Future of Child and Adolescent Mental Health Services: How best to support wellbeing

Future of Child and Adolescent Mental Health Services: How best to support wellbeing Dr Miranda Wolpert Reader in Evidence Based Practice and Research, UCL Director of Evidence Based Practice Unit, the Anna Freud Centre Director of Child Outcomes Research Consortium Mental Health Theme Lead Child Policy Research Unit

Future of CAMHS Historical context Current challenges and opportunities Future prospects Research into nature of current provision: modelling needs based grouping

CAMHS challenges Child guidance vs Child Psychiatry Promoting wellness vs treating Illness Children vs young adults Working with parents vs working with autonomous children Working in schools vs working in clinics Small part of mental health services vs small part of child educational services

CAMHS History Resources not sufficient for need Lack of standardisation Lack of common language Lack of data 1 in 5 of those identified in need accessed service (Ford et al 2007) RCT evidence suggests 30% not improve (Warren et al 2010)

Current Delivery Challenges Rising need for girls with emotional problems though plateau or improvement behaviour problems in boys (Fink et al., 2015) Schools first point of contact but ongoing challenges in relationship between schools and CAMHS (Fazel et al., 2014) Increased pressure on services with reports of increased rates of self harm and complexity of referrals (Hindley 2014)

Delivery Opportunities High interest and commitment Increasingly aligned models of what good looks like Ring fenced funding to support ongoing transformation Increased collaborative working DH, Dfe, NHSE and PHE Data to flow nationally from NHS services from 2016 New national survey Cross-sector data initiatives being piloted by CORC and others

Aligned policy

Support on mental health literacy 8

Supporting choice 9

Finding support

Considering need Support for shared decision making (Mulley) Patient activation focusses on self management (Greenhalgh) Increased awareness of negative consequences of treatment (Glasziou) Focus on value in healthcare (Gray) Interest in more focussed payment systems

Need Need is defined as the identified approach collaboratively agreed via a process of shared decision making between service provider and service user. It includes both judgement of the appropriateness of interventions offered and the informed choices of children, young people and their carers regarding the approach... that is best for them, within the parameters and scope of the commissioned service Wolpert, M., Vostanis, P., Young, S., Clark, B., Davies, R., Fleming, I., Whale, A. (2015). Child and Adolescent Mental Health Services Payment System Project: Final Report. London: CAMHS Press. p.7

CAMHS Future?

Understanding CAMHS case mix and resource use Payment Systems Project : http://pbrcamhs.org/final-report/ 14

Project Group Dr Miranda Wolpert, Professor Panos Vostanis, Simon Young, Dr Bruce Clark Dr Roger Davies, Dr Isobel Fleming Dr Lynne Howey Pat Howley Amy Macdougall Dr Peter Martin Tony Martin Charlotte Payne Benjamin Ritchie Dr Rob Senior Dr Ann York Dr Andy Whale

Analysis of data from existing datasets: Child Outcomes Research Consortium (CORC) (2012-13) 38,794 periods of contact for children (0-18) from 107 clinical teams in 21 services, submitted to CORC between March 2012 and December 2013. Modal number of appointments was 1; 24% of periods of contact were closed after the first appointment. Half of all cases were closed after three appointments or fewer had been attended. 37.8% of all appointments were attended by the 5.25% most resource-intensive patients, who attended more than 30 appointments each (Wolpert, et al., 2015, p.21) Greater resource use was associated with greater clinician-rated severity (e.g. on CGAS child global assessment scale) Some problems were more highly represented in the resource-intensive group such as eating disorders and psychosis. Great variability in terms of the amount of resource use, and type and severity of problem No correlation was possible to find in terms of any other indicators of need available in the dataset (Vostanis, et al., 2015; Wolpert, et al., 2015).

Analysis of resource use in relation to need by those accessing NHS outpatient CAMHS collected specifically as part of the Payment System Pilot Project (2012-14) 4573 episodes of care from 11 NHS outpatient CAMHS with resource information and case mix (current view) information Current View (Jones et al., 2013) 30 presenting problems7 (e.g. social anxiety, family relationship problems, carer management of child s behaviour) 4 contextual problems (relation to home, school, community, and service engagement 2 education, employment or training issues (attendance and attainment), 14 complexity factors (e.g. presence of learning disability, parental health issues, refugee status)

