Child and Adolescent Mental Health Services: inpatient report

4. Results and Discussion

The results and discussion sections are combined in this report, to aid the flow for the reader in following the complex quantitative and qualitative data and issues. Response rates are presented and explored first, including possible reasons for variation between Health Boards. Demographic and clinical information is presented next. Then the report weaves a route through the need for a mental health inpatient admission, pathways into the admission, where the patient was actually admitted to (if at all), issues during an admission and pathways out of hospital. The impact of the current situation on children, young people, families and services is reported and discussed. The numbers and profiles are given of patients requiring specialist mental health inpatient care not presently available in Scotland. Finally, other findings from the study, such as information on community LD CAMH services, are noted.

More detail on the quantitative results are included in the appendices (section A4), which contain much rich qualitative data, illustrating the experiences of children and young people, their families and professionals. Graphs and tables contained in the appendices also present more detail on the quantitative results reported in the main part of the report.

4.1 Responses from clinicians

153 questionnaires were returned by 43 clinicians from 13 out of the 14 territorial Health Boards in Scotland, relating to 84 individual patients who were admitted to hospital on 1 or more occasions. This represented 106 overall periods of inpatient care, 32 transfers within admissions, 8 situations where patients requiring hospital admission stayed at home/usual place of residence and 7 where they went to a non-hospital placement. Steady numbers of admissions were reported on from 2012-2014, but less for 2010-2011. Interviews with 37 of the 43 clinicians from a range of professional backgrounds and from all Health Boards that submitted questionnaires took place between March and May 2015.

The high number of responses to this survey from senior clinicians from a variety of child and adult services across Scotland (see appendices, section A4.1.2 (i) for detail) pointed to the research addressing an important issue for their patients. Clinicians reported lack of access to inpatient services for children and young people with learning disability (and for some with autism spectrum disorder but no learning disability) as being a longstanding issue, well before the study period. Excepting for the limited number of LD CAMHS specialist psychiatrists, numbers affected on individual clinician's caseload were small. However, the impact of the lack of inpatient care on children and young people, their families and local services was considered highly significant. The vast majority of those submitting questionnaires also gave considerable time to telephone interviews which added depth and insight into the situations outlined in the questionnaires. They were keen to see services develop to better meet the need of this patient group and their families and contributed valuable opinions and ideas as to how this could be done.

It was clear that the considerable difficulties experienced by many of the children and families described by clinicians had made a big impression on clinicians themselves. For example, one Consultant CAMHS Psychiatrist commented that these are the kind of patients and situations that Psychiatrists wake up in the night thinking about, even years later. Others said that their patient's situation was the most difficult and stressful they had ever had to manage.

4.2 Possible under-reporting

4.2.1 Practical difficulties in identifying historical admissions

Despite the apparent high number of responses received, participating clinicians were concerned that the study would not uncover the full extent of need. Under-developed IT recording systems and a lack of management and strategic oversight of services for this patient group were common issues. Clinicians mostly had to identify patients for inclusion from memory and manually 'trawling' through diaries. Where there are no learning disability CAMHS or autism spectrum disorder teams or teams are small, movement of 1 or 2 clinicians away from services severely affected ability to identify cases. Even having 1 key clinician on leave at the time of the survey may have meant that important cases were not highlighted. It is therefore not surprising that greater numbers were included from the last 3 years of the time period covered by the survey compared with the first 2 years.

4.2.2 Children and young people with autism but without learning disability

Only a small minority of clinicians submitted questionnaires for all of their patients with autism spectrum disorder but no learning disability meeting study criteria. Most only reported on these where they considered that more specialist care was required (usually due to forensic concerns/need for security). This impression was backed up by additional statistical data provided by 2 of the Scottish CAMH inpatient units on the diagnoses of children and young people admitted over the study period. For example, the National Child Psychiatry Inpatient Unit had a total of 37 patients with diagnoses of learning disability and/or autism spectrum disorder over the 5 year period, and Skye House (West of Scotland YPU) 36, many more than picked up by the study. Less than 5 of these in each case had learning disability diagnoses and this is more consistent with study figures.

4.2.3 Difficulties in identifying mental health issues and need for admission

Clinicians get used to having an exceedingly high threshold for admitting patients when there is no suitable unit and try to avoid admission by trying to manage in the community. Most clinicians have not themselves been to or experienced the outcomes of specialist LD CAMH units so may not be aware of when a patient would benefit from admission. Where there are no specialist LD CAMH community services, mental health elements to behavioural presentations may not be recognized at all. Children and young people can therefore move up increasing levels of restrictive and secure care and school settings with their mental health needs unaddressed.

In this study Health Boards without an LD CAMH service tended to report relatively more admissions for those with autism spectrum disorder but no learning disability, who often did not require specialist inpatient services. They were less likely to report admissions/need for admission for those with learning disability. A lack of community LD CAMH expertise may mean they were not working with and identifying the need amongst those with learning disability, particularly of more severe levels.

