Integrated care to address child and adolescent health in the 21st century: A clinical review

Abstract Background Increasing specialisation and technical sophistication of medical tools across the 21st century have contributed to dramatic improvements in the life‐expectancy of children and adolescents with complex physical health problems. Concurrently, there is growing appreciation within the community of the extent that children and adolescents experience mental disorders, which are more prevalent in those with complex chronic, serious or life‐limiting health conditions. In this context, there are compelling reasons for paediatric services to move to a model of care that promotes greater integration of child psychiatry within the medical, somatic teams that care for children and adolescents in children’s hospitals. Aims In this article, we discuss the range of medical disorders managed by contemporary paediatrics. Materials and Methods We conducted a broad review of the literature and existing services, and use individual accounts to illustrate adolescents’ healthcare preferences in the context of the challenges they experience around their mental health. Results Relevant disorders include life‐limiting disorders, such as cancer; disorders involving the brain, such as epilepsy; common chronic disorders, such as asthma and diabetes; psychiatric emergencies, such as deliberate self‐harm; and conditions that most commonly present to paediatric services, but where psychiatric input is required, such as severe eating disorders, somatic symptom disorders and gender dysphoria. The persisting legacy of the historical separation of physical and mental health services is described. Yet there are many models of service integration that can promote more collaborative care between psychiatrists and medical specialists, including some which have been taken to scale. Discussion In essence, clinical teams in children’s hospitals require more collaborative approaches that facilitate early recognition and treatment of the psychological aspects of illness as an integral part of patient‐centred, family‐focussed paediatric care, rather than as something that is bolted on when things go wrong. Conclusion Whilst trust and goodwill between services and providers will be required for novel models of care to be implemented, evaluation of these new models and incorporation of young people’s healthcare preferences is needed.


THE CHANGING FACE OF MEDICINE
There should be full collaboration between mental health and physical health services, and they should all be in the same place. In high-income countries, the complexity of needs of many patients in tertiary children's hospitals reflects these advances and challenges, and raises questions about the extent to which current models of care are fit for purpose (Sharpe & Naylor, 2016). Previously the majority of paediatric patients would be managed by a single specialist, who could address the majority of their health care needs.
In contrast, contemporary paediatrics is accompanied by a level of specialisation and sophistication of technical tools that commonly requires expert multidisciplinary teams (MDT) to manage complex chronic health conditions and disabilities. This shift, from a more singular orientation led by a key clinician towards more team-based care for children and adolescents, forces us to reconsider not only how care is delivered, but by whom.
An additional challenge is the importance of responding to children's and adolescents' emotional distress. There is growing awareness of the extent of mental health difficulties in young people in general, especially anxiety and depression, as well as the value of identifying individuals and families who are most negatively affected by the social and environmental determinants of health. There is also growing evidence that children and adolescents with complex chronic health conditions and disabilities experience a higher prevalence of mental illness, which in turn impacts their somatic health outcomes.
In this clinical review, we discuss the value of greater integration of mental and physical health care for children and adolescents managed by tertiary children's hospitals. To do this, we will describe clinical conditions that require a more integrated model of care where 'all parts are brought together' to work seamlessly for patients and their families (Royal College of Psychiatrists, 2019). We will explain some aspects of the history of the separation of paediatric and psychiatric services, briefly highlight some different models of integration of care and conclude with a call for greater integration to embrace the needs and demands of the current generation of younger patients and their families. Throughout this review, we include the voices of adolescents who are currently receiving care at a tertiary children's hospital in the United Kingdom. We will use the term 'child and adolescent' to describe the population commonly seen in children's hospitals, predominantly up to the age of 18 years.

THE NEED FOR PSYCHIATRY IN CHILDREN'S HOSPITALS
The nature of services provided by contemporary children's hospitals must reflect the breadth of clinical presentations that they manage.
Evidence increasingly highlights the high prevalence of psychiatric and neurodevelopmental disorders in the general population of children and adolescents, and particularly in those with complex and chronic medical disorders (Sadler et al., 2018). This is especially the case for anxiety and depression, although behavioural problems and eating disorders are also notable. There are five main areas of activity that are particularly relevant to psychiatry in a children's hospital: � life-limiting illnesses, such as cancer; � disorders involving the brain, such as epilepsy; � common chronic physical disorders, such as diabetes; � psychiatric emergencies, such as presentations of self-harm; � disorders that commonly present to paediatric services but where psychiatric input is often required, such as severe eating disorders, somatic symptom disorders and gender dysphoria.
Each of these areas will be briefly discussed, appreciating that some conditions span overlapping areas. This discussion has been informed by a narrative review of the literature as well as the views

