you are at the homepage - clicking the corresponding link on other pages will bring you back to this page HOME
visit the Scotland's Health on the Web website, with links to further health information OTHER SHOW SITES
general help in using Scotland's Health on the Web HELP
COMMENTS

NHSSCOTLAND RESOURCE ALLOCATION COMMITTEE

NHSScotland Logo - links to the SHOW home page
NRAC Home
Resource Allocation
Role & Work
NHS Boards
Research
Consultation
FAQ
Links
Members Area

Search the SHOW site

Frequently Asked Questions

1. Principles of Arbuthnott Formula
2. Coverage and Scope of Arbuthnott Formula
3. Operation of the Formula
4. Role of the Committee and its timescales
5. Other Allocation Formula
6. Unmet Need
7. Island Boards

1. Principles of Arbuthnott Formula

1.1 Q: Principles of the Arbuthnott Formula

1.1 Q: The Arbuthnott Formula does not provide us with sufficient resources to cover the healthcare needs of our population, yet it is supposed to be needs-based. Why is this?

A: The Arbuthnott Formula does not determine the total amount of resources required to meet all the needs of a NHS Board. The funds available to Scotland’s 14 NHS Boards are determined by Ministers during the Spending Review process. The Formula allocates this amount on a basis that is fair and equitable, and reflects the relative need of each NHS Board. It is then up to NHS Boards to decide how to spend their allocation in a way that best meets the needs of its resident population.

1.2 Q: How does the Arbuthnott Formula work?

A: The Formula assesses each NHS Board’s relative need for funding, using information about its population size and characteristics that influence the need for healthcare in terms of hospital services, community services and GP prescribing. The main drivers of the Formula are:

(i) share of the Scottish population living in the NHS Board area;
(ii) age structure of the population and relative number of males and females;
(iii) morbidity and life circumstances (e.g. deprivation); and
(iv) additional costs of delivering healthcare in remote and rural areas.

1.3 Q: What is the underlying principle of the allocation Formula?

A: The main objective of the Arbuthnott Formula is to provide for equity of access to healthcare. Resources are distributed among NHS Boards on the basis of relative need for health care services within that population group, where use of services has been used as a proxy for need. Scotland uses an indirect approach to measure healthcare needs. The indirect approach relies on health service utilisation data to measure those needs based on (i) the demographic profile of the populations, taking into account the national average costs of providing services based on age and sex, and (ii) relative levels of deprviation, and its estimated relationship on the greater use of services within each care programme. In addition to these two factors, the relative need for resources in each NHS Boards is also influenced by the additional costs of providing services in remote and rural areas.

1.4 Q: What are the care programmes and diagnostic groups used in the Formula ?

A: The table below sets out the care programmes and diagnostic groups that form the current Arbuthnott Formula.

 

Care Programme

Diagnostic Group(s)

Acute Services

Circulatory
Cancer
Respiratory
Digestive system
Injuries & poisoning
Other

Care of the Elderly

None

Mental Illness

Schizophrenia
Dementia
Non-psychotic disorders
Substance misuse
Other

Learning Disabilities

None

Maternity

None

Community 1

Health Visitors
District Nursing

GP Prescribing 2

Circulatory
Gastro-intestinal
Infections
Mental illness
Musculoskeletal
Other

1. Health Visiting and District Nursing are used as a proxy for all community services
2. Prescribing programme was disaggregated into the top five British National Formulary chapters


1.5 Q: My NHS Board provides healthcare services for 10% of the Scottish population, yet only receives 9% of the funding share – why is this?

A: A NHS Board’s share of population forms the basis of its allocation. However, this is then adjusted for factors that affect relative need for healthcare resources (age/sex, deprivation and remoteness).

For example, elderly people tend to make more use of health care services and are more costly to treat. Therefore, a Board with a greater elderly population will require more health care resources than one with a relatively younger population base. Deprived people are recognised to have a greater need for healthcare than relatively affluent people. Similarly, we recognise that there are additional costs in providing services in remote and rural areas. The impact of these factors are combined to create an overall index of need for each NHS Board, and this will determine the level of funding that a Board receives.