4.2.4 Other evidence of under-reporting

A number of clinicians identified patients for the survey but were not able to gather information and submit forms in the timescale required. Forms were received too late for inclusion in the analysis for 7 admissions relating to 4 individual patients. These were all looked after and accommodated children, with mild or moderate learning disability, aged 14-15, one of whom also had autism spectrum disorder. They all had additional major mental illness, severe behavioural problems, significant background issues of deprivation and/or neglect and had been charged with serious crimes. 3 had been admitted to adult IPCUs (Intensive Psychiatric Care Units) or adult low secure wards. All ideally required low or medium security on LD CAMH or 'mainstream' secure adolescent wards and required 2:1 or 3:1 nursing care. Only 1 received such care in a hospital in England. They were all in hospital for lengthy periods, up to 6 years, and 3 remained in hospital, including in adult forensic wards at the end of the study period. This indicates that there may be significant cases missed from the study and this needs to be taken into account when estimating the level of need for specialist beds and the interface with the proposed forensic CAMH unit.

Information on admissions to pediatric wards for mental health/behavioural reasons was submitted from very few Health Boards; therefore this number is also likely to be an underestimate.

4.3 Responses from families and carers

17 questionnaires were received from families/carers, describing 18 admissions/ parts of admission for 10 children and young people. Despite relatively small numbers, demographic and diagnostic characteristics broadly matched those of the patients described by clinician questionnaires and the children and young people were admitted to a similar range of units and usually had long waits for admission. Themes from parent/carer interviews were strikingly similar to those from clinician interviews and these experiences were incorporated into the relevant qualitative results sections. Important additional perspectives were gained and are given in section A4.17 of the appendices.

Overall, staff understanding of children and young people's needs was reported as being variable, and with clear room for improvement. Some good outcomes were described, but also significant difficulties and negative impact on children and young people's emotional well-being. Distance from home was a major issue for families who had to travel significant distances to visit their child (up to 8 hours).

Recruitment from families was more limited compared to that from clinicians. Recruitment relied on clinicians passing on questionnaires to families and this may have been thought to be inappropriate in some cases. For example, where they had lost contact with the family and were not sure of their circumstances, or where the children and young people were looked after by the Local Authority. It was clear from the information provided by those who did complete and return questionnaires that many parents remained in the midst of circumstances relating to their child's mental health. A number had been deeply affected by their child's difficulties and their journey through services. Therefore participating in the study may have been too difficult emotionally or time-wise for a number of other families.

Good service is what is expected and by virtue of things working well people move on. Generally speaking, people give feedback because they are unhappy or particularly pleased with some aspect of a service. Not surprisingly therefore, approximately 90% of those who selected themselves to be interviewed wished to do so to report on their " negative and difficult experiences". However, when talking about mental health services the importance of emotion must be acknowledged as a particular factor in people's understanding and perception of their experiences. When considering parental experience of their child's health the impact of this factor is further amplified. Parental understanding of their role as protector, care-giver, provider and problem-solver is hugely compromised by what is happening to their child and by the need for the involvement of professionals and agencies (Beresford B, Rabiee N & Sloper P, 2007).

The emotional impact on families of caring for a child with disability should not be underestimated. Added to that, the onset and diagnosis of mental ill health can be, as one parent reported it, "both devastating and frightening, we were thrown into the unknown and remain there even now". All the parents interviewed were in varying states of 'rawness'. They reported the experiences they had as "traumatic", some were able to describe particular aspects of the process quite clearly but others were hazier in their descriptions and timelines. Most parents described a strong sense of "isolation and separation, including from other parents of kids with autism". All of this was likely to have impacted on their experience of services and their reporting of it.

In addition to the evidence gained directly from the survey questionnaires and interviews, Kindred has provided advocacy support to four families over the past two years with children and young people with a diagnosis of autism spectrum disorder who have had long term experiences of inpatient care. They have also supported a further six families of children and young people with very challenging behaviour. The perspective gained over time of the impact on families who may have benefitted from a specialist inpatient facility in Scotland is included in the appendices ( section A5).

4.4 Views of children and young people

Whilst incorporating the experiences and views of families and carers, it is important to acknowledge that this survey was not able to seek such information directly from children and young people themselves. In any development and design of the recommended unit, views and experiences of children and young people should be sought, supported by the use of appropriate communication techniques. For example, links with LD CAMH units in England could be used to seek views from children and young people who are inpatients there, with help from independent advocacy organisations.

4.5 Variation between Health Boards

Differences in overall submission rates to the study from different Health Boards (see section A4.1.2 (i) of appendices) were likely to be due to a range of factors, including participation rates of individual clinicians, focus or otherwise of local services on this issue, and presence or absence of specialist LD CAMH services. It also depended on whether clinicians chose to submit data on patients with autism spectrum disorder but no learning disability who did not require specialist autism spectrum disorder services. This was highly variable. The information in this section is based on submission rates relative to the population size of the Health Boards. Numbers submitted from each Health Board were not sufficient to report statistically significant analysis; however, some interesting trends emerged.

Health Boards with the most established LD CAMH community services generally had:

• Lowest overall submission rates to the study - which may be due in part to a focus on those more complex children and young people requiring specialist inpatient units not currently available.

• A higher total number of admissions for children and young people with learning disability to units within their Health Board, particularly for those with moderate and severe learning disability – may be due to their ability to support their patients within local adult LD, adult mental health or YPU wards.

There were 2 notable exceptions:

• In one Health Board with an LD CAMH service, there were fewer admissions for those with learning disability, but the psychiatrist supported a number of patients with moderate learning disability at home or in local residential special schools who should have been in hospital.
• The Health Board with the highest rate of admissions of those with learning disability was one with no LD CAMH psychiatrist, but which had a large residential special school in its area from which a number of children and young people were admitted to the adult LD unit.