Key points
� As the science of medicine improves, children and adolescents with increasingly complex needs are being managed by paediatric teams.
� Better integration of somatic and psychological care is needed with models of collaborative care in children's hospital settings being developed.
� The range of disorders that benefit from integrated care include cancer care, disorders involving the brain, common chronic conditions, psychiatric emergencies and chronic pain and eating disorder treatment.
� Incorporating the preferences of children and adolescents, as well as their families, into these models of care remains important. and experiences of young patients managed within an integrated clinical service in the United Kingdom. A summary of these data is presented in Appendix S1, whilst management approaches to consider are presented in Figure 1. Although the focus of this article is limited to child psychiatry and paediatric medicine, it is important to keep in mind that there are many different health care professionals that work within tertiary care settings with broad and wide-reaching roles essential to both the psychological and physical care of the patients and families seen.

Life-limiting illnesses
Definitely having mental health support is very important for children with cancer. It is a massive thing and it affects you in every way, physically, emotionally and socially. You need someone there and then, you can't wait. Having mental health support was very important for me, because I was very depressed. It was difficult for me, because I knew my life was in danger and having someone that understands and gives you the support that you need is invaluable. It also helped that I could tell them medical things I was struggling with, but was uncomfortable to tell the oncologist.
However, sometimes I wished the support had been offered away from the hospital and the medical ward, as I was reminded constantly of the treatment and the threat of the illness. Ideally patients should choose whether they want to be seen in the same place or not.
It is important that the people providing the mental health support know about the physical side of things, so that you don't have to explain it over and over again.
[18 year old patient with an acute leukaemia] Conditions with a high disease burden, such as cancer, have been shown to increase the risk of psychiatric disorder (Kurtz & Abrams, 2011;Temming & Jenney, 2010) with importance placed on conceptualising psychological outcomes (Sansom-Daly & Wakefield, 2013). For example, a study of 1345 adolescent survivors of leukaemia found an approximately 1.5 times increased risk of depression, anxiety, attention deficit and antisocial behaviours compared to siblings (Temming & Jenney, 2010). A further review of 5320 paediatric brain tumour survivors reported increased incidence of depression, anxiety, schizophrenia and related psychoses and behavioural problems compared to the general population (Shah et al., 2015).
The nature of the patient journey for a particular cancer is a predictor of its potential psychological effects. The care pathway from diagnosis of cancer to long-term follow-up usually spans a period of many years and involves repeated investigations, procedures, admissions and treatments. These experiences can have farreaching psychosocial repercussions, which have encouraged some countries to publish standards for psychosocial care (Wiener et al., 2015). For example, anticipatory anxiety is a common issue in clinical practice benefiting from a MDT approach involving judicious use of symptom-specific medication, such as analgesia and antiemetics and psychological support, such as distraction and relaxation techniques (Kurtz & Abrams, 2011). The medical interventions required to manage cancer can disrupt engagement with social activities, school, peers, family, as well as the individual's development of body image, identity and sexuality (Kurtz & Abrams, 2011). With improvements in prognosis, a growing cohort of patients transition from being a 'cancer patient' to a 'cancer survivor'. This change in identity can be difficult to adjust to psychologically, which may not be reflected in the level and nature of support available as these adolescents will often then transition from hospital to community services (Kurtz & Abrams, 2011;Sawyer et al., 2017). The Childhood Cancer Survivorship Study of 20,276 patients reported a significantly increased risk of psychosocial issues compared to controls (Kurtz & Abrams, 2011;Wakefield et al., 2010). There are particular risks associated with different cancers and treatment type, but central nervous system (CNS) tumours, neurosurgery, chemotherapeutic agents and cranial radiation can all result in neurocognitive sequelae, further placing young people at particular risk for psychiatric comorbidities.