1.6 Q: What is the Arbuthnott Index ?

A: The level of healthcare need within Boards is measured, by proxy, using a composite index of morbidity and life circumstances known as the Arbuthnott Index. This is composed of four indicators that were found to be consistently associated with the use of healthcare resources across a range of diagnostic groups and care programmes. The four indicators are:

(i) the standardised mortality rate among people under 65;
(ii) the standardised rate of unemployment benefit claimants in the working-age population;
(iii) the percentage of elderly people claiming income support; and
(iv) households with 2 or more indicators of deprivation at the 1991 Census.

The 6 indicators of deprivation considered in (iv) were: unemployed or permanently sick head of household, low socio-economic group head of household, overcrowded households, large households, lone-parent families and all-elderly households.

1.7 Q: How do you estimate the population?

A: For hospital services, the Formula uses mid-year estimates from General Register Office for Scotland (GROS). For the 2006/07 allocations, the estimate will be based on 2004 mid-year estimates. Alternatives measures for population, (including population projections) have been considered by researchers who have reviewed the population basis of the Formula for NRAC.

For GP prescribing the population source is the Community Health Index (CHI) which contains every person registered with a GP in Scotland.

See Technical Report B prepared for consultation for more information on population sources.

1.8 Q: Why is population calculated differently for hospital services and GP prescribing?

A: For hospital services the population is based on the NHS Board of Residence, however, for GP prescribing the population base is NHS Board of Management.

For GP prescribing the activity measure is the volume of prescribing based on GP practices irrespective of the home address of the patient. Population is based on the number of patients on the lists of GPs managed by each NHS Board, rather than simply the NHS Board of residence.

1.9 Q: Why is it important to take into account the age/sex profile of the population?

A: The Arbuthnott Formula uses this information to take account of the use of different specialities by each age/sex group (e.g. for maternity services), and also in calculating the costs of treating patients of different ages. It makes the Formula more ‘sensitive’ to the healthcare requirements of the different population groups.

The age-sex cost weights have been reviewed for NRAC by researchers. See Technical Report C prepared for consultation for more information on age-sex cost weights.

1.10 Q: How do you measure levels of remoteness?

A: There was only a very limited adjustment for remoteness in SHARE, the formula prior to Arbuthnott. The Arbuthnott Committee thought it was very important to improve this as this is a particularly important Scottish issue.

The remoteness adjustment is intended to consider the extent to which on average, NHS Boards in more remote and rural areas, face relatively higher costs for hospital services and community services. It is an indicator and so measures outcomes that influence costs, and not a direct measure of those costs.

For hospital services - Estimates of the effects of remoteness on hospital services are based on the relationship between road kilometres per 1,000 population and the relative costs of hospital services for the residents in different NHS Boards. During the Arbuthnott review, this was found to be the best indicator of the influence of remoteness on costs. As the Island Boards are small, they were combined and one adjustment was created for all of them.

For community services separate remoteness adjustments are applied to

a) all travel intensive community services (e.g. district, nursing, health visiting, community midwifery, etc) and
b) services from fixed locations (e.g. clinics, health centres).

For the former, amounting to around two-thirds of community services costs, an adjustment is based on a cost index developed by NERA Consulting for delivering district nurse and health visiting services in remote and rural areas. This index is based on a model for predicting the number of average minutes per nursing contact that includes factors for contact rates, contact times, time traveling, hours available for patient care and grade mix within population clusters. The travel-intensive component of the model concentrates on travel within and between clusters unable to justify a local nursing team and clusters of less than 500 persons.

For services from fixed locations the remoteness adjustment derived for GMS services was applied. This is based on a three factors which at Board level were significant predictors of the proportion of GMS costs influenced by rurality and remoteness: population density, proportion living in communities of less than 500 people and proportion of practice lists qualifying as road mileage patients.

The remoteness adjustment has been reviewed for NRAC by a team of researchers. See Technical Report E prepared for consultation for more information on the remoteness adjustment.

1.11 Q: Why doesn’t GP prescribing have an adjustment for remoteness?

A: The GP prescribing element of the Formula covers the cost of prescribed drugs which are reimbursed at nationally fixed prices. Therefore, there is no need to build in a remoteness adjustment.