The Health Board of residence ( i.e. origin) of patients requiring admission to hospital also appeared to vary in relation to the community LD CAMH services in that Health Board:

• Where there was no specialist community LD CAMH service, rates of reported admissions tended to be lowest - this may indicate hidden need, with community services not identifying and treating mental health problems in this group.
• Where services were partially developed, admission rates were highest - perhaps indicating that more problems were identified, but services were inadequate to treat children and young people with more severe/complex problems in the community.
• Where community services were most well developed, admission rates were in between - this may mean that mental health problems were being identified and that some admissions were prevented due to more comprehensive and, in one Health Board, intensive work being done in the community.

4.6 Demographics, diagnoses, presenting issues and admission characteristics

More detail can be found on these results in sections A4.2 to A4.4 of the appendices.

4.6.1 Ethnic group

>95% white British.

4.6.2 Gender

63% male. As the severity level of the learning disability increased, males were more likely to be over-represented: 55% were male where there was no learning disability, up to 80% male where there was severe/profound learning disability. This is consistent with higher rates of learning disability diagnosis in males in general, and increased % of males with increasingly severe levels of learning disability. This contrasts with admissions for children and young people under age 18 in general, as illustrated by 32 out of 50 patients in the 2014 census on people in mental health and learning disability inpatient beds being female (Scottish Government 2015).

4.6.3 Age

Range 8-17 years; mean 14.9, mode 16. Those with moderate, severe and profound learning disability were more likely to have admission at a younger age. The relatively younger age of admission for those with these more severe levels of learning disability may be due to the increased severity and complexity of their co-morbid conditions.

4.6.4 Looked after and accommodated children

30% of submissions were for looked after and accommodated children, with highest rates for those with moderate learning disability. High rates of submissions for looked after and accommodated children were likely to be due in part to a significant proportion being accommodated with family agreement in residential schools and care settings (particularly those with moderate learning disability and challenging behaviour). In addition, children with learning disability are known to be exposed to high rates of abuse (Taylor et al, 2014), and some children and young people may have therefore been looked after and accommodated for child protection reasons.

4.6.5 Level of learning disability and rates of autism spectrum disorder

The patients described in clinicians' questionnaires had a range of levels of learning disability. Due to inclusion criteria, all those without learning disability had autism spectrum disorder. 30% of patients had no learning disability, 30% had mild learning disability, 30% had moderate learning disability, and 10% had severe/profound learning disability. 75% had autism spectrum disorder, including 100% of those with no learning disability, 52% of those with mild learning disability, 73% of those with moderate learning disability and 90% of those with severe/profound learning disability.

4.6.6 Diagnoses

Multiple co-morbidities were common. Additional psychiatric diagnoses included:

• Affective disorder (24%)
• Anxiety and trauma related disorder (23%)
• Psychosis (21%)
• ADHD (15%)
• Other developmental disorders (6%)
• Other additional psychiatric diagnoses (13%)

50% had additional physical health diagnoses:

• 15% had multiple physical health diagnoses
• The number of additional physical health diagnoses increased with the severity of learning disability

The high rate of autism spectrum disorder in patients with learning disability, increasing with severity of learning disability and high rates of comorbid mental illness and physical health diagnoses, reflects clinical experience and extensive research evidence for this population.

4.6.7 Presenting behaviours

• 65% had self-injury (100% where severe/profound learning disability)
• 67% had aggression (100% where severe/profound learning disability)
• 51% had destructiveness (80% where severe/profound learning disability)
• 28% had sexualized behaviour (42% where mild learning disability)

4.6.8 Police involvement

41% had police involvement at some stage (52% where mild learning disability)

4.6.9 Admission rates over time

Steady admission numbers were reported from 2012 – 2014, but less for 2010 and 2011. This is likely to be due in part to issues described above in identifying patients for inclusion in the study. In addition, it is possible that there was: improved identification of mental health difficulties and the need for inpatient care over time; less capacity in some local children and young people and adult services to pull together local alternatives to hospital admission; an actual increase in numbers of children and young people with complex mental health and behavioural needs requiring admission.

4.6.10 Reasons for admission

Most common across all levels of learning disability were:

• Mental health assessment and stabilisation (58-67%)
• Risk management (excluding suicidality/deliberate self harm) 50% of those with no learning disability, 70-80% of those with all levels of learning disability.

Suicidality/deliberate self harm: 37% of those with no learning disability, very low rates in all levels of learning disability.

Medication management: 16% of those with no or mild learning disability, 29% for moderate learning disability and 35% for severe/profound learning disability.

Home/care placement breakdown was a common reason for admission only for those with moderate learning disability (23%).

4.6.11 Mental Health Act

53% of patients were noted to have been detained under the Mental Health Act for all or part of an admission.

4.6.12 Staff: patient ratio

Limited information given, but >15% needed more than 2 nurses per patient.

4.7 Where children and young people were admitted to

4.7.1 Admissions to hospital settings

The questionnaires returned by clinicians described episodes of care in a range of Scottish settings: 56 in adult mental health wards, 30 in young people's (12-18 years old) mental health inpatient units ( YPUs), 25 in adult learning disability (adult LD) wards, 7 in pediatric wards, and 5 in the National Child Psychiatry Inpatient Unit ( NCPIPU). 17 patients were admitted to wards in England during the study period: 9 to LD CAMH inpatient units (including secure LD CAMH inpatient units) and 8 to other units (including secure adolescent and ASD specific).