Disorders involving the brain
When I first came to the hospital I thought my problems were physical. When I was told I didn't have epilepsy I had a mixed bunch of emotions. Initially I was very confused because I had been taking antiepileptic medication for 5 years. It was hard for me to trust medical professionals. Now I really understand what epilepsy is, and that I don't have it, but I needed help to understand it. If at that point I had been told I needed to go somewhere else to get mental health support, I would have felt that they were brushing me off. I would have been angry and annoyed towards the health care system. I probably would have become depressed. I don't think I would have been able to trust medical professionals again [15 year old male presenting with seizures. His mother's account is included in In addition to CNS tumours, neurological disorders, such as epilepsy and traumatic brain injuries (TBIs), are associated with a high prevalence of psychiatric comorbidities. This significantly increases the complexity of patient management, which is often not adequately reflected in best practice guidelines, highlighting the need for collaboration between psychiatry and neurology (Plevin & Smith, 2019, Salpekar et al., 2015. A recent review of children post-TBI found the prevalence of depression was 33%-50% (Ryttersgaard et al., 2020). Novel psychiatric diagnoses post-TBI include a spectrum of personality change, ADHD and other behavioural disorders (Max, 2014;Ryttersgaard et al., 2020).
A review of children with epilepsy (CWE) reported that psychopathology occurs in 37%-77% including depression, anxiety, intellectual impairments, ADHD and autism spectrum disorders (ASD; Verrotti et al., 2014). Another review reported that the risk of depression remains two to five times higher than the general population even when epilepsy is adequately controlled (Salpekar et al., 2015). The relationship between psychiatric symptoms and epilepsy is complex. For example, 'stress' is one of the most commonly self-reported precipitants of seizures (Verrotti et al., 2014). The diagnosis of psychiatric disorders in CWE can be complicated by the presence of neurocognitive deficits and the lack of validated screening tools or diagnostic criteria in this population.
For example, there are few assessment tools for children with intellectual impairment and anxiety (Plevin & Smith, 2019). Careful consideration is required when prescribing for this patient group as anti-epileptic medications may have psychiatric side effect profiles and psychiatric medications may interact with anti-epileptic drugs (Plevin & Smith, 2019). As psychiatric comorbidity in CWE increases impairment and early diagnosis of psychiatric problems is key to improving clinical outcomes, functioning and quality of life (Plevin & Smith, 2019), effective collaboration between specialities optimises patient management. Asthma and type 1 diabetes (T1DM) are amongst the most frequently encountered chronic physical disorders in children's hospitals. A strong evidence base demonstrates the increased prevalence of psychological disorders in patients with these conditions (Dudeney et al., 2017, Peters & Fritz, 2011, Rechenberg et al., 2017, Winston, 2020. Optimal management relies upon supporting, educating and empowering patients and families to understand the condition, including the importance of adhering to treatment to minimise the risk of acute and long-term complications. Studies have shown a correlation between psychiatric disorders and lower adherence to treatment with resulting poorer health outcomes (Rechenberg et al., 2017, Winston, 2020, Young et al., 2013. Recognising the prevalence of psychiatric comorbidity in young people with T1DM needs to be integral to their management. A review by Dudeney et al. (2017) found that the prevalence of anxiety disorders in adolescents under the age of 18 years with asthma is more than three times higher than the general population.
The prevalence of depression, ADHD, behavioural disorders and intellectual impairment is also higher in this population (Peters & Fritz, 2011). The prevalence of depression in children aged 3-18 years with T1DM is estimated to be twice that of the general population, whilst the prevalence of anxiety is similarly increased and estimated to be 18.4% (Rechenberg et al., 2017). Studies have also shown an increased prevalence of eating disorders in adolescents with diabetes. For example, a recent review reported a prevalence of 10% in 12-19 year old females with T1DM compared to 4% in controls without diabetes, with even higher rates of subthreshold disorders (Winston, 2020). Comorbid psychiatric disorders affect the management of chronic disease in a number of ways.
Whilst emotional triggers are a common precipitant of 'asthma attacks', people with asthma can also confuse symptoms of anxiety with asthma, which can lead to a cycle of overuse of short-acting reliever inhalers (Goodwin et al., 2012). Both anxiety and depression are correlated with a higher asthma symptom burden (Peters & Fritz, 2011). Similarly, in T1DM, psychiatric disorders are associated with poorer disease control, increasing the risk of elevated HbA1c levels and acute and chronic complications (Rechenberg et al., 2017;Winston, 2020;Young et al., 2013).

