents from children’s and non-children’s hospitals. Responses from community staff were also compared about admission and discharge (phase 2) for children with and children without LD using Wilcoxon signed-rank tests. Differences between children’s and non-children’s hospitals or between hospitals with and hospitals without LD nurse provision were compared using Mann–Whitney tests.
Parent and child satisfaction data (phase 3)
For the purposes of this report we focused on responses from parents and/or children with a long-term condition, with and without LD. Descriptive statistics were used to characterise the sample. Composite variables representing key domains related to satisfaction with care to meet their child’s needs, staff interaction and communication (parent survey) and environment, people, and care and treatment (child survey) were calculated (see Chapter 11). All composite variables on the parent survey had satisfactory internal reliability (Cronbach’s alpha > 0.8); all Cronbach’s alpha values for the composite variables in the children’s survey were > 0.6. Responses were compared separately for children with and children without LD and parents of children with and children without LD using Mann–Whitney, chi-squared or Fisher’s exact tests, as appropriate. For all surveys in each phase, a Bonferroni correction for multicomparisons was made, resulting in an alpha level of 0.005. All data were analysed using IBM SPSS Statistics version 22 (IBM Corporation, Armonk, NY, USA).
Phase 4
Findings from phases 1–3 were synthesised and shared with hospital staff at the phase 2 hospitals, and an open dissemination event was held at GOSH on 18 June 2019. The aim of this was to generate discussion to inform the development of a LD training package for staff. The views of families were also sought through discussion with individual patients on the ward and consultation with members of the Parent Advisory Group (see Parent Advisory Group).
Patient and public involvement
Aim
Patient and public involvement has been central to our work from the outset, with our primary aim being to ensure that the study was carried out in an appropriate, accessible and sensitive manner.
Preparation of proposal for grant funding
This study evolved from four pieces of work involving the chief investigator (KO) and/or other members of the research team (FG, JW, ITW). Key to each of these was our commitment to hearing what matters most to participants. The chief investigator consulted widely with parents, staff and LD experts about the RQs, design, data collection methods and dissemination to ensure that these were acceptable, feasible and of importance. One parent said: ‘Very important – LD are overlooked. My local trust only appointed a LD nurse advisor following a death, which was far too late’. Our parent co-applicant (SK) provided valuable insights and expert PPI throughout the study.
Involvement since grant funding
Parent Advisory Group
A group of 8–10 parents of children with and children without LD who had used hospital services was established to advise on all phases of the study. The Parent Advisory Group met twice per year during the study, with contact between meetings taking place by e-mail for specific requests and study updates. Representatives of the Parent Advisory Group attended three of the four of SSC meetings and fed back to other Parent Advisory Group members.
Phase 1
The Parent Advisory Group gave feedback on the preliminary results from the interviews in a sense-checking exercise to ascertain if the findings resonated with their experiences; this resulted in changes to the phase 2 parent interview topic guide.
Phase 2
The Parent Advisory Group gave feedback on data collection with children, parents and hospital staff. This was either during face-to-face meetings (Table 3) or at the Study Steering Committee meeting, where the group was involved in the ranking exercise for the selection of phase 2 sites and the challenges associated with phase 2. Additional feedback was elicited between meetings by e-mail on aspects such as the layout and instructions of the parent diary, the appropriateness of the parent safety review tool, and the content and layout of the information sheets and consent forms.
| Area of feedback | PPI | |
|---|---|---|
| Staff | Hospital staff recruitment | Recruitment procedure |
| Community staff survey | Suggested content | |
| Parent | Recruitment | Acceptability of procedure |
| Diary | Acceptability/feasibility of method and process | |
| Safety review form | Acceptability/feasibility of method and process | |
| Interview process | Acceptability/feasibility of method and process | |
| Interview topic guide | Suggested content | |
| Children | Use of cameras | Acceptability/feasibility of activity |
| Talking Mats | Acceptability/feasibility of activity | |
| Arts and crafts activities | Acceptability/feasibility of activity |
Phase 3
Feedback was provided on the style, format and administration of the survey to parents, as well as the clarity of questions and completion time. Parents’ knowledge of being in hospital, receiving care and navigating the system was hugely informative, resulting in timely changes to the survey. For the children’s survey, revisions were made to the wording of some questions, ideas were provided for pictures to accompany questions and some questions were moved to the parent survey.
Study Partnership Schools
Two local special educational needs schools, Richard Cloudesley School and The Bridge School, agreed to be Study Partnership Schools. One member of staff was pleased that the students had been asked to help, as it was felt that they were often overlooked for similar opportunities.
Discussions about the methods for collecting data from children were held, and school staff chose to trial the activities that they felt would benefit their students the most. In one school, a class trialled the photograph elicitation activity by taking photographs of their school and documenting what they liked and disliked. These photographs were subsequently used to produce an ‘about my school’ book to help familiarise new students with the school. In the other school, students trialled Talking Mats using photographs and picture symbols to share what they liked and disliked about their curriculum and school environment. A summary of each child’s participation was produced for the class teachers. Both schools were offered a presentation of the phase 1 study results. The opportunity to trial the methods was invaluable for assessing feasibility, strategies for delivery and the extent of student interest and engagement. Staff also gave helpful advice on elicitation techniques used in classroom settings for children with LD.
Students in the GOSH school provided feedback on proposed designs for picture–symbol cards for use with Talking Mats and the children’s survey. Students were asked to describe each picture and what changes could be made to improve clarity. They indicated their preferences for thumbs-up/thumbs-down, smiley faces/sad faces or ticks/crosses for the response options.
Young People’s Advisory Group and Young People’s Forum
Great Ormond Street Hospital runs two groups to provide researchers and clinicians with feedback on research and clinical services. The GOSH Young People’s Advisory Group, comprising 8- to 19-year-olds, is part of a national network of Young People’s Advisory Groups that meets regularly with the remit to be involved in the design and delivery of clinical research to ensure that it is relevant to children, young people and families. The Young People’s Forum, comprising children and young people aged 10–21 years, aims to improve the experiences of teenage patients. Both groups gave feedback on the proposed phase 2 research methods for children, including trialling aspects of the ‘hospital tour’ photograph elicitation activity by taking photographs of public access areas at GOSH and discussing these based on their experience of their own hospital admission. This process informed the practical set-up and delivery of this method of data collection.
We successfully integrated PPI into our research, underpinned by the research team’s belief that PPI is a worthwhile pursuit. Our Parent Advisory Group was pivotal; the members have provided input throughout, and we would highly recommend having a dedicated PPI co-ordinator, supported by other members of the research team. We extend our thanks to all of our PPI partners who have contributed to our study.