More detail on the profile of patients admitted to different types of ward is given in section A4.6.2 of the appendices. Patients with increasingly severe levels of learning disability were more likely to be admitted to adult LD wards, but only 2/3 of those even with severe/profound learning disability were admitted to a learning disability specific ward, with ½ of these being an age-appropriate facility.

The most common age for those admitted to adult MH, adult LD, YPU and LD CAMHS (including secure) wards was 16-17 years. Those remaining at home or admitted to non-hospital placements tended to be younger teenagers.

Rates of autism spectrum disorder varied according to the type of ward admitted to. Very high rates in those admitted to adult mental health units reflects that a significant number of patients in the survey with no learning disability (and therefore by definition due to inclusion criteria with autism spectrum disorder) were admitted to these wards. There were higher rates of autism spectrum disorder in those on adult LD wards compared with those on YPUs. This may have been due to the greater severity of learning disability in those admitted to adult LD wards, with high rates of comorbid learning disability and autism spectrum disorder in this group. YPUs seemed to be able to generally manage those with Mild LD and more straightforward mental illness presentations. The added complexity of autism spectrum disorder with learning disability may have made these units less likely to cope with/accept admissions.

4.7.2 Admissions to age-appropriate facilities

The percentages of children and young people in the survey admitted to age-appropriate mental health wards (including transfers within admission) were 27% for those with autism spectrum disorder but not learning disability, 36% for those with mild learning disability, 33% for those with moderate learning disability and 43% for those with severe/profound learning disability.

In the 2014 Mental Health and Learning Disability Inpatient Bed Census (Scottish Government 2015), 38 out of 50 patients aged less than 18 years were in a children or young people's unit (76%). Methodological differences do not allow direct comparison with admissions of the general population of children and young people in Scotland to age appropriate mental health inpatient wards, however these percentages do appear to indicate likely poorer access to age appropriate care for children and young people with learning disability and/or autism spectrum disorder compared to their peers. It should be cautioned that reporting of admissions for those without learning disability but with autism spectrum disorder in this survey was biased towards those admitted to adult wards due to many submissions coming from one Health Board who reported difficulty accessing beds in their regional YPU during the study period.

4.7.3 Admissions to non-hospital settings or remained at home

In addition to these admissions to hospital settings, there were 15 patients who required admission but remained at home or were admitted to non-hospital placements. These patients tended to be younger, have moderate learning disability, usually had major mental illness and all had highly challenging behaviour. The main reasons for not being admitted were lack of a suitable age-appropriate specialist ward at a manageable distance, and cross-border issues related to the Mental Health Act.

4.8 Pathways

This section covers routes into admissions (including waits), routes through admissions (including transfers), lengths of stay, separate periods of admission for the same patient, and routes out of admissions. There was a wide variety of complex pathways for patients ideally requiring specialist LD CAMH, adolescent mental health secure or ASD-specific secure inpatient provision, which only a minority actually received in specialist units in England. Section A4.14 of the appendices gives detailed information on all aspects of patient pathways, including examples of pathways followed by children and young people requiring inpatient care currently unavailable in Scotland ( section A4.14.4).

4.8.1 Waits for an admission

Patients often spent significant time making no or limited progress at home, or in residential care, or on one or more inappropriate wards (or some combination of all of these).

Finding a bed in hospital was usually difficult. Clinicians described extremely 'high thresholds' for admission, meaning they did not arrange admission for these children and young people until they were much more unwell than peers without learning disability and/or autism (or than adults with learning disability), due to lack of suitable facilities. Admissions were usually undertaken only in absolute crisis, when the risk of staying in the community exceeded the risk of an inappropriate ward. There was a lack of community mental health services to support the child and their family, with 'cobbled-together' arrangements and gaps being filled by other services. Despite this, once admission could no longer be avoided, children and young people often waited considerable periods of time even for an inappropriate bed, with 27% waiting more than 4 weeks. The majority were not admitted directly to appropriate wards, having to wait again for transfer (if that happened) to a specialist bed.

4.8.2 Length of admission

Admissions were lengthy (44% over 6 months), particularly for those with learning disability. Examples were given where patients needing specialist admissions were kept only for brief periods on non-specialist wards and then discharged home or to social care placements with inadequate mental health assessment/treatment. Whilst these admissions may have been reported in the survey, the length of them would have been less than that really required. Conversely, there were patients in hospital for much longer than ideally required because treatment took much longer in non-specialist units and because of 'delayed discharges' where there was a lack of suitable social care/education provision to discharge to. These factors make estimation of the likely average required length of stay on a specialist unit difficult. One of the English LD CAMH NHS inpatient units aims for a 3 month length of stay for those with severe learning disability. However, this varies considerably, those with forensic needs and those with highly complex or treatment-resistant mental illness usually requiring significantly longer. A reasonable overall estimated average length of stay would be between 6 and 12 months, with significant individual variation.

4.8.3 Transfers and multiple admissions

27% of patients had at least 1 transfer during their admission, most commonly those with mild learning disability. 16% had more than one separate admission. Numbers of transfers during an admission are likely to be an underestimate of the true picture. In some cases, information on questionnaires was unclear about timing of transfers so they were not counted. There were also a number of patients with learning disability known to have moved on to adult LD inpatient units after the age of 18. Only transfers before age 18 were counted in the data.

