TRIAGE AND PATHWAYS IN MEDIUM SECURE CARE Harry Kennedy Executive Clinical Director National Forensic Mental Health Service Central Mental Hospital Dundrum, Dublin and Clinical Professor of Forensic Psychiatry Trinity College Dublin Forensic Conference 1-3 February 2012
Co-researchers • Conor O’Neill • Kim McDonnell • Grainne Flynn • Paul Braham • Clare McInerney • Darran Flynn • Mary Davoren • Damian Smith • Zareena Abidin • Andrea Nulty • Damian Mohan • Norah Byrne • Pauline Gill • Sally Linehan • David Timmons • Brenda Wright • Leena Naughten • Helen O’Neill • Olivia Gibbons • Charles Smith • Ronan Mullaney • Stephen Monks 2 Forensic Conference 1-3 February 2012
Why study triage at all? • 1. Triage – who should we admit to high secure, medium secure, low secure or open / community services, and why? • 2. Urgency - on a waiting list of people qualified for admission, who should be given priority? • 3. When is a patient ready to move to a less secure place, or to the community? – 3. How do we know that a patient has successfully completed relevant treatments to address the reason they were admitted to a given level of therapeutic security? – 4. What does ‘recovery’ mean in a forensic context e.g. ‘conditional discharge’ or CTO? 3 Forensic Conference 1-3 February 2012
Drivers • Policy and rights – least restrictive option • Transparency – MHTs, Habeas Corpus, ECHR. • Commissioning – quality: consistency, reliability, criterion setting. • Validity – what do we teach, what do we do. 4 Forensic Conference 1-3 February 2012
What do consultant forensic psychiatrists do? •Admission assessments / decisions, •movement along the recovery path •Discharge recommendations / decisions •What do we teach SpRs? •Have we any evidence for any of what we do? •How do we study what we do? •RCTs, ethics committees, AUCs, linear regression…… 5 Forensic Conference 1-3 February 2012
Admission triage • Who should be admitted, what level of therapeutic security does the patient need? • Several existing structured professional judgement instruments, ACSeSS (1998), Coid & Khatan (2000), STD (2001), OPRISK (2002), HONOS-S (2004), SNAP (2005). • Risk not considered • Two elements – security need and urgency. • Decisions are made in the context of a system and a pathway – prison screening systems, – court liaison/diversion, – pre-admission assessment and 6 – admissions panels Forensic Conference 1-3 February 2012
DUNDRUM-1 triage security • Seriousness of violence • Seriousness of self-harm • Immediacy of risk of violence • Immediacy of risk of suicide/ self harm • Specialist forensic need • Absconding / eloping • Preventing access • Victim sensitivity/public confidence issues • Complex Risk of Violence • Institutional behaviour • Legal process α = 0.949 7 K > 0.85 Forensic Conference 1-3 February 2012
DUNDRUM-1 triage security SCORE DUNDRUM-1:TRIAGE SECURITY ITEMS 0 1 2 3 4 S1 Seriousness of violence S2 Seriousness of self-harm S3 Immediacy of risk of violence S4 Immediacy of risk of suicide/ self harm S5 Specialist forensic need S6 Absconding / eloping S7 Preventing access S8 Victim sensitivity/public confidence issues S9 Complex Risk of Violence S1 Institutional behaviour 0 S1 Legal process 1 4 high medium 3 PICU 2 Open wards 1 Independent / 0 community Forensic Conference 1-3 February 2012
Three months Jan 2008 to Dec 2009 921 screened on committal – Birmingham & Grubin Prison-based court diversion – Ensures full screening of at risk population 246 assessed after screening 159 GP follow up 57 psych follow up in prison 30 admitted hospitals / CMHTs 9 Forensic Conference 1-3 February 2012
DUNDRUM-1 Prison triage ANOVA F=360.1/2/0.001 10 Forensic Conference 1-3 February 2012
DUNDRUM-1 Triage Security score April to June 2009, those not transferred from prison to hospital (n=216) v transferred from prison to any hospital (n=30) Area Under the Curve =0.984 (95% confidence interval 0.971 to 0.997) At threshold score 6, sensitivity =0.95, sensitivity=0.92 11 Forensic Conference 1-3 February 2012
Two years Jan 2008 to Dec 2009 7454 screened on committal – Birmingham & Grubin Prison based court diversion – Screens full population at risk 1454 assessed after screening GP follow up psych follow up in prison 27 divert to hospital / CMHT 26 divert to PICU 47 admitted to MSU / HSU 12 Forensic Conference 1-3 February 2012
