1. Being young, male and experiencing first psychosis Max Birchwood www.youthspace.me

2. NIHR SUPEREDEN programme grant Sustaining Positive Engagement and Recovery The next step after Early Intervention for Psychosis Lead: Birmingham and Solihull Mental Health Foundation Trust Cambridgeshire and Peterborough NHS Foundation Trust EIS Cheshire and Wirral Partnership NHS Trust EIS Lancashire NHS Partnership Trust EIS Norfolk & Waveney Mental Health Partnership Trust EIS Devon and Cornwall Partnership Trust EIS University of Birmingham University of Bristol University of Cambridge University of East Anglia University of Manchester University of Warwick King's College London

3. Birmingham 5 teams (Birchwood/Lester) Lancashire + Wirral 5 teams (Marshall/Lewis/Sharma) East Anglia 4 teams (Jones/Fowler) Cornwall 2 teams (Amos) The National/SUPER EDEN sites

4. “Roughly half of all lifetime mental disorders in most studies start by the mid-teens and three quarters by the mid-20s. Later onsets are mostly secondary conditions. Severe disorders are typically preceded by less severe disorders that are seldom brought to clinical attention” Kessler et al, Current Opinion Psychiatry, 2007

5. “Mental disorders are the chronic diseases of the young” Insel TR, Fenton WS. Psychiatric epidemiology: it's not just about counting anymore. Arch Gen Psychiatry. 2005; 62(6): 590-2.

6. Slide courtesy of Patrick McGorry

7. The psychoses

9. Black Caribbean 49% White British 20% p < 0.001 Indicators of adversity in early adulthood and risk of psychosis late adulthood Morgan, Kirkbride, Hutchinson et al., 2008

10. The affect in non-affective psychosis

11. symptom severity time prodrome psychosis treatment DUP ‘Affect-symptom gap’ Affect

12. N=318; 31%N=709; 69%

13. Male gender and early trajectories of.. Social functioning and productive use of time; links with later ‘NEET’ status Harm to self and others Emotional and affective functioning

14. Why are early trajectories of social functioning important?

15. THE CRITICAL PERIOD Early trajectories predict long term trajectories The plateau effect: ceiling of disability/symptoms early in manifest course (Bleuler) Adolescent social functioning best predictor of early phase social functioning “Early phase of psychosis is a stormy one,plateauing thereafter” From : Birchwood,M and Macmillan,JF (1993) Early intervention in schizophrenia Australia & New Zealand Journal of Psychiatry 27 374-8

16. What do the early trajectories of social functioning look like; and links with male gender?

17. Identifying trajectories of social recovery. Latent Class Growth Analysis (LCGA) is a technique for identifying distinct homogenous subpopulations with similar trajectories of growth over time (known as latent classes) within longitudinal data collected from a larger heterogeneous population (Jung and Wickrama, 2008). The analyses were conducted using Mplus version 4 (Muthen and Muthen, 1998).

21. NB. On average, a community sample of 16-35yrs, spends 63.5 hrs/week in constructive activity

25. Poor and stable social functioning, is common in the early phase, even with ‘best care’, disproportionately affects young males, and has its origins in early adolescence.

26. But does it matter?

27. Does it affect valued life opportunities?

32. Female Male NEET 112 325 437 48.5% 61.6% 57.6% EET 119 203 322 51.5% 38.4% 42.4% Total 231 528 759 NEET and gender at 12 months: UK National EDEN findings.

33. Those with low, stable activity are at very high risk of NEET

34. Low and unchanging social disability is characteristic of a substantial proportion receiving early intervention services (‘non-responders’) It’s largely confined to young males with psychosis and stretches back into early adolescence. This group need special attention as they are at very high risk of NEET (and therefore LT social exclusion). Summary

35. Gender and harm to self and others

36. Why do individuals act on their delusions and others resist? What are the developmental pathways to harm to self or others? Why are there no interventions to reduce harm vs treating psychosis? Are they the same thing? Can we prevent such behaviour?

37. Winsper et al. (2013) JAMA Psychiatry, 70 (12) 1287-1293

38. Background Rates of harm/aggression prior to FEP do not appear to substantially decrease following service contact (Winsper et al. 2013) Mirrors concerns that treatments for psychosis do not tackle harm risks (Serper et al., 2011)

39. It has been hypothesized that there are 3 groups of individuals with psychosis with harm risks: 1.The early starters display a pattern of antisocial behaviour emerging in childhood, which remains relatively stable across the lifespan. 2.An illness onset group displays no antisocial behavior prior to illness, then repeatedly engages in aggressive behavior. 3.A second illness onset group displays no antisocial behaviour prior to and for the first few decades of illness, then commits serious harm. Hodgins S. Violent behaviour among people with schizophrenia. Philos Trans R Soc Lond B Biol Sci. 2008;363(1503):2505-2518.