4.8.4 Discharge destination

The discharge destination at the end of the last admission in the study period varied according to the presence or absence of learning disability, and the level of severity. The discharge destination was home for 75% of those without learning disability, but 55% for those with more severe levels of learning disability. 18% (14 patients) remained in hospital, some due to lack of appropriate social care or education resources to move on to. This was particularly distressing for children and young people and their families, and impacted on the service provision available for other patients.

4.9 Impact and outcomes of the current situation on children and young people and families

There was evidence of willingness and efforts of practitioners in attempting to meet children's needs at a local level, in order to reduce the impact of long-distance separation from families. This involved creative responses and examples of cross-disciplinary working. However, the difficulties faced in Scotland by children and young people with learning disability (often also with autism spectrum disorder) who required mental health admission were considerable, as was the impact on their families and services who attempted to meet their needs. These were explored in detail with clinicians and families and fully documented in the appendices ( sections A4.6 to A4.13). Time and resources were wasted on admissions to inappropriate units, which could in some cases contribute to further deterioration.

4.9.1 Impacts of waiting for a bed

Children and young people endured long periods of inadequately treated illness and distress at home, potentially leading to poorer long-term prognosis. It was highly stressful for families supporting children and young people displaying severe self-injury, aggression and destructive behaviours at home. Families, exhausted from caring responsibilities, had to make difficult decisions. They were fearful of admission, especially when no appropriate unit was available and/or they had previous negative experiences. It was difficult to hand over care, especially to staff inexperienced in working with children and young people and/or those with learning disability

Risks were unacceptably high in the community from aggression, self-injury, and destructiveness, use of high levels of psychotropic medication without the ability to safely monitor and from families having to use unsafe physical restraint.

There were particular difficulties for vulnerable patients, including those with severe learning disability and more complex needs, and looked-after children and young people in out-of-area placements.

4.9.2 Outcomes from admissions to wards in Scotland

Some good outcomes from admissions to all types of ward were reported, despite the lack of suitable facilities and expertise for children and young people with learning disability. Examples were given of acute risks being managed, behaviour stabilised, medication adjusted, care needs identified and discharge or transfer to more specialist units supported. However, in general, children and young people with more than mild learning disability admitted to wards in Scotland faced significant difficulties, including lengthy admissions, sometimes in highly restrictive environments, with multiple transfers and transitions for those most in need of consistency. Lack of specialist age/developmentally-appropriate multidisciplinary care and environment could contribute to unnecessarily high use of medication and restraint.

Children and young people on local adult LD and mental health wards had better family contact and continuity with services, but the lack of staff expertise, age-appropriate physical environment, education and activities led in some cases to an escalation in challenging behaviour, with high use of medication and restraint. There could be a lack of progress over months or years. Children and young people exposed to adult patients were anxious or socially and sexually vulnerable. Some adult LD wards took in children and young people with highly challenging behaviour with successful outcomes, due to flexibility in adapting physical environment and staffing, but at significant cost to services to adult patients. Other children and young people were reported to have been discharged too quickly and inadequately treated due to concern about the adult setting.

Children and young people with autism spectrum disorder but no learning disability generally did well in the regional mental health YPUs, except for those children and young people requiring inpatient mental health care with a degree of security, including forensic. Although staff expertise of autism spectrum disorder reportedly appeared to be improving over recent years, it remained variable and parents had particular concerns in this area. However, when these children and young people were admitted to adult mental health wards, they (like those with learning disability) were disproportionately affected compared with typically developing peers. This was due to the child or young person's poorer communication and social skills, developmental level and need for an adapted environment and activities.

The YPUs usually managed children and young people with mild learning disability and typical mental illness presentations well. The success of admissions to YPUs for those with moderate learning disability and greater complexity varied according to staff experience and support from local LD CAMH professionals. Support was generally unavailable for patients from outside the Health Board in which the YPU was situated, leading potentially to unequal access and outcomes.

Young people with severe learning disability and/or highly aggressive or destructive behaviours were rarely accepted for admission to the YPUs. Staffs on the National Child Psychiatry Inpatient Unit were considered more experienced in working with children with greater neurodevelopmental complexity. However, even on the National Child Psychiatry Inpatient Unit, some under 12s with severe/profound learning disability were only manageable as day patients and ultimately required LD CAMH inpatient admissions in England.

Families were often deeply upset by the experience of their child being in hospital, describing a lack of emotional support to deal with diagnoses and the need for improved communication. They felt that a focus on just one aspect of a young person's difficulties created problems in meeting their needs and that better access to therapy, education and activities was required. Negative experiences impacted on future relationships and engagement of young people and families with adult services.

4.9.3 Outcomes when a child or young person was not admitted to hospital when required

Extremely high levels of input from community clinicians and multiagency partners were given for some individual children and young people who required hospital but were not able to access it. Despite this high level of input, children and young people endured unnecessarily prolonged periods of illness and distress and families were stressed and exhausted caring for them.

There were high risks to children and young people and families from self-injury, aggression and destructiveness, as well as from an inability of community services to adequately monitor mental state and medication. Support agencies and schools sometimes withdrew support due to an inability to manage the severity of behaviours. In the absence of appropriate mental health inpatient care, some children and young people's challenging behaviour escalated and they were moved through increasing levels of security in expensive social care or educational residential placements which could themselves be out of area and without access to specialist mental healthcare.

