C PPP P Care Pathways and Packages Project Developing currencies for mental health payment by results Learning Disabilities Payment System Development 18 th June 2013 1
June 10 Work initiated in NTW Oct 10 CPPP LD subgroup established Nov 11 1st national sub group meeting Nov 12 Bangor s cluster analysis completed June 13 Clinical validity check completed Timeline Next Steps July 10 Bangor University consulted July 11 CPPP initial report Aug 12 Data collection completed April 13 Clinically focussed analysis completed June 2013 Final report
Service Users Academics Local Authorities Health Care Professionals of all disciplines Focus Groups Roundtable discussions Involvement Next Steps Carers Inclusion North Commissioners On-line surveys Workshops Meetings
Phase 1 18 Provider Trusts Phase 2 11 Provider Trusts Scale Next Steps 2825 service user assessments 829 service user assessments
Social Care Specialist Health Care Scope
Overall Aim Identify whole population need (Specialist Learning Disabilities Healthcare) Identify groups of need (Purchase units) Identify suites of interventions (Care packages) Produce Quality and Outcome Framework Integrated personalised approach (Individualised health and social care)
Initial Aims Identify needs of people Draft additional Rate a large accessing scales to create number of specialist an integrated users with the health services MH& LD tool not captured Allocation Tool by the MH Tool Assess the tool s performance Produce meaningful units Integrate new LD units into mandated units and re-test
Overview of the process Draft preliminary MHLDT & Units Staff questionnaire Statistical analysis of tool Data collection Cluster analysis User & carer feedback Clinical adjustments Re-test Final report 8
Recap: initial data collection & statistical analysis Figure 1: Distribution of cases across preliminary cluster groupings Total cases n= 2825
Level of LD (%) across mandated and LD units 10
Distribution across preliminary units 400 350 300 250 MHCTs 200 150 100 50 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 99 A3 Units Total cases n= 2825
Dendrogramfrom statistical analysis A diagram summarising how cases group together at different levels of distance (distance is standardised onto a new scale) Used to identify the number of clusters to define in the 2nd stage of cluster analysis
Statistically generated groups (n=5) Statistical group letter [size] and short description Relatively high scoring items Relativelylow scoring items A[N = 256] ASD, aggression, communication limitations, otherwise cognitively relatively able, low physical problems A F G H 5 B[N = 254] Profound LD, physical health problems, low CB/MH problems 4 5 10 18 E G H I 1 7 8 9 17 A B C C [N = 172] Severe LD, ASD, relatively high levels of CB and MH needs 1 7 8 9 10 11 17 A B C D E F D [N = 399] Mild LD with relatively low levels of need 4 10 17 C E F G H I E[N = 220] Mild LD, SIB/self-harm, others at risk/vulnerable 17 B C E G H
Clinicaladjustment process
Typical resultsof clinical adjustment process Statistical grouping Statistical grouping Clinical request Clinical request Clinical request Clinical request Resulting purchase unit Resulting purchase unit Resulting purchase unit
Final coverage and overlap of proposed LD Units 9e = 70 22 n=30 9a = 368 9b = 39 9c = 156 2 9d=121 9f=73 3 23 n=107 24 n= 87 Total cases n=1256 coverage (85.4%)
Draft scoring profiles & descriptive fields
Proposed Learning Disabilities Unit Titles 9A Maintenance, engagement and minor support needs, complicated by LD 9B Risk to self, complicated by LD 9C Risk to others complicated by LD 9D Risk to others, complicated by mild LD & ASD 9E Risk to others, complicated by moderate - profound LD & ASD 9F Risk to others & self, complicated by moderate - profound LD & ASD 22 Physical health complicated by mild LD 23 Physical health complicated by moderate - profound LD 24 Physical health with dysphagia complicated by moderate - profound LD 18
Clinical re-test Figure 5: Pies chart showing reallocation of cases to final learning disabilities and mental health clusters Total cases n= 829
Distribution across proposed units Total cases n= 829
Changes in unit allocations Figure 5: Pies chart showing reallocation of cases to final learning disabilities and mental health clusters
Changes in unit allocations Figure 6: Bar chart comparing proportion of clusters allocated by rules for allocation and clinical decisions
Changes in unit allocations Graph 5 shows the number of cases from the re-submission that were allocated to each of the mandated MH clusters.
Preliminary clinical feedback Wide ranging views Very positive to very negative We asked for all the potential problems That s what we got! Really valuable when testing different possibilities to know people s concerns and preferences
Final clinician feedback % for each rating Cluster 1 (Poor Fit) 2 3 4 5 (Good fit) 9a n = 78 7.7 12.8 19.2 29.5 29.5 9b n =47 6.4 36.2 44.7 10.6 9c n=83 2.4 3.6 19.3 51.8 18.1 9d n=69 1.4 1.4 24.6 60.9 11.6 9e n=32 3.1 34.4 31.3 31.3 9f n=112 2.7 17 58.9 14.3 22 n=56 3.6 1.8 26.8 53.6 12.5 23 n=126 3.2 11.9 31.7 50.8 24 n=58 6.9 10.3 36.2 43.1 Overall 4.6 4.3 18.2 40.4 27.1 Overall the mean rating = 3.85, mode = 4.
User & carer feedback Facilitated by Inclusion North (Scott Cunningham) Four workshops (Middlesbroughx2; Newcastle; Sheffield) 24 people with learning disabilities; 8 family members and family carers; 20 staff members Aims Understanding how the process works Saying whether they thought there are any opportunities from this process. Saying if there is anything people are worried about Similar questions, worries and big messages across all workshops 27
The Big Messages Lots of the things people talked about were about how we get a good balance between using the money in the best way we can whilst still treating people as citizens with a right to a good service. Lots of people thought it was a good idea to be able to say Lots of people thought it was a good idea to be able to say how much money things would cost and that people should be able to know how much they are entitled to. At the same time people felt that this should never just be about the money but about making sure people get access to good quality specialist health services if they need it that supports them to be citizens of their community.
Conclusions Six new scales to create an integrated MH&LD tool Nine new LD units to create an integrated MH&LD needsled classification system Acceptable statistical properties and clinical face validity Compatible with other work streams Unchanged allocation process for MH units Potential to link to forensic MH units (forensic LD) Suitable as the basis of a payment system for LD linked to a robust Q&O framework Potential to help reduce inpatient use, support personalisation & joint commissioning
Recommendations Continued use, evaluation and refinement of units Use of forthcoming MHLDDS to capture costing information and assess resource homogeneity Develop care packages linked to best-practice Create robust Quality & Outcomes framework / link to existing Outcomes Frameworks Continue to engage with clinicians, LAs, Users, Carers and other stakeholders