Psychiatric emergencies presenting to children's hospitals
Adolescents with psychiatric emergencies, such as deliberate selfharm (DSH), severe intoxication and delirium frequently present to medical rather than psychiatric settings, such as emergency departments (ED). A recent UK e-cohort study reported an increase in ED attendances for self-harm in 15-19 year olds for both females and males over the past decade, with a corresponding increase in general hospital admissions of females with self-harm (Marchant et al., 2020). Similarly, in Victoria, Australia, an anal- One UK study reported that 27% of ED attendances for DSH were associated with a subsequent outpatient appointment (Marchant et al., 2020), with studies examining compliance with agreed discharge plans reporting around 66% compliance in a study from Ohio, United States (Sobolewski et al., 2013) and 80% from an Australian study (median age 14 years, majority of presentations for self-harm ;Hopper et al., 2011).
In the acute hospital, especially in intensive care units, delirium is another common psychiatric emergency, which is seen in up to 50% of critically ill children and adolescents (Wilson et al., 2020). Management often involves prescription of psychotropic medication and a thorough understanding of possible underlying somatic conditions, which necessitates appropriate psychiatric input to support acute care, educate staff and families, and plan follow-up.
Disorders that commonly present to paediatric services but where psychiatric input is often required In the seventeenth century, the philosopher John Locke described the mind of a child as 'white paper, void of all characters, without any ideas' (Karafyllis, 2018). This 'tabula rasa' kind of thinking contributed to assumptions that children could therefore not experience mental illness. In the nineteenth century, public health and education reforms in the United Kingdom led to improvements in sanitation, campaigns for women's health and universal education, with some children identified as unable to meet expectations either in learning, behaviour or because of 'fragility'.
Prior to this, children had not been treated separately from adults but were seen as miniature adults, expected to work and be useful within the family and society. Children who deviated from behavioural norms were regarded as deserving of punishment rather than benefitting from medical diagnosis or treatment, and those with intellectual impairments or psychiatric disorders were highly stigmatised. Children as young as five would be managed by general physicians and be admitted to asylums, the forerunners of adult psychiatric wards (Parry-Jones, 1989).
In the twentieth century, several new strands of thought were beginning to be applied to mental illness in children and adolescents.
Psychotherapeutic approaches, which had begun to have a following for adults, were becoming available for children, led by psychologists, such as Anna Freud and Melanie Klein, and paediatricians, such as Donald Winnicott, amongst others. In parallel with child psychotherapy and psychology, the Child Guidance movement started in the United States in the 1930s (Rey et al., 2015). This became a model for working with children with behavioural problems and delinquency.
The Child Guidance Centres included social workers, educational psychologists and psychotherapists who worked as a team with the belief that the child must be understood in the context of the family and social environment. Becoming a focus for research as well, these centres eventually also housed child psychiatry, as envisioned in the 1946 UK Blacker report.
Emphasis on the highly specialist nature of child psychiatry, available to only a few, meant that services were never designed or resourced to meet the true need in the population (Barrett, 2019).
In spite of the good intentions enshrined in the Blacker report, the imposition of psychiatrists into Child Guidance Centres was not always popular. There could be a struggle for dominance in teams, with interdisciplinary rivalry and disagreement about which explanatory model (attachment theory, psychodynamic psychotherapy, family therapy, ecological systems theory, genetic, etc.) was best placed to inform intervention decisions for children. Furthermore, at this time the 'medical model' became seen as being opposed to more holistic thinking about the needs of the individual child (Williams & Kerfoot, 2005).