4.9.4 Outcomes from admissions to specialist units in England

Patients who were admitted to specialist LD CAMH and ASD specific wards in England generally received comprehensive, age-appropriate, multidisciplinary assessment and treatment in an appropriate physical environment with developmentally appropriate activities, education and peer-group. There were significant improvements in mental health, sleep and well-being for children and young people and their families. This demonstrated what was possible to achieve for some children and young people with highly complex problems, with recommendations and support plans continuing to helpfully inform local services several years after discharge.

Most difficulties described with these admissions related to distance from home, causing additional distress to children, young people and their families. Dislocation from family and local services complicated discharge planning. Accessing beds was complicated and involved prolonged negotiations with families, local and national services, long waits for beds, and complex cross-border Mental Health Act issues

4.10 Impact and outcomes of current situation on services, clinicians and their other patients:

4.10.1 Summary of financial costs

Children and young people admitted to specialist LD CAMH and ASD CAMH NHS units in England are funded via the NHS National Services Scotland risk share scheme on behalf of Scottish Health Boards. Costs were variable with a peak of costs at the end of the study period where total NSD spend on Forensic LD CAMH, LD CAMH and ASD CAMH specialist care cost £1.06 million in 2014/15.

Admissions to specialist private hospitals are funded directly by the patient's Health Board of residence, with costs varying from £330,000 to £624,000 per patient per annum. Where children and young people were nursed in Scottish mental health inpatient units (including adult LD, adult mental health and YPU wards), additional staffing costs to Health Boards on top of usual care were up to £300,000 - £500,000 per patient per annum. Other additional costs to Health Boards were less easy to quantify but these included assessments (£2000 per patient), travel and subsistence allowances for families, and costly adaptations and repairs to existing wards.

Considerable time and resources were used in 'containing' situations, with patients being maintained in unsuitable units pending an appropriate bed. Costs to local councils for some patients awaiting suitable hospital care were similarly high, between £260,000 and £360,000 per patient per year.

4.10.2 Costs of patients admitted to specialist LD CAMH/ ASD specific wards in NHS England

4.10.2 (i) Financial risk sharing scheme

The access of patients who were admitted to specialist LD CAMH and ASD specific wards in NHS England is covered under a financial risk sharing scheme administered for and on behalf of the 14 territorial NHS Scotland Health Boards by NSD (National Specialist and Screening Services Directorate of NHS National Services Scotland). This risk share scheme was established in 1999 following the Management Executive letter setting out the policy on replacement of Extra Contractual Referrals issued on 25 January 1999 (MEL1999/4).

NHS Board Chief Executives and Scottish Government colleagues approved the former National Services Advisory Group ( NSAG, now National Specialist Services Committee, NSSC) recommendation for funding to be top-sliced off Health Board allocations and transferred to NSD to cover equitable Scottish access to specialised services in England where equivalent services were not available in Scotland. The analysis of national expenditure and referral patterns has enabled NSD to identify opportunities to encourage service development in NHS Scotland and reduce the need for cross-border care in other acute areas of health care.

The establishment of new commissioning arrangements for specialist CAMHS services within NHS England from 1 April 2013 led to changes in the arrangements that had been in place for Scottish patients to access specialist services in England. This made it more difficult for patients from Scotland to access beds due to priority being given to NHS England patients. This restriction in access to NHS England beds was not reflected in NSD spends. However, there is likely to have been financial impact on Health Boards, who would have had to pay for private LD CAMHS hospital admission instead, or (along with Councils) for alternative arrangements whilst awaiting an NHS bed.

4.10.2 (ii) NSD spending on LD CAMH and ASD CAMH inpatient care

The cost attached to the provision of specialist LD CAMH and ASD CAMH inpatient care in NHS units in England has risen sharply over the last 5 years from a figure of £ 63,269 in 2010/11, to a sum £1.06 million in 2014/15. Patient numbers are less than 5 for both of these financial years and therefore cannot be reported. Northumberland, Tyne and Wear NHS Foundation Trust are the main provider of care. It should be pointed out that costs do not necessarily reflect demand. For example, there can be limited access to beds, and costs incurred over the last 5 years will have been contained by these restrictions. Further factors are discussed in section 5.9.4 of this report.

It would be prudent to estimate a figure of circa £1 million per annum for the provision of specialist LD CAMH and ASD CAMH inpatient care in England based upon case mix and complexity of care delivered over the past two years. However, costs to NSD are directly proportional to access to service provision in NHS facilities in England. Costs can also vary significantly because of the level of care package required. Some patients require much more nursing support than others, and with relatively long lengths of stay, it is possible for the care of individual patients to prove very costly. As each patient will have had a package of care tailored to their individual needs, the range of costs will have varied considerably. However the costs for the non-forensic care tend to be limited by the fact that Northumberland, Tyne and Wear NHS Foundation Trust look to complete their assessment and seek a discharge plan for their patients within a 13 week window, limiting costs for complex LD CAMHS patients to around £120,000 per in-patient episode. It is recognised that limited access for complex LD CAMHS patients to NSD-funded NHS England LD CAMHS inpatient units for a variety of reasons has meant that costs for those requiring inpatient mental health care has been mostly borne by Health Boards and Local Authorities.

