NRAC (2006) 32 NHSSCOTLAND RESOURCE ALLOCATION COMMITTEE FHS – PROGRESS TO DATE AND NEXT STEPS 1. FHS formulae – the story so far 1. Arbuthnott review did not consider FHS formulae due to resource constraints. However this was picked up by SCRA (2001-3) who formed a sub-group to take this work forward. 2. SCRA proposed that research be commissioned to develop allocation formula for dental (GDS), pharmacy (PS) and ophthalmics (GOS) and three separate research specs were drawn up. The specs were prepared in 2002 and make no mention of any impending new contracts or contract changes. 3. The initial response to the invitation to tender was disappointing and it was decided that the work would be taken forward within SEHD. This proved unrealistic due to resource pressures and so they went back to the best of the initial bids (Deloitte) to discuss their proposal. This resulted in a delay in starting the work which finally commenced in October 2004 with a 6-month deadline. 4. An internal Advisory group was set up consisting of members of SEHD and ISD and this met on 4 occasions with Deloitte. 5. Deloitte completed 3 research reports by August 2005 covering each area separately. They interpreted their brief fairly tightly concentrating on producing final formulae for all 3 areas using the same basic structure as used in the Arbuthnott formula (population, age-sex, MLC, excess costs), although the specs did not make constraints on methods. The main structural difference was the inclusion of cross-boundary flow adjustments as requested in the specs. 6. Deloitte presented their work to NRAC in August 2005. NRAC agreed to take over this work from the Advisory group as an element of their remit is to “advise on possible formulaic approaches to the parts of health expenditure not currently covered by the Formula (e.g. primary care dental, pharmaceutical and ophthalmic services)”. 7. In the reports Deloitte recognised the limitations of the methods used and set out caveats to the results, including recommendations for further work. These can be found in the „Conclusions and Recommendations‟ chapter of each report. 8. NRAC decided to put the 3 reports out as part of a targeted consultation aimed mostly at NHS Boards and some other experts in Dec - Jan 06. 9. Twenty-one responses were received on the reports from NHS Boards and professional organisations, with concerns summarised as follows (see paper 2006-16). There was a generally positive response to adopting a formulaic approach to allocating funding from most consultees. However most had concerns with these particular formulae which could be summarised as follows: Timing – introducing formulae at a time when contracts are changing. Data quality and availability. More research is needed on some aspects of the formulae Specific elements of the formula – MLC adjustment, what to do about unmet need, the issue of cross boundary flow, deprivation in rural areas. Financial issues – using a formula to distribute cash limited funds, how will Argyll and Clyde issue be dealt with. 9. Issues related to FHS have been discussed at NRAC meetings as follows: Policy context – papers 2005-37 and 2005-47 discuss the current policy context and obligations on Boards as well as policy changes in line for the FHS services. Private sector – papers 2005-56, 2005-64 discuss private sector issues in relation to the arbuthnott formula and conclude that issue deserves special attention in relation to FHS. Cross-boundary flows – paper 2005-65 discusses population bases and concludes that for FHS should be based on legal requirements. Consultation – the responses from the targeted consultation process were summarised in paper 2006-16. Core criteria - the formulae as a whole have been judged against the NRAC core criteria (paper 2006-17) and found to partially meet those criteria (e.g. some met, some partially, some not). Relative merits - the relative merits of the components of each formula have been assessed (paper 2006-27). This concludes that there are issues (e.g. unanswered questions) in all three areas though with a clear ranking from least problematic (GOS) to most problematic (GDS). 10. At NRAC on 22/3/06 it was agreed that the FHS work would work to a different, presumably longer, timescale compared to the rest of the Review and not be locked into the timetable for general consultation. At the same meeting there were also concerns raised about possible „face validity‟ problems with the formulae, such as potential large changes in funding for some Health Boards. 2. Comparison with Arbuthnott – differences in approach and concepts 11. There are some differences in method and approach in how these formulae were developed compared to how formulae have been developed first under Arbuthnott Review and now under NRAC. Not all of these have significant implications but are useful pointers on what NRAC will need to address in taking this work forward. Deloitte based their final formulae on the most up to date year‟s data available at the time, mostly 2003-4, and so were unable to take account of: a) future contract/policy changes (though there was limited discussion of suitability under new contracts). b) year-on-year stability. No core criteria were used to judge the formulae recommendations. No analysis of the effect of supply on utilisation was included (as it is in Arbuthnott Formula). Deloitte used epidemiological data alongside utilisation data within some of the formulae, to produce a “hybrid” approach. The formulae were based on the existing Arbuthnott areas geography rather than data zones now being introduced. The formulae include references to elements of the current Arbuthnott Formula which are likely to change as a result of the NRAC review (e.g. MLC Index). Should try and refer to potentially new elements which NRAC may recommend (e.g. new excess cost adjustment). The authors themselves had concerns about the final formulae in each of the areas to varying degrees (GOS least, GDS most) and recommended further research work on certain aspects. There has been no external QA of the technical work, and, since the work was carried out by a single analyst within Deloitte, little scope for internal QA. 12. Some important conceptual issues regarding FHS in comparison with HCHS which would need to be taken account of in resource allocation. Again there are varying levels of implication for the formulae but worthy of documentation and consideration. Influence of contracts Control over supply Private sector provision NHS Board obligations Administrative data sources Cross-boundary flow These issues are expanded in the Annex. 13. Worth noting that some of these issues will also have been relevant to GMS so there is scope for learning any lessons and tapping into any previous thinking. 3. Recommendations 14. Given all the above it is difficult to see how the current Deloitte formulae can be “accepted” as-is. One difficulty is that there is nothing to compare them with (other than current method). 