NEW MODELS OF INTEGRATION
This historical legacy has left children's hospitals in many countries with mental and physical healthcare that continues to be delivered in a fragmented, often separated manner which can compromise access, quality and efficiency of care (Druss & Goldman, 2018). Newer The second level more specifically addresses care coordination.
Whilst co-location of psychiatry and primary care improves access, the benefits are not fully reaped in the absence of collaborative care models, such as when a care manager coordinates care between primary care clinicians and specialists. A recent review of paediatric integrated and collaborative care models found that increased access to psychiatric treatment can improve outcomes for medically ill children and adolescents (Burkhart et al., 2020).
Beyond care coordination, a third step to integration is when paediatric and child psychiatry systems more seamlessly promote coordinated care, including through shared access to medical records. Different services are developing new models of embedding mental health specialists in speciality clinics. This includes specialist paediatric eating disorder services where access to multidisciplinary care is increasingly the norm or having psychologists based within epilepsy clinics as services become increasingly focused on outpatients (Hughes et al., 2014, Plevin & Smith, 2019. As with adult inpatient hospital services (Sharpe et al., 2020), paediatric inpatient services are also exploring proactive provision, such as the Children's Psychological Medicine Service in Oxford University Hospitals, where integrated care ensures psychiatrists and psychologists attend the daily rounds of all new paediatric admissions to proactively identify those children and adolescents who would benefit from an integrated approach.
Yet the dead hand of history often impedes change, which can take considerable goodwill, time, effort and investment to shift (Druss & Goldman, 2018). Change requires a step-wise approach towards integration, with an appreciation that trust needs to be established between the people running services. Addressing organisational challenges, key financing questions and differing service values takes time (Druss & Goldman, 2018). Physical co-location can be an important early step to facilitate closer professional collaboration and the development of a shared vision (Peek & National Integration Academy Council, 2013). Table 1 highlights some professional opportunities that psychiatrists can view as helpful steps towards closer collaboration.
In the United States, a number of important innovations have been developed around integrated psychiatric and paediatric training that can help support more integrated models of clinical care. For example, in the mid-1980s a combined training programme was set up in a few centres to bring paediatrics, psychiatry and child psychiatry together in the formation of Triple Board training (Gleason & Sexson, 2017). More recently, a 'Pediatric portal' programme was established to encourage board certified paediatricians to supplement their skills through additional formal training in psychiatry, as well as discussions about the common core competencies required within each speciality (Gleason & Sexson, 2017, Shaw et al., 2019.
Amongst other outcomes, these approaches are expected to build trust between psychiatrists and paediatricians.
Child and adolescent psychiatrists working in hospital systems need to ensure they have the expertise to encompass the complexity, treatment implications and medical systems being navigated -not just for a specific patient, but by the ward, team or even hospital.
Their professional competencies need to contribute to the diagnosis, formulation and treatment plans developed by utilising core skills that bring together biological, psychodynamic, pharmacological, developmental, family and systems perspectives into the consultation. The growing awareness of the psychological landscape of the child and adolescent being treated, can be more easily acknowledged and explored when medical, nursing and allied health staff have strong appreciation of the value of providing psychological assessment and interventions within more integrated models of care in children's hospitals. In addition, the presence of mental health professionals within services might enhance skill building for all staff, including through clinical discussions and supervision, improving everyone's capacity to acknowledge and address the psychological needs of children and their families, as well as the clinical teams that support them. F I G U R E 2 Levels of Collaboration in models of Integrated Care. Adapted from Heath et al. (2013). Deeper blue is intended to signal a higher level of collaboration  (Funk & Ivbijaro, 2008).

A CALL FOR ACTION
Increasing numbers of children and adolescents with life-limiting, complex chronic health conditions and disabilities are presenting to medical services with complex psychological needs. Psychiatric comorbidity affects adherence to medical treatment, health outcomes, quality of life and global functioning in the foundational years of human development. Yet contemporary clinical contexts, characterised by increasing sub-specialisation and high levels of technical competence, risk reinforcing the legacy of distinct physical and mental health services, eroding opportunities for more integrated approaches for patient-centred and family focussed healthcare delivery. Improved experiences of care, better quality health of the population and reduced healthcare costs for individuals are compelling reasons for more integrated services (Kullgren et al., 2020).
As the majority of major lifelong mental illnesses start in the adolescent years when patterns of self-care and care-seeking become established, approaches to engaging young people with chronic health conditions around both their physical and emotional wellbeing are important. Adolescence can be especially complex for this cohort whose interpersonal relationships with family, schools and peers can be as disrupted by the complexity of their disorders and the healthcare systems they must negotiate (Patterson & Vakili, 2014, Rickerby & Roesler, 2015, Sawyer et al., 2007. For those with uncertain prognoses or life-limiting conditions, the additional challenges of managing difficult symptoms (e.g. pain, fatigue) and repeated hospital appointments or admissions can exact a heavy burden on their emotional wellbeing, friendships and family life (Barker et al., 2019).
Despite increasing numbers of young people growing up with chronic and life-limiting conditions (Barker et al., 2019), there is continued inertia within the health system around approaches to better prevent, diagnose and treat comorbid psychiatric illness.
Embedded within our legacy of separated physical and mental health services, this inertia is supported by separate institutions, reflected in independent training and reinforced by distinct funding schemes.
Child and adolescent health and wellbeing would be greatly supported by research in and development of creative service models that enhance integration. The COVID-19 pandemic has contributed to a mental health crisis in young people and resulted in a groundswell of interest in child and adolescent emotional wellbeing. Now is the time for the field to grasp the opportunity of responding, not by simply doing more of the same (as we are not meeting the needs of paediatric patients and their families) but by considering opportunities for doing things differently.
Finally, the opportunities to be gained extend beyond some of these examples in high-resource settings. Child psychiatry is even more poorly accessible in resource-poor countries where paediatric care remains focussed on acute care, where mental health literacy is particularly low and mental illness remains highly stigmatised (Sawyer et al., 2019). These countries have even more to gain than high income countries in developing cost-effective, integrated models of care which can address both physical and mental health and wellbeing.