1.12 Q: How do you take account of economies of scale?

A: The issue of economies of scale was considered in the work of the Arbuthnott committee with respect to how it would affect adjustments being made for remoteness within hospital services. The committee considered 4 hospital types (acute, mental illness, elderly care and maternity) and considered possible economies of scale at 4 average cost levels for each (total, nursing, medical and allocated costs). The findings were that there are:

  • Acute hospitals - significant economies of scale in all cost areas. Higher levels of inpatient and day case activity was associated with lower costs.
  • Mental illness hospitals - economies of scale only present in total costs. Higher numbers of inpatient weeks (squared) associated with lower total costs.
  • Elderly care hospitals – economies of scale exist in all cost areas. Higher levels of inpatient weeks reduce costs, but only over a certain range.
  • Maternity hospitals – economies of scale in all cost areas. Higher numbers of births associated with lower costs, over a certain range.

    This evidence that smaller hospitals tend to have higher average costs than larger hospitals was one of the major arguments for including the remoteness adjustment in the hospital services section of the Arbuthnott Formula, since it is generally true that smaller hospitals are found in the more remote areas.

1.13 Q: How often is the Arbuthnott Formula updated?

A: The Arbuthnott Formula is updated each year for changes in population size, age/sex profile and the level of deprivation in each NHS Board. However, the index for remoteness stays constant. Within the deprivation index, all components are updated annually with the exception of the Census-based indicator, which is still based on 1991 census information.

See Technical Report A, prepared for consultation, for more on how the Arbuthnott Formula is updated and implemented.

1.14 Q: Does the Formula give enough emphasis to deprivation or remoteness?

A: The weights attached to different elements in the Formula are based on the best available evidence at the time of the Arbuthnott Formula, depending on how each factor influences the need for healthcare. The weights were not chosen, but based on empirical analysis. The adjustment for morbidity and life circumstances therefore takes account of the need for services within diagnostic groups over and above the affect of the age and sex profile of the population. The adjustment for the excess costs of supply then takes account of the additional costs of delivering services to meet the needs that are predicted by the age & sex and morbidity and life circumstances adjustments.

It should be remembered that the target shares for each Board are influenced not only by the different adjustments within the Formula but also by the profile of Boards. Most Boards are very variable, containing a mix of remote/urban areas and affluent/deprived areas, and this is taken account of when the results are presented at Board level.

See Technical Report A, prepared for consultation, for more on how the different components of the Arbuthnott Formula have performed since implementation.

1.15 Q. Isn’t the allocation skewed because people access more services where they are available ?

A. Controlling for the impact of supply is a key aspect of the Arbuthnott Formula. Estimates of the relationship between the use of services and indicators of morbidity and life circumstances do take into account of the influence of supply (or accessibility). Accessibility is essentially calculated as a distance weighted beds per head of population. Supply scores are used in the regression to control for supply contamination.
In Fair Shares for All, the access measures used covered a number of types of hospital bed provision, local authority residential homes and private nursing homes — together with some measures relating only to the prescribing work (such as whether the practice was a dispensing practice). In the post-consultation work a range of other social work indicators relating to access to day care places and home help provision were also obtained and tested out in the modelling. This was done to address any concerns about the model not picking up sufficient information on alternative provision.


1.16 Q: How do you weight the different components of the Formula ?

A: The Arbuthnott Formula has the following basic structure:

Population * age/sex * MLC * remoteness

The aim of the modelling is to explain the current overall need for resources of each NHS Board in terms of a percentage share.

An index is calculated for each element of the Formula for each care programme in such a way that it compares each Board’s position with the national average. For example, if the levels of deprivation in a Board is higher than the national average its index will be more than 1 to reflect that its population will need more healthcare resources. By calculating each index in this way, the values can then be multiplied by the population share to determine how much more (or less) resource each Board requires compared with its basic population share due to age/sex, deprivation and remoteness.

The way each of the indices is calculated means that there is no need to set weights for the impact of age/sex, deprivation or remoteness – the strength of the impact comes through from the analysis.

In order to determine the overall adjustment for each Board, each of the care programme formulae are weighted together by the national average expenditure on those care programmes.

1.17 Q: Why do you use a multiplicative model ?

A: The original work for the Arbuthnott Formula investigated additive (linear) and multiplicative (log transformation) regression models for the deprivation adjustment to test which model resulted in the best fit of the data. The additive model consistently performed better in terms of the goodness of fit tests, therefore, this was used in the final work on the Arbuthnott Formula. Therefore, the deprivation adjustment is not a multiplicative model.

