To flag or not to flag: Identification of children and young people with learning disabilities in English hospitals

Abstract Background Children and young people with learning disabilities experience poor health outcomes and lengthier hospital admissions than those without learning disabilities. No consistently applied, systematic approach exists across the NHS to identify and record this population. This paper describes practices in English hospitals to identify children and young people with learning disabilities. Method Interviews: 65 NHS staff. Questionnaire: 2,261 NHS staff. Conducted across 24 NHS hospitals in England. Results No standardized approach exists to identify children or young people with a learning disability or for this information to be consistently recorded, communicated to relevant parties within a hospital, Trust or across NHS services. Staff reported a reliance on parents to inform them about their child's needs but concerns about “flagging” patients might be a significant barrier. Discussion Without an integrated systematic way across the NHS to identify children with learning disabilities, their individual needs will not be identified.


| INTRODUC TI ON
Following on from the "Six Lives" report (Parliamentary & Health Service Ombudsman & Local Government Ombudsman, 2009), the current policy drive around the care of patients with learning disabilities (LD) (known internationally as intellectual disabilities) reinforces the view that to be in a position to meet needs and to prevent premature death, needs must first be identified. The UK's Confidential Inquiry into Premature Death of People with Learning Disabilities (CIPOLD), for example, recommends the "clear identification of people with LD on the National Health Service central registration system and in all health-care record systems" (Heslop et al., 2014: 894). Identification, the process of "flagging," to mark for attention or treatment in a specified way, has since been incorporated into the Care Quality Commission (CQC) best practice guidelines for children and young people with LD in hospital and in the recent LD improvement standards for NHS Trusts. Here, it states that "Trusts must have mechanisms to identify and flag patients with learning disabilities, autism or both from the point of admission through to discharge; and where appropriate, share this information as people move through departments and between services" (NHS Improvement, 2018: 6). There is limited guidance or evidence available as to what mechanisms of identification might be most effective or how they should be used in practice, ing that flags are "clearly visible on the patient's records" (Glasper, 2017: 65). Fundamental to the process is that the flag is accompanied by "a statement of the reasonable adjustments required" (CIPOLD) such that care can be adapted (Sheehan et al., 2016) and monitored for adherence with equality legislation (Tuffrey-Wijne et al., 2013).
There is little evidence to support what actually happens in practice. The identification of LD may not rest solely on a diagnosis as this does not necessarily indicate a need and the diagnostic category is not homogeneous. For some children and young people, a formal diagnosis of LD may never be made despite a continuing acknowledgement of their global development delay as they progress towards adulthood. Furthermore, where LD are known to exist, a better healthcare offer may arise, but there is limited evidence of that automatically leading to care that responds to or takes into account individuals' needs. Evidence suggests that staff rely on parents to supervise, protect, advocate and look after their child's medical needs and behaviours whilst in hospital (Mimmo, Harrison, & Hinchcliff, 2018).
There is clearly an important difference between identifying that a child or young person has LD and recording what reasonable adjustments are required, and ensuring those reasonable adjustments are consistently delivered during their hospital admission in a timely way (Turner & Robinson, 2011). One study of adults with LD has found a lack of staff knowledge, expertise, willingness to identify and flag LD, and a reluctance to routinely record and "label" people exists across junior and senior staff (Tuffrey-Wijne et al., 2013;Tuffrey-Wijne & Hollins, 2014), and this reluctance can also be seen in other sectors such as education (Ho, 2004). Evidence is also lacking around staff education or training needs and their confidence and capability to meet the needs of this particular population.
"Pay More Attention" is a NIHR funded mixed methods study aiming to identify the factors that facilitate and prevent children and young people with long-term conditions with and without LD from receiving equal access to high-quality hospital care and services. This paper reports on the practices of a sample of English hospitals who employ (or not) a process to identify this population with LD. The wider context and overall findings are reported in Oulton et al. (2018).

| ME THOD
Health Research Authority approval was given for Phase 1 of the study (IRAS: 193932), the results of which are reported in this paper.
Full ethical approval for this study was obtained from London-Stanmore Research Ethics Committee (16/LO/0645). Staff who took part in interviews were provided with a participant information sheet and consent form and provided verbal consent prior to taking part in the interview. Survey participants were informed that their completion and return of the anonymized survey would be taken as their consent to participate.
Semi-structured interviews were conducted with at least two senior managers per hospital with responsibility for the organization or management of care for patients with learning disability. These were conducted by CK or JR. Questions addressed the identification of children with LD and for all children: (a) equality of access to appointments, investigations and treatments; (b) the involvement of families as active partners; (c) satisfaction with care; and (d) safety concerns. Specifically, the interview explored what policies, procedures and systems were in place for identifying and/or flagging children within their Trust. The interview was piloted with two senior NHS managers.
An online questionnaire (with paper copies available) was developed for all clinical and non-clinical staff with contact with this patient group to elicit perceptions of their ability to identify the needs of those with and without LD and their families and provide highquality care to effectively meet these needs. This was piloted with seven NHS staff at non-participating sites with revisions made to improve clarity. Five questions related to the identification of children and young people with LD.