Categorising Mental Health problems 1. Anxious away from care givers (Separation anxiety) 11.Extremes of mood (Bipolar disorder) 21.Family relationship difficulties 2. Anxious in social situations (Social anxiety/phobia) 12. Delusional beliefs and hallucinations (Psychosis) 22. Problems in attachment to parent/carer (Attachment problems) 3. General anxiety (generalised anxiety) 13. Drug and alcohol difficulties (Substance abuse) 23. Peer relationship difficulties 4. Compelled to do or think things (OCD) 14. Difficulties sitting still or concentrating (ADHD/Hyperactivity) 24. Persistent difficulties managing relationships with others (includes emerging personality disorder) 5. Panics (Panic Disorder) 15. Behavioural difficulties (CD or ODD) 25. Does not speak (selective mutism) 6. Avoids going out (Agoraphobia) 16. Poses risk to others 26. Gender discomfort Issues (GID) 7. Avoids specific things (Specific phobia) 17.Carer management of CYP behaviour (e.g. management of child) 27. Unexplained physical symptoms 8. Repetitive problematic behaviours (Habit problems) 18. Doesn t go to the toilet in time (Elimination problems) 28. Unexplained developmental difficulties 9. Depression/low mood (Depression) 19. Disturbed by traumatic event (PTSD) 29.Self-care issues (includes medical care management, obesity) 10.Self-harm (Self injury or selfharm) 20.Eating issues (Anorexia/Bulimia) 30. Adjustment to health issues See pages 15, 17 and 19 in Current view Tool Completion guide

Note: N = 4573. Forty periods of contact were recorded to have attended more than thirty appointments. These are not shown in this graph, but are included in the analysis.

What does clustering ( grouping ) mean? Note: These images represent simulated illustrations only. No CAMHS data were used to make them.

Grouping Development: Three approaches Approach 1: Regression Trees Approach 2: K-medoids Cluster Analysis Approach 3: Conceptually guided grouping

Table 1: Predicted resource use for needs-based groupings, from payment systems project analysis Needs-based groupings Predicted % in grouping based on application of the algorithm 95% confidence interval of group percentage Predicted average no. of sessions 95% confidence interval of estimated average appointments Predicted % resource use for a typical service* Informal confidence range for predicted resource use** Getting advice 28% 27%-29% 6.2 4.6-8.47 24% 20%-29% Getting help 61% 60%-62% 6.9 5.1-9.5 59% 53%-65% Getting more help 11% 11%-12% 10.4 7.5-14.5 16% 13%-22% Total 100% -- 7.2 6.6-7.8 100% --

Illustrative indication of relative grouping sizes based on analysis of Current View data collected Sep 2012-June 2014 Draft groupings Getting Advice: Signposting and Self-management Advice [A1] Getting Advice: Neurodevelopmental Assessment [A2] Getting Help: Guided by NICE Guideline 16 and/or Guideline 133 (Self-harm) [H1] Getting Help: Guided by NICE Guideline 26 (PTSD) [H2] Getting Help: Guided by NICE Guideline 28 (Depression) [H3] Getting Help: Guided by NICE Guideline 31 (OCD) [H4] Getting Help: Guided by NICE Guideline 38 (Bipolar Disorder) [H5] Getting Help: Guided by NICE Guideline 72 (ADHD) [H6] Getting Help: Guided by NICE Guideline 113 (GAD and/or Panic Disorder) [H7] Getting Help: Guided by NICE Guideline 158 (Antisocial Behaviour and/or Conduct Disorders) [H8] Getting Help: Guided by NICE Guideline 159 (Social Anxiety Disorder) [H9] Getting Help: Guided by NICE Guideline 170 (Autism Spectrum) [H10] Getting Help: With Co-occurring Behavioural* And Emotional** Difficulties [H21] Getting Help: With Co-occurring Emotional** Difficulties [H22] Getting Help: With a Difficulty or Co-occurring Difficulties Not Covered by H1-H10, MH1-MH3 or H21-H22 [H23] Getting More Help: Guided by NICE Guideline 9 (Eating Disorders) [MH1] Getting More Help: Guided by NICE Guideline 155 (Psychosis) and/or Guideline 38 (Bipolar Disorder) [MH3] Getting More Help: With Co-occurring Difficulties of Severe Impact [MH9] Percentage of periods of contact in sample 30 % 3% 6% 2% 6% 1% 1% 6% 4% 5% 2% 2% 2% 8% 16 % 1% 1% 8% Total sample size: 4573 periods of contact in the community (i.e. does not include inpatient periods of contact) from 11 CAMH services. Data were collected between September 2012 and June 2014. Current View tools were usually completed after the first contact within a period of contact. Percentages sum to more than 100%, because each group has been rounded to the nearest whole percentage, and because a child can be in the grouping Getting Advice: Neurodevelopmental Assessment (A2) at the same time as being in one of the other groupings. Apart from A2 all other groupings are mutually exclusive. * Behavioural difficulties (Conduct Disorder or Oppositional Defiant Disorder). ** For the purpose of grouping assignment emotional difficulties are defined as: Depression/low mood (Depression); Panics (Panic Disorder); Anxious generally (Generalized anxiety); Compelled to do or think things (OCD); Anxious in social situations (Social anxiety/phobia); Anxious away from caregivers (Separation anxiety); Avoids going out (Agoraphobia); Avoids specific things (Specific phobia).