However, the overall Formula is multiplicative in the sense that the raw population share for each NHS Board is multiplied by the relative needs indices for each NHS Board. We start with the raw population count for a NHS Board and then try to take account of the difference in healthcare needs/costs of the population. The Formula attempts to predict the expected workload for NHS Boards based on their population. Each index is based on evidence and turned into a relative measure so that the index can be multiplied by the population. The span (i.e. the range between the highest and lowest Board) of the relative measure determines the strength of the adjustment. The extent to which each impacts on the raw population share is therefore determined by the evidence from within each adjustment e.g. for a given care programme the age/sex adjustment might be the least strong of the three.

In this way, we transform a raw population share to reflect the anticipated extra/less need and extra/less cost facing each of the NHS Boards – and express this as a weighted population share. The adjustments are evidence based and the evidence gives the weight of the index for each NHS Board. The index will be higher/lower for each adjustment depending on the relative strength of each indicator based on the evidence. Each adjustment does not explicitly have a weight attached to it because there is no need to do so given the strength of the impact comes through in the calculations.

2. Coverage and Scope of the Arbuthnott Formula

2.1 Q: How much funding is distributed through the Arbuthnott Formula?

A: The Arbuthnott Formula covers funding for Hospital and Community Health services, and GP Prescribing. This amounts to around 70% of the total healthcare budget in 2005/06. Other formulae are used to distribute other areas of funding such as General Medical Services and capital allocations.

See Improving the Arbuthnott Formula, prepared for consultation, for more on how NHS budgets are distributed.

2.2 Q: How does the Arbuthnott Formula take account of cross-boundary flows?

A: The Arbuthnott Formula allocates resources on the basis of NHS Board of Residence and not by NHS Board of Treatment. It is up to individual Boards to recover costs for patients treated from other NHS Boards, and this has traditionally been done through Service Level Agreements (SLAs) and now tariffs.

2.3 Q: How will the National Tariff impact on the Arbuthnott Formula and/or the work of NRAC?

A: The national tariff applies only to cross-boundary activity, and the treatment of flows is not covered by the Arbuthnott Formula. These tariffs will be phased in over a number of years to avoid any financial turbulence for Boards.

Consideration has been given to using tariff costs in the age-sex cost weights component of the Formula, but it has been decided not to use them at this stage.

2.4 Q: Are community hospitals covered in the hospitals section or the community services section?

The costs of community hospitals are included under the appropriate care programme of the Formula e.g. acute, care of elderly, maternity etc. depending on the activities that are carried out, rather than the location. They will not be included in the community section of the Formula as this only covers activity outside of hospital e.g. in the patients home.

2.5 Q: How are travelling people, homeless, immigrant workers, prisoners, students and temporary residents taken account of in the Formula?

A: These specific population groups are taken into account in the following way:

  • Travelling People & Homeless – if these people were not picked up in the census then they will not be included in the MYE, however, if they are registered with a GP they would then be included.
  • Immigrant workers are picked up through the international passenger survey
  • Prisoners are included as residents in the NHS Board where they are imprisoned.
  • Students are included in the GRO population figures at their term-time address.
  • Refugees and asylum seekers are included in the GROS population estimates.

2.6 Q: How are temporary residents dealt with in the Formula?

A: There are two aspects to healthcare provision for temporary residents – hospital admissions, and prescribing.

  • Hospital Admissions - NHS Boards are able to claim back the costs of treating non-resident populations through the finance mechanisms that are in place. This applies to either residents in other Scottish NHS Boards, or visitors from other countries – the latter is achieved through UNPAC (unplanned activity) provisions.
  • Prescribing - there is no capacity in the financial system to claim back the time spent with, or prescription costs of, visitors. Inter-board costs (or “cross-border flows” as they are known in Prescribing) are dealt with as part of the conversion of a Gross Ingredient Cost based formula modeled on NHS Board of Management to a Net Ingredient Cost based allocation on NHS Board of Residence in the finance system. For visitors, we therefore need to make an adjustment to the Formula – starting with the population base.

2.7 Q: Are differences in emergency admission rates and delayed discharge rates evaluated anywhere in the Formula (concern that may be more in very remote areas)?

A: No specific analysis has been carried out to look at the differences in emergency admission rates and delayed discharges between remote and urban areas. However see Chapter 9 of Technical Report E, prepared for consultation, for more on rural and urban differences in patterns of care.

2.8 Q: How much private provision of healthcare is there in Scotland and does that affect the NHS provision?

A: No information is routinely collected by the Scottish Executive Health Department on private health sector provision or usage, and sources of data that identify Scotland separately from the rest of the UK are difficult to obtain.