It should be noted that pre-admission assessments generally cost an additional £2000 per patient, and an individual patient may require more than one assessment. This cost is usually borne by NSD via the risk sharing scheme for patients assessed by NHS England units but paid for directly by Health Boards for private hospitals.

4.10.3 Direct costs to Health Boards

Information received by the survey on direct costs to Health Boards was limited and incomplete. Therefore it was difficult to estimate an overall cost that includes all the various elements involved. However, the following costs were identified:

Cost of private hospitals in England: Where NHS England LD CAMH/ ASD beds were unavailable or unsuitable for the children and young people in this survey, a small number were admitted to private hospitals in England. These admissions were paid for directly by the home Health Board and were most commonly to St Andrews Healthcare in Northampton. Costs varied from £330,000 to £624,000 per patient per year. Some admissions were very lengthy, e.g. for 2 ½ years.

Cost of adult LD hospitals in Scotland: Costs were variable when children and young people were admitted to adult LD wards, basic costs being approximately £3,300 per week per patient (£170,000 per annum). However, to nurse the children and young people safely in that environment additional cost were common, e.g.

• £8,000 – £10,000 for additional nursing staff for 2:1 or greater levels of care (£400,000 – £520,000 per annum)
• £6,700 per week (£345,000 per annum) for one admission where a part of an adult ward had to be used for one young person with high staff ratios.

Admissions could be lengthy, e.g. 3 years at £455,000 per annum for one patient, 18 months at £520,000 per annum for another.

Cost of admissions to adult mental health wards, including IPCU: Costs were quoted as between £1,000 and £6,000 a week (£52,000 - £312,000 per annum), presumably due to variable levels of support required.

Cost of admissions to YPUs: Costs for nursing children and young people 1:1 or 2:1 on YPUs were described of up to £312,000 per annum.

Cost of care on pediatric wards: Admissions to these wards tended to be shorter, but there was still significant cost of between £4,000 and £8,000 a week for additional nursing staff to support individual children and young people.

Cost of other admissions: The Lothian LD CAMH Intensive team uses existing staff to support admissions to a dedicated individual unit when their patients require admission. On top of the existing cost of intensive team nurses, additional nursing care, including bank staff, can cost £200,000 per year.

Travel and subsistence costs: The Health Board of residence is responsible for refunding travel and subsistence costs for families visiting their child in hospital. Limited information was received on this but an example was given of approximately £3,000 for 1 child or young person for 6 months.

Costs of adaptations to wards: It was not possible to ascertain the costs to Health Boards associated with adaptations and repairs during admissions to existing wards. These were most commonly made when adolescents with severely destructive, self-injurious or aggressive behaviours were admitted to adult LD wards. Costs could be significant over prolonged periods, e.g. building of a 'bespoke' robust, ASD-specific unit within an adult LD ward to accommodate one teenager. Daily repairs, including by out-of-hours joiners, were commonly required. Similar costs were associated with a smaller number of patients admitted to one of the YPUs, but for shorter lengths of time.

4.10.4 Costs to local authorities

The survey received very limited information about costs to local authorities associated with admissions (or lack of availability of admissions) for this patient group. However, there were indications that costs of looking after these children outside hospital while awaiting hospital care were of a similar cost to inpatient care. For example:

• £360,000 per annum for a residential special school
• Estimated £130,000 for additional support locally for 6 months for a young person while awaiting a bed in an LD CAMH NHS unit in England.

Some local councils also contributed to the cost of admissions, mostly to local adult LD wards, for example, by funding education and social care professionals to work with a young person in hospital. This could be to maintain education and community access, to keep up contact with known care staff, or to develop relationships with care staff prior to discharge. One council indicated that they paid £400 per week for education to be provided to a young person on an adult LD ward.

4.10.5 Other impacts on local services and on clinicians

Each local admission for those with more severe learning disability and complex needs/severe challenging behavior was a time-consuming and stressful 'special arrangement' for clinicians. Admissions to specialist units in England involved lengthy negotiations around funding and arrangements. Whilst arranging/waiting for a bed, or as an alternative to admission, community clinicians had to manage unacceptably high levels of risk in the community with very limited services. This was anxiety-provoking, with clinicians feeling isolated and unsupported. Relationships between services were strained due to a lack of management responsibility and 'ownership' of these children and young people.

Community-based LD CAMH psychiatrists had to retain responsibility for inpatients over long periods. Inpatient teams and facilities in non-specialist wards had to be 'cobbled together' for individual patients, building up working relationships from scratch each time. Inpatient nurses were anxious due to lack of experience with this particular patient group, and some nurses sustained some serious injuries. Adult patients with learning disability lost access to inpatient care. Many additional unpaid hours were worked, impacting on clinicians' personal lives and care of other community patients. There was a concern that recruitment to such stressful community services will be affected. More detail on these impacts can be found in appendices section 4.18.

4.11 Numbers and profiles of patients requiring specialist inpatient care not presently available in Scotland

A total of 54 patients were identified by their clinicians over the 5 year period as having required inpatient care not presently available in Scotland. 45 of these required LD CAMH specialist inpatient provision and 9 required non-learning disability specific inpatient provisions (see also section A4.19 of appendices).