| Setting
Twenty-four NHS hospitals in England including specialist children's hospitals (n = 15) participated. The nine non-specialist hospitals all had at least one paediatric inpatient ward as well as paediatric outpatient clinics and services. A local collaborator managed the study at each site.

| Qualitative
Audio-recorded interviews were transcribed verbatim, anonymized and uploaded to NVivo 11 to utilize the framework option. Framework analysis, developed by Ritchie and Spencer (1994), was used to analyse the data. It is well recognized as an appropriate approach for applied research and focuses on describing and interpreting experiences in specific settings. It offers a systematic approach, which fitted the research and data collected (Ritchie & Spencer, 1994).
The five stages of framework analysis were followed: familiarization with the data; identifying a thematic framework; indexing; charting and mapping; and interpretation (Ritchie & Lewis, 2003;Ritchie & Spencer, 1994). CK and JR charted, mapped and interpreted the data individually using Parkinson, Eatough, Holmes, Stapley, and Midgley (2015) as an example. The final review was undertaken by KO, JW and FG.

| Quantitative
Descriptive statistics were used to characterize the sample.
Responses from participants at specialist children's hospitals were compared with those at non-children's hospitals using Mann-Whitney tests in terms of (a) the perceived usefulness of identifying this population at an organizational level; (b) the usefulness of visibly flagging; (c) whether or not their organization provided information to identify these young people; and (d) their own confidence in identifying whether or not a patient in their care had LD.

| Participants
The interview participants covered a range of senior staff within the participating hospitals with responsibility for people with learning disabilities. The present authors set out to conduct two interviews per site giving a minimum sample size of 48. This was to ensure all the interview questions were answered. The present authors continued interviewing staff until questions were answered with local collaborators asked to identify and invite additional participants. The final sample size was 65. Interviews lasted 30-45 min ( Table 1).
The quantitative and qualitative findings are presented using headings that provide a very broad linear narrative from identifying children with LD, what is done with that information, staff communicating with CYP with LD through to their own capacities and capabilities. This structure comes from the mapping and interpretation stage of framework analysis where patterns in the data were identified and offer a way to understand the current situation across the participating hospitals.

| Identifying children with learning disabilities
Interview data revealed that 10 of the 24 hospitals in our sample had a process in place for identifying children and young people with LD and this was more prevalent in specialist children's hospitals (n = 8, 53%) than non-specialist hospitals (n = 2, 22%). Staff referred to the use of "flagging" and "alerts," although there appeared to be no clear distinction in the data between these two terms. Furthermore, no common or consistent formal or informal approach appeared to exist to identify this population with various practices being described including via a general practitioner (GP) referral letter or from a school, the use of hospital records; via pre-assessment clinics, another hospital service or department; or via parents during the nursing assessment upon admission or in outpatients. It was evident that if a diagnosis of LD was made in a community setting, this information may or may not be transferred to the hospital. In four sites, staff felt that parents may not support the identification and subsequent labelling of children with LD because it felt simply wrong to "label" or there was a fear of getting it wrong even in Trusts where adults with LD are flagged.
Interviewees tended to emphasize children as "individuals" and that all children should be treated "the same." The implication was that if those with LD were flagged, it would mean these children were not being treated the same and this explained the hesitance or resistance to flag: [W]e're a Trust with a specific Children's Hospital I guess regardless of whether they've got learning disabilities or not, you know, you could say a child who has, I don't know, chronic orthopaedic issues around brittle bones is catered for differently than a child who attends regularly with their diabetes, do you know what I mean? It's a very personalised approach we have here, I like to think as well.
(Specialist Children's, Consultant) In some hospitals, parental agreement was sought for their child with LD to be flagged with the benefits of flagging outlined to them.  Survey responses indicated that most staff saw the benefit of having a process in place, with 81% recognizing the usefulness of identifying children with LD at an organizational level and 72% recognizing the usefulness for such patients to be visibly "flagged" (72%) ( Table 2). There were no differences between the specialist and non-specialist hospitals in relation to such views. Only half of all survey respondents reported being given information about how to define LD, with an additional 14% not knowing whether they had information or not. The distribution of responses from specialist and non-specialist hospitals was similar.
The present authors asked survey respondents about the systems in place at their Trust to identify and record that a CYP has a LD. Most frequently reported were medical notes (56%) or nursing notes (42%), followed by electronic documentation (27%). Seven per cent reported that no system was in place and 25% did not know what systems were in place at their Trust. Very few respondents referred to the use of databases (8%) or a sticker on the patient's notes (5%).