In 2002-3 there were 10 independent acute hospitals in Scotland with a capacity of 442 beds. Relative to the NHS this is very small representing only about 2.4% of the total acute beds across public and private sectors in Scotland. Also, Scotland has only 4.7% of the 9463 UK independent sector acute beds so independent sector capacity is greater per head of population in the rest of the UK. In addition, in Scotland, there are a small number of facilities offering mental health services (e.g. drug rehabilitation).

The proportion of adults covered by Private Medical Insurance (PMI) is generally lower in Scotland than the rest of the UK. Estimates of prevalence of PMI at 8% for Scotland compared to 12.3% in Great Britain as a whole in 2001 and 19% for the South East of England (Laing&Buisson 2002-3).
.
Uptake of PMI has been found to associated with a number of individual characteristics including age (greater among the middle-aged) and socio-economic status (greater among the less deprived). Uptake also tends be higher among those who rate themselves in good health and so prevalence of PMI may not be a good indicator of actual use of services.

By ignoring any non-NHS health services (e.g. private, self-care) the Formula implicitly assumes that the use of NHS and non-NHS services is through patient choice and not design. In other words, the current configuration and provision of NHS services means that those who need to access an NHS service can do so and those who use a private service do so by choice.

3. Operation of the Formula

3.1 Q: Will NRAC be able to influence and/or make recommendations on how the Formula is currently being implemented?

A: No, responsibility for implementation and issues on the operation of the Formula lies with the Scottish Executive Health Department (SEHD), and is outwith the scope of the Committee. However, NRAC is keen to offer ideas about how the success of the Formula can be demonstrated and we are keen to hear your thoughts on this. Also to give clear advice on how the Formula could be better used for planning by Boards.

3.2 Q: Why do the relative shares as calculated by the Arbuthnott Formula differ from the actual shares that NHS Boards receive of the final allocations?

A: This issue relates to the movement towards parity. SEHD decided to phase in the Arbuthnott Formula by way of ‘differential growth’ whereby all Boards would continue to enjoy real-terms growth in their allocations year-on-year, with those above parity (i.e. above their Arbuthnott target share) receiving less growth than those below parity until the new distribution was achieved. This is still ongoing.

3.3 Q: The Formula is a waste of time as the movement towards parity is too slow?

A: It was expected that the Arbuthnott Formula would be phased in over 5-6 years. This has not proved possible, given the commitment of the Health Department to ensure that no Board would lose funding as a consequence to moving to the new formula.

All Boards are given a standard increase in funds each year. The size of the standard increase is sufficient to meet the main pressures including pay and price inflation and the cost of prescribed drugs, together with funds for developments. Those Boards below their Arbuthnott target allocation are given additional funds (around 0.5% of the unified budget) to help move them towards their parity target at a faster rate.

3.4 Q: Which Boards are above/below target and why?

A: There are a number of Boards that are currently receiving budget shares above their Arbuthnott target shares. These Boards are Greater Glasgow & Clyde, Shetland, and Western Isles, with Tayside being only marginally above target. All other Boards are below their target shares at present. Details of each Board’s distance from parity following the 2006-07 allocation are shown in parity section of the website. Boards are above/below target due to the phasing in of the Arbuthnott Formula and the principle of differential growth.

4. Role of the Committee and its timescales

4.1 Q: Why do we need a review of Arbuthnott when the Formula has not been fully implemented (that is, not all Boards have reached parity)?

A: The original Committee made a commitment to review the Formula after it had been in operation for 5-6 years. The purpose of this review is to find out if the Formula can be improved or refined on the basis of better evidence, and/or new data which may impact on the Formula.

4.2 Q: Are you planning a radical overhaul of the Arbuthnott Formula?

A: No, the purpose of NRAC is not to change the basic principles of the current Formula. The remit states that the new Committee will look into ways of ‘improving and refining’ the Formula to take into account any changes in available information and data since the last Review.

4.3 Q: What are the timescales of the Committee?

A: The Committee was established in February 2005. A consultation on research commissioned by the Committee is to run from July – September 2006. The Committee is due to report on its final recommendation to the Minister in summer 2007.

4.4 Q: How will any changes and/or recommendations made by NRAC be implemented?

A: NRAC will publish a report outlining its recommendations on the Arbuthnott Formula in view of the evidence it has considered during the life of the Committee. These recommendations will be considered by the Minister for Health and Community Care, and it is the role of the Scottish Executive Health Department to be responsible for if and how the recommendations are implemented.