15. Worth re-stating that NRAC‟s remit with regard to FHS is to: “advise on possible formulaic approaches to the parts of health expenditure not currently covered by the Formula (e.g. primary care dental, pharmaceutical and ophthalmic services)” As such the Deloitte formula can be seen as one approach (another is the current method). 16. Recommendations: take a step back - change emphasis from accepting or rejecting the Deloitte formulae to advising on suitability of alternative approaches. This might include the pros and cons of Arbuthnott and non-Arbuthnott type (e.g. epidemiological) approaches. Most are agreed that the idea of a formulaic approach is a good one, but discussions have got confused with issues about implementation. Therefore concentrate on : o developing formulae that reflect needs (in remit), rather than o implementing such a formula, e.g. the when and how (not in remit) NRAC would report on progress with available options and recommendations for example based around the following: o a formulaic approach is recommended for FHS – why ? o formulaic approach should be based on Arbuthnott Formula – why ? o particular issues for FHS – discussion o what the formulaic options are – current, Deloitte formulae as-is, Deloitte formulae developed for new contracts, etc o any recommendations for further work Proposed next steps for expanded formulaic options are described below. 4. Next steps 17. The further work on developing the FHS formulae will be led by ASD, working with policy colleagues and those at ISD who were involved in the original Deloitte research work. If the work plan is agreed upon then the work on the FHS formulae would continue with reports on progress at NRAC meetings. 18. NRAC should consider if/how to reply to those organisations and individuals who responded during the FHS consultation period. This could possibly be done via e-mail, or by some segment in the online consultation document. 19. The following work on all three areas (GOS, PCS and GDS) will go ahead separately but simultaneously, although most progress will be on GOS, and the least on GDS. For GOS it should be possible to complete sections A, B and C. Section A will be done for PCS and GDS, but anything else will depend on resources available. This is a reasonably intensive program of work, and will need to be fitted around other NRAC work during and following the consultation period. WHAT A. Knowledge gathering (GOS, PCS, GDS) Find up to date data for all aspects of formulae, data that has become available since reports were written Examine when data accounting for contract changes will become available WHO ASD/ISD BY WHEN Autumn 06 ASD/ISD Autumn 06 Consider how any policy changes since reports were written could affect underlying bases of formulae B. Formula updates (GOS) (PCS?, GDS?) Improve spreadsheet models, remove mistakes found in internal QA Re-run formulae with updated data, check year-onyear stability Re-run formulae with data reflecting contract changes (if possible), check stability following the changes C. Formula development (GOS) (PCS?, GDS?) Consider specific problems with each formula (especially cross-boundary flow) raised in consultation and by Deloitte, consider ways of improving Sensitivity analysis considering effect of possible future changes in demand Consider how FHS formulae (based on „old‟ Arbuthnott) could be affected by the move to the „new‟ allocation formula recommended by NRAC (e.g. MFF adjustment, datazone geography) ASD/SEHD policy Winter 06 ASD ASD ASD Autumn 06 Winter 06 Winter 06 ASD/ISD/SEHD policy Winter 06 ASD/SEHD policy ASD/ISD Spring 07 Spring 07 20. The endpoint will be to ensure that NRAC‟s final report (Summer 2007) can „advise on possible formulaic approaches‟ with respect to all three areas of Family Health Services. The work on FHS is operating on a different timescale to the majority of NRAC‟s other work the final report of NRAC should contain progress to date, and recommendations for further work. If appropriate data can be found and allocations given by a formula remain stable following introduction of contract changes and updating of spreadsheets then could be recommended by NRAC that formula be adopted. If suitable data is not available and/or formula results are shown to be unstable then NRAC can advise on; a) whether a formula is appropriate for the area of FHS, and b) if it is, then what work is necessary in the future to develop it. Annex Contractor status – FHS contractors are not NHS employees and the services they provide are governed by contracts which state how the contractors will be reimbursed for services provided. This suggests that if a contract changes then the “need” for that service may change. E.g. new sight test includes a wider health check (e.g. for signs of diabetes) and the need for this may be wider than for a simple eye-sight test. Therefore, any formula has to take account of how services are defined contractually at any point. Change of contract may therefore require a change in formula (rather than simply an update). Supply issues – Boards have not had a great deal of control over location and number of contractors up until now, often this is more of a commercial decision and based on other non-NHS factors. This might result in some areas being over-supplied and others undersupplied and suggests that this should be taken into account in deriving a formula. Private sector – the private sector plays a greater role in these services than for HCHS and, in some areas, may be the only source of service. Even where NHS services are provided a much greater chunk of contractor time is taken up with non-NHS activities. Raises the question whether the formula should predict the needs of the whole population or just those likely to use NHS services. Board legal obligations – For HCHS and GMS Boards are legally responsible for providing services to all their resident population. From April 2006, for Pharmaceutical Services, Boards have also become responsible for resident populations. For dental and ophthalmic services Boards are only responsible for ensuring that services for treating patients are provided within their boundaries. This has implications for where funds are allocated and how cross-boundary flows are treated. Administrative data sources – much of the data around FHS comes from administrative data sources set up to reimburse contractors based on their contracted services and so often are centred on reimbursement costs. Often this means patient and clinical information is poor or missing compared to say hospital records which are required more for clinical reasons. Cross-boundary flow - Deloitte included an adjustment for cross-boundary flow (not currently part of Arbuthnott, since a separate mechanism exists outside the formula). With the exception of pharmacy, this was based on new, previously untried, record linkages of databases at ISD with mixed success (96% matching for dental, 42% for ophthalmics).