40. In the Dunedin prospective study: 40% of individuals who developed schizophreniform disorder by age 26 years displayed conduct disorder prior to the age of 15.1 (p<0.001) Kim-Cohen J, Caspi A, Moffitt TE, HarringtonHL,Milne BJ, Poulton R. Prior juvenile diagnoses in adults with mental disorder. Arch Gen Psychiatry.2003;60(7):709-717. 12. Hodgins S, Cree A,

41. Research Questions Are there distinct subgroups of FEP young people differing in premorbid anti-social behaviour patterns? Do these subgroups differ in prevalence of harmful behavior following EIS entry? What are the direct and indirect (via mediators, e.g., positive symptoms) associations between premorbid delinquency and violent behavior?

42. Assessments Outcome: harm behaviour during EIS contact – “Adverse Outcomes Screening Questionnaire” – Dichotomous outcome (0=no violence; 1=violence at 6 or 12 months). Shortened version of the MacArthur study questionnaire. Main predictor: Premorbid anti-social behaviour – Premorbid Adjustment Scale (PAS) (“adaptation” subscale) – Continuous measure at baseline referring to: childhood, early adolescence, and late adolescence

43. Assessments: Confounders and mediators Past Drug Use – Continuous measure at baseline (0:no past drug use; 1:not more than 3 times; 2:less than weekly; 3:1 to 3 times weekly; 4: almost every day) Duration of untreated psychosis – Dichotomous measure (0:less than 6 months; 1: more than 6 months)

44. Assessments: Confounders and mediators Age of illness onset – Continuous measure reported at baseline Positive symptoms – Positive and Negative Syndrome Scale (PANSS) – Continuous measure reported at 6 months

45. Methods: 3 stages Latent Class Growth Analysis: LCGA (Question 1) – To group individuals according to patterns of delinquent behavior across time from childhood to late adolescence Logistic Regressions (Question 2) – To assess unadjusted associations between delinquent groups (identified in the LCGA) and violent behavior during EIS contact Path Analysis (Question 3) – To assess direct and indirect (via possible mediators, e.g., positive symptoms) associations between delinquent groups and violent behavior

46. Self- and other- post FEP 13.7% at 6 or 12 months 8.6% at 6 months; 8.5% at 12 months >80% male

47. Trajectories of Premorbid Anti-social behaviour (LCGA) Assessed using the “adaptation” subscale of the PAS 48% 28.7% 9.7% 13.2%

48. Results 2: Logistic Regressions Stable moderate adolescent anti-social behaviour significantly increased risk of later harm behavior: OR=1.97 (95% CI=1.12-3.46)* Stable high anti-social behaviour most strongly increased risk of violent behavior: OR=3.53 (95% CI=1.85-6.73)* * Stable low delinquency used as the reference group. These associations are unadjusted

49. (In comparison to low delinquency group) Stable moderately delinquent group significantly more likely to be male: OR: 1.81 (1.22, 2.69) (In comparison to low delinquency group) Stable highly delinquent group significantly more likely to be male: OR: 2.36 (1.41, 3.95) Male gender and trajectories of harmful behaviour

50. Summary 1.Stable high anti-social behaviour in adolescence independently increased risk of later harm behaviour 2.Males at highest risk 3.Stable moderate antisocial behaviour only increased risk of violent behavior via positive symptoms (there was no direct association)

51. Affective function and gender

52. Depression in 80% at one or more phases ‘Prodromal’ (adolescent) depression predicted acute and post psychotic depression

53. Depression by gender (p<0.001, n=736)

55. First episode Psychosis N=80 Social phobia (non-psychotic) N=31 No social Phobia N=60 Social Phobia N=20 Healthy controls N=24 Michail & Birchwood, BJP,2009

56. Social anxiety disorder in FEP shows female excess, similar to non-psychosis SAD.

57. Social interaction anxiety scale (p<0.001)

58. Social Phobia Scale (p<0.001)

59. In conclusion Psychoses are predominantly male with onsets stretching into early adolescence Young males prone to severe social disability developing in adolescence and continuing into early adulthood post ‘first episode’. This group at v high risk of NEET. Anti-social and harmful behaviour more prevalent in males and can persist post-onset. But de novo behaviour linked to acting on delusions. Depression and social anxiety prevalent at all stages, but LESS prevalent in males.