4.5 Q: My Board is not represented on the Committee – how can I be sure that the Committee is impartial, and will take into account any concerns that my Board has?

A: All committee members have signed up to undertake their role with honesty, integrity and political impartiality. Members will contribute to the Committee as individuals using their skills and experience, and their role is not to represent the views of their NHS Board. The Committee will seek to ensure openness and transparency in all its work: by presenting the work we plan to do and consulting on elements along the way, explaining the rationale for our recommendations, and communicating in language and formats that are easily accessible.

In 2005 NRAC conducted a tour of NHS Boards to meet with representatives to help inform them of the work being conducted, and to discuss any issues/concerns they may have on the review of the Formula. A summary of the issues raised can be found in the NHS Boards area of the website.

4.6 Q: How will you know when you have made an improvement to the Formula? How will you judge success?

A: The Arbuthnott Committee used a number of ‘core criteria’ for judging between the formulaic options in the last Review. NRAC has agreed to adopt a slightly modified version of these to keep judgements about different options that emerge, within a consistent framework.

4.7 Q: What issues is the Committee intending to consider during its work?

A: At an early stage, the Committee identified the following issues to be taken forward within this remit:

  • Improving and refining the allocation Formula: by considering new data developments, defining population at a smaller area level than currently used, and reviewing the deprivation and remoteness indices;
  • Advise on possible health expenditure not currently covered by the Formula, such as Family Health Services;
  • Reviewing information to support existing elements of the Formula e.g. Health service reform plans (including community provision, health improvement, joint futures), resource allocation work from other countries etc, and;
  • The issue of unmet need and related pilot studies, and considering formula adjustments on this basis.

Progress on these issues can be found in Improving the Arbuthnott Formula, a report prepared for consultation.

4.8 Q: What are the important data developments that the Committee will consider?

A: The main data developments that the Committee will consider are:

  • Availability of prevalence data, e.g. from the Quality Outcomes Framework;
  • the availability of data at data zone level from Scottish Neighbourhood Statistics,
  • the 2001 Census

4.9 Q: How does the Scottish Index of Multiple Deprivation (SIMD) work on deprivation affect the Arbuthnott Index?

A: SIMD has been adopted as the standard measure of multiple deprivation by the Scottish Executive. The SIMD pulls together a large number of indicators to produce indices for a number of domains, e.g. income. The Arbuthnott Index is not an indicator of deprivation but an indicator of healthcare needs. The research to review the index and the adjustment for morbidity and life circumstances has considered a variety of data sources that could reflect health care needs, including many of the data sources used in SIMD.

See Technical Report D, prepared for consultation, for more information on the review of morbidity and life circumstances in the Formula.

4.10 Q: How does the Welsh approach differ from the Scottish approach?

A: Wales use a direct approach to measure healthcare needs, they make direct use of survey data to make adjustments for demography and additional need. Wales:

  • Obtain region-specific prevalence rate estimates for each age-group using a health survey.
  • Apply region-specific prevalence rate estimates to age profile of resident population to obtain estimated numbers of cases.
  • Calculate each region’s share of national cases.
  • Obtain national budget for the condition.
  • Derive regional budgets by multiplying national budget by the regional shares of cases.

In comparison Scotland makes indirect use of surveys to improve adjustments for additional need

4.11 Q: Can we follow a Welsh epidemiological approach?

A: The Arbuthnott Committee considered the feasibility of adopting an epidemiological approach in its original discussions and in the work it commissioned on unmet need. However t did not recommend adopting this approach, in the main due to the lack of robust, national prevalence data that was required.

In the current review, NRAC looked again at the epidemiological approach while recognising that any such move in that direction would involve a substantial change from the current resource allocation Formula. This method requires detailed data on the morbidity of the population with a wide range of coverage across all the NHS Boards in the country. It also requires the ability to link the morbidity data to data on subsequent use of health services, at a patient level, to determine a method of allocating costs.

After reviewing the available data sources for measuring the morbidity of the population with suitable linkages to health service activity, it was concluded that testing a formula based on a full epidemiological approach was not feasible within the timescale of the Committee’s work. However NRAC believes there may be scope in future for using morbidity data and direct measures of healthcare need where possible.