Illustrative indication of distributions of appointments by grouping based on analysis of Current View and appointments data collected Sep 2012-June 2014 Total sample size: 4573 periods Number of appointments attended between period of contact start date and discharge date inclusive (binary log scale) of contact in the community (i.e. does not include inpatient periods of contact) from 11 CAMH services. Data were collected between September 2012 and June 2014. Current View tools were usually completed after the first contact within a period of contact. Key to box plots <- Upper quartile <- Median <- Lower quartile A1 H6 H10 H5 H8 H3 Legend to grouping labels A1: Getting Advice: Signposting and Self-management Advice H1: Getting Help: Guided by NICE Guideline 16 and/or Guideline 133 (Self-harm) H2: Getting Help: Guided by NICE Guideline 26 (PTSD) H3: Getting Help: Guided by NICE Guideline 28 (Depression) H4: Getting Help: Guided by NICE Guideline 31 (OCD) H5: Getting Help: Guided by NICE Guideline 38 (Bipolar Disorder) H6: Getting Help: Guided by NICE Guideline 72 (ADHD) H7: Getting Help: Guided by NICE Guideline 113 (GAD and/or Panic Disorder) H8: Getting Help: Guided by NICE Guideline 158 (Antisocial Behaviour and/or Conduct Disorders) H9: Getting Help: Guided by NICE Guideline 159 (Social Anxiety Disorder) H10: Getting Help: Guided by NICE Guideline 170 (Autism Spectrum) H7 H4 H2 H1 Draft groupings H9 H21 H23 H22 MH1 MH9 MH3 H21: Getting Help: With Co-occurring Behavioural* And Emotional** Difficulties H22: Getting Help: With Co-occurring Emotional** Difficulties H23: Getting Help: With a Difficulty or Co-occurring Difficulties Not Covered by H1-H10, MH1-MH3 or H21-H22 MH1: Getting More Help: Guided by NICE Guideline 9 (Eating Disorders) MH3: Getting More Help: Guided by NICE Guideline 155 (Psychosis) and/or Guideline 38 (Bipolar Disorder) MH9: Getting More Help: With Co-occurring Difficulties of Severe Impact * Behavioural difficulties (Conduct Disorder or Oppositional Defiant Disorder). ** For the purpose of grouping assignment emotional difficulties are defined as: Depression/low mood (Depression); Panics (Panic Disorder); Anxious generally (Generalized anxiety); Compelled to do or think things (OCD); Anxious in social situations (Social anxiety/phobia); Anxious away from caregivers (Separation anxiety); Avoids going out (Agoraphobia); Avoids specific things (Specific phobia). N.B. Getting Advice: Neurodevelopmental Assessment (A2) is not shown as additional appointments that may be associated with this grouping cannot be discerned in this data set.

Testing the relevance of complexity factors (etc.) for predicting resource use

Choosing clusters via shared decision making Hypothesised Need Care package Possible elements algorithm (based on current view tool) Example of shared decision potentially overriding algorithm Adjusting to life circumstances Signposting No problem rated more than mild Severe difficulties but choose self management A single problem on CV form rated moderate that does not fit any NICE guidance Concern about depression agree to wait Temporary or mild difficulties Managing chronic difficulties Selfmanagement support Choice appointment 28

Choosing outcome focussed via shared decision making 29

CAMHS Future? Increased focus on coping and resilience More explicit about limitations and dangers of intervention Greater emphasis on shared standards and consideration of routine data across systems Greater clarity of aim of any intervention in terms of needs being met

Contact Miranda.wolpert@ucl.ac.uk EBPU@annafreud.org CORC@annafreud.org