4.11.1 Profile of patients requiring specialist LD CAMH inpatient provision

Compared with patients able to access 'mainstream' YPUs, the 45 requiring specialist LD CAMH inpatient facilities had greater degrees of learning disability, with all of those with severe/profound learning disability requiring such provision. They were more likely to be male (70%). 21 patients were aged 16-17, 15 aged 14-15 and 9 aged 13 or under. The main reasons for requiring hospital admission were risk management and mental health assessment/stabilisation. Children and young people had very high levels of distress and severe challenging behaviours, requiring high staff ratios, 40% needing 1:1 and 31% needing 2:1 care.

These 45 children and young people had 76 periods of inpatient care, mostly due to transfers between units during 1 admission, a minority having more than 1 admission. 44 periods of care were in Scottish adult LD or adult mental health wards (including secure and intensive psychiatric care units), 12 in Scottish CAMH wards, 4 in Scottish pediatric wards, 6 were not admitted to hospital at all, and 10 were in specialist LD CAMH wards in England.

Of the 45 children and young people identified by this survey as requiring LD CAMHS specialist provision:

• 29 required an LD CAMHS specialist unit without additional security or individualised provision
• 9 (mostly with mild learning disability) required a secure LD CAMH inpatient facility
• 7 (with moderate/severe learning disability) required a robust, individualised LD CAMHS inpatient environment, giving the option of nursing children and young people away from others when required.

Those requiring robust, individualised environments were amongst those with the most complex and challenging difficulties. Some could not access inpatient care at all, or had required highly expensive bespoke provision to be built locally, with huge impact on local services.

4.11.2 Autism spectrum disorder (non- learning disability) specific provision

9 children and young people in this survey required autism spectrum disorder (non-learning disability) specific provision that is not available in Scotland. The majority of these had autism spectrum disorder and no learning disability or mild learning disability and required secure/forensic inpatient care. There was insufficient evidence of need for a specialist unit for children and young people with autism spectrum disorder without learning disability. However, information on these children and young people highlights the need for the proposed secure forensic CAMH inpatient unit to have high levels of autism spectrum disorder and learning disability knowledge and skills and an appropriate physical environment to meet the needs of this group.

There is also likely to be a very small ongoing number of children and young people with autism spectrum disorder but no learning disability who will require specialist inpatient care not covered by the remits of the proposed secure forensic CAMH or LD CAMH inpatient units, for example, those requiring a higher level of autism spectrum disorder expertise or a higher level of security than can be provided. The information gathered by this study indicates this is likely to be for an average of less than one young person from Scotland each year. It is important that clear commissioning arrangements are in place so that such admissions can be easily arranged elsewhere in the UK if required.

4.12 Other study findings

4.12.1 Community LD CAMH services

Results confirmed previous reports ( LD CAMHS Framework Document, Appendix A2.2) of inadequate community mental health services for children and young people with learning disability. Where present, community expertise could be so fragile that having one key clinician on leave could result in an absence of LD CAMHS expertise. This meant that services struggled to safely manage the children and young people at home whilst awaiting admission. Section A4.16 of appendices provides further detail.

Intensive community LD CAMHS services, along with earlier, more robust specialist social care, education and health provisions could have helped manage some children and young people more safely in the community before and after admission. Given the inclusion criteria for this survey and from information provided by clinicians, it is likely that the vast majority of patients identified in the survey would have required hospital admission even if intensive community treatment services had been available locally. Indeed, thresholds were so high that numbers are likely to underestimate true need. However, such provisions could also have allowed for shorter admissions and a sooner return home or to a community placement nearer home.

Intensive community treatment services capable of working in children and young people's homes, school and respite provisions are crucial to provide the full range of appropriately stepped care for children and young people with learning disability and mental health/behavioural difficulties. Their development nationally, alongside the development of an LD CAMH inpatient unit, would prevent unnecessary admissions to such a unit. Along with adequate availability of specialist educational and social care resources, these would be essential in ensuring appropriate admissions and aiding timely discharge out of the unit.

The ' LD CAMHS Models and Outcomes Study' (Glasgow University) is specifically investigating intensive models across the UK and results will be available soon. Clinicians in this study also noted that, while these services could be very valuable, intensive support in patients' homes is not always practical or appropriate, due to family issues or physical layout of their house. Home may not be a safe environment for workers as well as family. Families may not want strangers in their house overnight, particularly where there are siblings. Intensive treatment services can only work in the context of appropriate social care support and respite.

4.12.2 Robust individualized settings

The need for robust individualised settings was a key theme throughout the results of the survey, in inpatient and community settings, including social care and education. The relatively small number of children and young people that required this were amongst those with the most complex and challenging difficulties who were most difficult to place in inpatient care. Some were unable to access inpatient care at all and admissions that did take place had a huge impact on local services and other patient care. For a minority, there was no hospital ( NHS or private) in the UK that could provide the physical environment required and major adaptations and building work had to be carried out in a local adult environment to accommodate them.

Very physically robust physical environments are required to withstand highly destructive behavior for these children and young people to be safely and effectively treated in hospital. Importantly, these environments need to also take into account the sensory processing difficulties common in children and young people with autism spectrum disorder and aim to reduce arousal/anxiety levels. Some children and young people need to be nursed separately from other patients for variable periods of time. They therefore require an individual part of a unit with their own living, sleeping and bathroom areas, with access to safe outdoor space, activities, education and a full multidisciplinary team. Both patient and staff safety need to be considered in the design of the building for children and young people with very severe self-injurious and aggressive behaviours.