DUNDRUM-1 Prison triage ANOVA F=75.2/2/0.001 13 Forensic Conference 1-3 February 2012
DUNDRUM-1 Triage Security score Two year period 2008 to 2009, those transferred to open wards (n=27) compared to those transferred to psychiatric intensive care units (n=26) Area Under the Curve =0.805 (95% confidence interval 0.680 to 0.930) At threshold score 13, sensitivity 0.78, specificity 0.71. 14 Forensic Conference 1-3 February 2012
Figure 4 - DUNDRUM-1 Triage Security score Two year period 2008 to 2009, those transferred to psychiatric intensive care units (n=26) compared to those transferred to forensic medium and high security at Central Mental Hospital (n=47) Area Under the Curve =0.866 (95% confidence interval 0.784 to 0.949) At threshold score 20, sensitivity=0.73, specificity=0.83 15 Forensic Conference 1-3 February 2012
D-1 triage security: item to outcome nil v open v PICU v any adm PICU MSU AUC AUC AUC 1.Seriousness of violence 0.915 0.722 0.644 2. Seriousness of self-harm 0.515ns 0.568ns 0.601ns 3. Immediacy of risk of violence 0.961 0.693 0.644 4. Immediacy of risk of suicide/ self harm 0.546ns 0.573ns 0.609ns 5. Specialist forensic need 0.973 0.786 0.695 6. Absconding / eloping 0.930 0.860 0.726 7. Preventing access 0.905 0.825 0.670 8. Victim sensitivity/public confidence 0.806 0.775 0.690 9. Complex risk of violence 0.767 0.762 0.596 10. Institutional behaviour 0.907 0.698 0.599 11. Legal process 0.945 0.927 0.969 16 Forensic Conference 1-3 February 2012
EXACT AGREEMENT BETWEEN ITEMS AND DISPOSALS Observed 95% CI of Spearman rank proportion in observed correlation coefficient b Agreement proportion in (n=316) a agreement 0.70 – 0.80 TS1: serious violence 0.75 0.803 0.56 – 0.67 TS2: serious self harm 0.61 0.259 0.70 – 0.80 TS3: immediacy of violence risk 0.75 0.879 0.62 – 0.73 TS4: immediacy of self harm risk 0.67 0.236 0.74 – 0.83 TS5: specialist forensic need 0.78 0.908 0.76 – 0.85 TS6: absconding risk 0.80 0.879 0.74 – 0.83 TS7: preventing access 0.78 0.831 0.76 – 0.85 TS8: victim sensitivities 0.80 0.806 0.67 – 0.77 TS9: complex risks 0.72 0.828 0.66 – 0.76 TS10: institutional behaviour 0.71 0.758 0.90 – 0.95 TS11: legal procedure 0.92 0.921 17 Forensic Conference 1-3 February 2012
How do we decide to move patients from high to medium, medium to low secure and to the community? Acute Cluster/High Secure SABU 3:1 Male admission 2:1 PRISON AND COURT DIVERSION Medium Cluster/Medium Secure Medium Medium 1.4:1 1.1:1 R&R Slow stream Pre- Rehab TEAM Rehab discharge 0.7:1 0.7:1 0.4:1 High support Community residence Forensic Conference 1-3 February 2012
Dynamic risk ITEM ITEM DESCRIPTION Scoring C1 Lack of insight C2 Negative attitudes C3 Active symptoms of major mental illness C4 Impulsivity C5 Unresponsiveness to treatment ITEM ITEM DESCRIPTION Scoring ‘IN’ ‘OUT’ R1 Plans lack feasibility R2 Exposure to destabilisers R3 Lack of personal support R4 Non-compliance with remediation attempts R5 Stress Mueller-Isberner, Webster, Gretenkord 2007 Dolan & Blattner 2010 19 Forensic Conference 1-3 February 2012
TREATMENT PROGRAMMES: FIVE PILLARS Pillars of Care Treatment Programmes Drugs and Harmful Physical Mental Psychosocial Health Alcohol functioning Health Behaviours Information Psycho- Healthy Eating Offence wellness Education Physical related Vocational & Abstinence Wrap Activity Therapy Occupational Social & National MCT Victim Coping Screening Empathy Social & Programmes Skills Grow Social Role Recreational Self Primary Health Modelling Managemen GP Self Risk Family t Management Therapy Optician AA Needs strengths and risks Information treatment and self -management Individual Care Plan Achievable Goals 20 Forensic Conference 1-3 February 2012
DUNDRUM-3 programme completion • Physical health Not ready for a move down 4 • Mental health ready for a move • Drugs and Alcohol e.g. from high to medium security 3 • Problem behaviours • Self-care and activities of ready for a move e.g. 2 from medium to low security daily living • Education, Occupation and ready for a move to 1 Creativity supported community living 3: ready for a move • Family and Social Networks 0 O: ready for 4: not ready to move ready for independence down a level of security independence Forensic Conference 1-3 February 2012
Recommend
More recommend