| What happens when children with LD are identified?
When a child is known to have LD, the admission type, planned or then puts a flag on the system, so anytime anybody accesses that patient on the system, it will come up with a flag, they can go in and see what their learning disability is.
(Specialist Children's, Matron) What we do with our alerts, is we've got something called [name of specific system] and our alert is linked to our email addresses and we all have an iPhone, so whenever that patient comes in, it's a live system, after two minutes of being clerked into the hospital, we'll get an alert on our phone to say the patient's name and where they are.
(Non-Specialist, LD nurse for Adults) These pathways may be triggered through a flag or an alert already on the patient's records, identified during a pre-admission appointment or recorded on their admission paperwork. Accessing specific information about a patient could be difficult; for example, staff may be able to see a child has LD but not all staff can access a child's record to ascertain more detail: If people have got a learning disability and it's iden- In contrast to formal routes, informal routes existed alongside or as standalone practices. Transfer of information between staff about children varied from a member of staff "knowing" the patient from a previous admission, verbal handover "huddles" during or when shifts changed, through to recording LD information in nursing and/or medical records.
Nearly half (48%) of all staff strongly agreed or agreed to being routinely informed that a child or young person in their care has LD (Table 2), with nearly a quarter (24%) of staff disagreeing or strongly disagreeing with that statement. With reference to particular staff groups, ancillary staff reported being routinely informed of patients with LD less than clinical staff (X 2 = 61.77; p < 0.001). Amongst clinical staff, nurses and healthcare assistants, but not doctors, were significantly more likely to report being routinely informed of a child/young person's LD than allied health professionals (Z = 2.74; p < 0.006).

| Communicating about children with learning disabilities
It was acknowledged in the interviews that good communication between all parties was facilitative in identifying, understanding and meeting a young person's needs from pre-admission to discharge. Additionally, issues associated with the degree of severity of a child's LD, with those at the milder end of the spectrum being potentially harder for staff to identify.
Unless it's overt or the parents actually formally identify, there may be milder learning difficulties that get missed because they're not specifically stated to the staff during the admission process, or pre-admission.
(Specialist Children's, Senior Nurse Manager) It was noted that parents often had to repeat the same information to staff, and at times they may become frustrated with having to repeat information readily available in their child's hospital notes or hospital passport.
So, obviously, it's a real irritation for parents to have to go through the same information all the time, you know, at every contact. So we do have a system whereby, for those with particular issues, and they tend to be more medical issues, they can have an alert put on the system.

| Staff capacities and capabilities
Overall 74% of respondents felt extremely confident or confident to identify a child or young person in their care or who they met has LD (Table 2 Q. E). Respondents from specialist children's hospitals reported feeling more confident to identify that a patient in their care had LD compared with respondents from non-specialist hospitals (Z = 3.03; p = 0.002). However, in both specialist and non-specialist hospitals, staff were more confident when their Trust gave them information about how to define LD (specialist hospital: Z = 7.40; p < 0.001; non-specialist hospital: Z = 4.36; p < 0.001).
The data appear to show a correlation between Trusts that flag LD and staff views related to reasonable adjustments and safety. For example, staff from hospitals who flagged felt more able to identify what reasonable adjustments are needed for CYP with a long-term condition and learning disabilities than those from hospitals who did not flag (Z = 4.37; p < 0.001), as well as feeling more confident that any reasonable adjustments would be accommodated in a timely way (Z = 2.62; p = 0.009).
Furthermore, staff from hospitals who flagged were more likely to report working in an environment that was deemed safe for meeting the needs of children and young people with a long-term condition and learning disabilities than those from hospitals who did not flag (Z = 2.30; p = 0.021). The former were also more likely to report that they were always able to deliver safe care than those from hospitals who did not flag (Z = 2.68; p = 0.007) and felt more confident to safely manage challenging behaviour (Z = 3.07; p = 0.002).
The NHS groups staff into Pay Bands for non-medical staff.

