Workshop: use of routine outcome monitoring in assertive outreach NFAO conference, University of Warwickshire, 13/7/10 Rob Macpherson, consultant psychiatrist Mike Blackburn, Gloucester AO team manager & SW NFAO rep Gloucester Assertive Outreach Team, 2Gether NHS Partnership Trust
Current team was reformed in 2006 key changes Shared vision discipline specific roles team caseload systems approach Multidisciplinary Meeting Supervision Daily meetings Appraisals Roles with the team
Team structure Team secretary Team manager 5 Nurses (band 6 x 4, band 5 x 1) 2 Occupational Therapists (band 6 & 5) 2 Social Workers both AMHP.5 Health and Wellbeing Practitioner 3 Support Workers 1.2 Consultant Psychiatrists 1 Associate Specialist 1 SPR 1 SHO.4 Psychologist team caseload - 81
Roles within the team Users champion Carers champion Homeless Dual diagnosis / Substance misuse Family work Child protection Hearing voices MHA, CTO, Capacity, Court of Protection, Appointeeship Best interest assessor Employment &Vocation Activity Program Medication management Physical Health & Wellbeing
Developing opportunities Journey 2 Work 3 Gym groups Allotment Snooker / pool group Women's group Swimming group Tennis Indoor racquets Walking groups Service user holidays Decking group Voluntary work Supported employment Art Shape Gloucester college
Need to constantly evaluating what we do
Questions I How many colleagues have a service evaluation/research project in their teams
Questions II How many teams use routine outcome measurements in all/most cases
Questions III Any other ways of measuring outcomes in teams
Background: National Guidelines/policy NICE guidelines CHAI survey National confidential inquiries AIMS, STAR wards
Local policy/practice Local information systems-?data quality Local audits Critical incident reviews Use of rating scales: LUNSERS, DAI, BPRS, QOL, CANSAS, PANSS, AIMS, REHAB, HONOS, CFI, KGI, EM, CGI, CPRS,HAMILTON, BECK, HAD, SFS, Recovery star
Background literature Systematic review of use of QOL/needs rating (Gilbody et al, 2002)- no impact on psychological functioning - results not used clinically -but clinicians welcomed new information RCT standardised needs assessment & care planning (Marshall et al 2004): pt satisfaction increased, no other change
Background literature Fonagy (2005) Scottish Schizophrenia Outcomes study: 10% (>1000) of all patients with Schizophrenia followed up for 4 years. Outcomes: routine data collection possible: Reduced hospitalisation, suicide, self harm: level of impairment on HONOS increased. Kiseley et al (2008): in Nova Scotia: HONOS used in all new out patient assessments. Achieved 86% completion. Factors to increase clinician support: 1, Feedback of individual & aggregated (caseload) data to clinicians 2, Produced tabulated glossaries (domains in rows, scores in columns)
Background literature- 4 steps to for routine outcome measurement 1, realism: time, resources, leadership, expertise, IT. Do not start if not in place 2, identify outcome domains: preferably include service user and staff priorities 3, context. Who is the assessment for, will results be aggregated, timescale for measurement 4, Identify outcome measures closest to requirements. Slade et al (2003) advised use CANSAS (to aid care planning) & HONOS (to identify level of need) in routine outcome measurement
Policy UK DoH (2001): recommends routine outcome measures to evaluate interventions with service users CSIP report (2006): recommended measuring outcomes, guidance to use HONOS as part of national minimal data set for all on enhanced CPA NIMHE outcome measures review (2008) Compendium produced of 188 instruments, scored on quality, with guidance re use
HONOS PBR Care Clusters DECISION TREE (RELATIONSHIP OF CARE CLUSTERS TO EACH OTHER) A Non- Psychotic Working-aged Adults and Older People with Mental Health Issues B Psychosis C Organic a. Mild/Mod/ Severe b. Very Severe And Complex c. Substance Misuse a. First Episode b. Ongoing or Recurrent c. Psychotic Crisis d. Very Severe Engagement a. Cognitive Impairment 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Service user perspective Service users want- a job live -a relationship -somewhere decent to Medical goals less valued (ie, fewer symptoms) Who chooses the outcome measures?
Aims of routine outcome measurement Questions 1, who is it for? Service user, professional, team, commissioners? 2, Are you assessing - workload (caseload profile tool) - team working (Dartmouth fidelity scale) -service user satisfaction/experience (CUES/ QOL scale) - a specific aspect of the team s work- may be audit of standards, service evaluation or research 3, If assessing service user outcome, are you rating - psychopathology -social functioning -need -everything
Examples 1 Gloucester AOT Pan London (Priebe et al, 2003) Dartmouth score 4.3? Average age 38 37 Unemployed 82% 80% Living alone 41% 52% Av illness duration 15 yrs 15 yrs Compulsory admission 23% over 1 yr 25% over 9 mths Police contact 34% over 1 yr 21% over 9 mths Admission rate 31% over 1 yr 39% over 9 mths Substance misuse 29% 29%
Carer s needs Examples 2 36/75 patients on caseload have carer (NSF definition = min 12 homes/week support). Carers rate higher unmet needs around care of home, lower ratings on risk to others. Carers assessments completed for all carers reported had given up careers (29%), independence (18%) - problems tiredness (36%), unpredictability of patient (29%) - need someone to talk to, practical help (finance/break/respite).
Examples 3 Audit Vs NICE guidelines in schizophrenia All keyworkers completed audit proforma for each of 61 cases of schizophrenia in ACT team. Compliance with guidelines: Formal family intervention 20% (51% no contact, 10% declined, 15% other family work) CBT 40% (23% unable to participate, 18% no persisting symptoms) Advance directives 0% Antipsychotic Polypharmacy 15% (others: 5% patient choice, 3% Clozaril augmentation, 7% reduction of AP caused relapse).
Service Evaluation of Assertive Outreach Teams (SEAT): Aims Prospective assessment of outcome of all cases taken on by three GPT ACT teams over 12 months Steering group with cons/team leaders/r&d/ carer/user representation. All welcome
SEAT Method All cases assessed at/before entry to team. Then at 6 & 12 months Consent not required. Service evaluation so ethics submission not needed Measures: uptake of health services contact with CJS CANSAS: staff + pt + carer (where appropriate) HONOS Engagement measure
SEAT: Method + dissemination Research assistant to co-ordinate & work with teams to ensure ratings carried out on time. Key role of consultants & team leaders. Data inputted into SPSS: develop a database Analyse changes in ratings At 6 & 12 months. Timescale 2-3 years Dissemination: present to teams, WAA & GPT Board, write up & publish.
Thanks for listening rob.macpherson@glos.nhs.uk