Examples of different roles by Bands include Band 5 for a Staff
Nurse and Band 6 for a Nurse Specialist through to Band 8 for a Modern matron. Survey data revealed that senior nursing and allied health staff (Bands 7 and 8) felt more able to identify children with LD (Z = −3.193; p = 0.001) but were not more likely to implement reasonable adjustments than their junior colleagues (Bands 1-4, 5 and 6) (Z = −1.707; p = 0.088). This finding was supported only in part by interview data, with interviewees suggesting that senior and more experienced ward staff were more likely to identify children with LD and to implement reasonable adjustments to meet their needs. Interviewees went on to describe a number of factors thought to impact whether CYP with LD are identified by staff and have their needs meet, from their initial and ongoing training, experience gained over time, having sufficient time with the patient and their family and access to appropriate resources.
There was a general expectation by the interviewees that staff involved in a child's care would ask about any LD and/or would have read a patient's notes. However, it was recognized that staff may not have a sufficient level of knowledge about LD, which could mean this was omitted during admission; this was despite interviewees recognizing the value of specific training around caring for young people with LD to raise and maintain awareness and for training to be put into practice.

| D ISCUSS I ON
Reported here are a sample of strategies NHS English hospitals employ (or not) to identify children and young people with LD accessing their services, as well as staff views regarding the process. Prior to the introduction of the LD improvement standards for NHS Trusts (NHS Improvement, 2018), the present authors found that less than half of hospitals in our study had any process in place for flagging these patients and for those who did, there was no standardized way for this to happen or for this information to always be recorded, or for what is recorded to be confirmed as accurate, communicated to relevant parties within a hospital/Trust and acted upon.
A range of practices existed from a passive approach of simply noting a diagnosis of LD in a child's hospital records, asking parents' permission for a flag to be applied through to active engagement with this information, including a formalized chain of events leading, ideally, to the implementation of reasonable adjustments to meet a child's needs. As a result, the patient experience between hospitals is likely to be very different. and accommodating those needs through the provision of reasonable adjustments.
Interviewees raised a number of issues about the complexities of identifying children and young people with LD, in relation to their parents. They felt that parents may not support a process of "labelling" their child, they may not agree with the diagnosis or they may need support themselves to articulate their views due to their own LD or language barrier. Such findings highlight the need for clear guidelines about how parents should be involved in the process of flagging children and young people with LD in hospital, particularly as some of the staff interviewed felt that children with "mild LD" can go unrecognized. In adult services it has been reported that "those without a formal diagnosis may remain invisible" (Sheehan et al., 2016: 6). Included in these guidelines must be information about the use of definitions around LD, as well as how to ask parents the relevant questions in a sensitive manner.
Less than half of staff who the present authors surveyed reported being routinely informed that a child in their care has LD, but at least three quarters of them reported seeing the benefit of formally identifying this population and feeling confident to do this.
Such findings suggest that the primary issue is the communication of information between staff rather than a lack of staff knowledge or willingness to identify the population, as found in adult services (

| Limitations
The data presented here provide the perspective of NHS hospital staff. It is important to gain the views of children, young people and their parents to understand how LD is correctly identified (or where identification is missed or an incorrect "label" applied) and the value of having LD identified and the perceived implications that this information will have on practice.
This wider study from which this data comes sought to map the provision of care for children with long-term conditions, particularly those with LD through a mixed methods approach. The NHS staff who participated in interviews were identified locally and comprised a wide variety of roles (e.g., medical consultants, matrons, clinical nurse specialist and managers) which may have produced a lack of consistency in the data. At two sites (non-specialist hospitals), no staff agreed to be interviewed. The 30-min interview was designed to encourage participation but limited in-depth discussion around site practices. Additionally, where a difference by "pay band" has been identified, the present authors cannot be certain that a lower band indicates a lack of experience in working with children and young people with LD.
Although the study did not aim to be proportionally representative of specialist children's hospitals and non-specialist hospitals, children are more likely to receive treatment in their local hospital, which for many, is unlikely to be a specialist hospital for children. If this study was repeated, a sample that better reflects this form of access might be incorporated.

| CON CLUS ION
There is no standardized way of identifying children and young people with LD and their individual needs across hospitals in the NHS in England. The identification of these children needs to be consistent within and across the NHS and for this identification to be the beginning of a standardized process whether through flagging or another form of alert. This is essential to not only inform health and social care professionals that a child has a LD but what reasonable adjustments they require as individuals and to ensure that these are put in place to provide them with the best possible care. This is dependent on healthcare professionals having a level of training, confidence, skills and knowledge to actively and positively respond to initially identifying LD and being able to put into place the adjustment required, drawing on the available resources from the individual to the institutional. A robust and comprehensive system that works across the whole NHS is required to identify this population and their needs and provide equity of access to resources to meet these needs.

ACK N OWLED G M ENTS
Our thanks go to the local collaborators at each of the participating sites for their assistance in Phase 1 and to all the NHS staff who gave their time to participate in an interview or complete the survey.