4.12 Q: What impact will ‘Delivering for Health’ have on the review?

A: The Committee will keep a watching brief on the developments occurring as a result of the Kerr Review and subsequent ‘Delivering for Health’ report and any effect these may have on resource allocation issues. This may include for example, the recommendations on:

  • Care of Older People and long-term conditions - for the age/sex adjustment
  • New investment to allow people from deprived areas to have enhanced access to a range of health interventions - for deprivation issues – links to unmet needs pilots
  • New ways to deliver rural health care – on the costs of providing services to rural areas.

Overall it will be the aim of our Committee to ensure that the resource allocation work is linked as closely as possible with policy developments and to future proof the formula wherever possible.

4.13 Q: If research is being commissioned on supply and need, what about demand?

A: The research that has been commissioned to look at “supply” is looking at the excess costs of supplying healthcare and not the absolute quantity of healthcare that is/should be supplied. The “needs” research is reviewing how healthcare needs due to morbidity and life circumstances (MLC) and other factors are currently accounted for in the Formula, with a view to refinement and improvement.

The Arbuthnott Formula is a Scotland-wide allocation formula that allocates resources to NHS Boards on the basis of relative patient need and not patient demand for healthcare. The Formula assesses the relative need for resources in each NHS Board area by taking into account its population share weighted by age/sex, the level of deprivation and the effect of remoteness on the costs of delivering health care services. It is based on the assumption that relative use made of services is a proxy for relative need. The review is to “improve and refine” the Formula not to question its basic principles.

Modern healthcare theory states that the demand for healthcare seems to be infinite as everyone wants improvements to their quality of life. Healthcare is not just about saving lives but also about reducing pain and discomfort. As with all other public services, there is a limit to the funds available to the NHS which can make it difficult to manage patient expectations. However the role of a national funding formula is to distribute funds on the basis of need and the role of service providers, e.g. Boards, is to manage demand locally.

5. Other Allocation Formulae

5.1 Q: Will the Committee re-visit the Additional Cost of Teaching (ACT) Review, or review the implementation ?

A: No. ACT has just been reviewed under a sub-group for the last Committee chaired by Sir John Arbuthnott. The Committee made its recommendations in January 2004, and these were approved by the Minister. The proposals for Direct Costs were subject to consultation, and will now be implemented in 2005/06 on a phased basis. NHS Education Scotland (NES) was given the responsibility for taking forward this element of the Formula and its implementation, which was overseen by a Working Group comprising representatives from both Teaching and non-teaching Boards.

5.2 Q: Why was a distinction made between Direct and Indirect Costs?

A: The research into the review of ACT made a distinction between Direct and Indirect Costs on the basis that direct costs were those that could be directly attributed to the teaching of medical students (and medical teaching of dental students). A model was developed to estimate the actual costs to the NHS of supporting these students and this was used to inform the Direct Costs formula.

On Indirect Costs, research was commissioned to find evidence to support the view that major teaching hospitals incurred inevitably higher running costs due to teaching activities (in addition to the direct costs). The research could find no evidence to explain why teaching hospitals should have higher costs, which were estimated to be around £25m for the 8 teaching hospitals in Scotland. To avoid financial turbulence for Boards it was decided that rather than removing these funds, the balance of the ACT funding and Direct Costs (around £28m) would be added to their Unified Budgets, and distributed on the basis of the Arbuthnott shares for the 4 Teaching Boards.

5.3 Q: Will you be extending the scope of the Formula to include capital allocation?

A: A review of capital allocation is not covered by the remit of NRAC. However, Capital is allocated using a variation of the Arbuthnott Formula with 90% of the capital budget being distributed based on Arbuthnott shares adjusted for cross-border flows. The remaining 10% of the capital budget is distributed among the four main Tertiary centres based on their share of regional specialty flow. There improvements to the Formula will result in changes to the capital allocation.

5.4 Q: Are the formulae you issued for Family Health Services (FHS) going to be implemented?

A: NRAC commissioned research from Deloitte MCS Ltd on the three areas of FHS spending – dentistry, ophthalmics and pharmacy – to determine if it was possible to develop needs based formulae similar to the Arbuthnott Formula. The reports proposed weighted capitation formulae for each of the three services based on the best available evidence and data.

NRAC put the three reports out to consultation in December 2005. The consultees included the services, NHS Boards, academics and other stakeholders. The consensus from respondents was that they welcomed the principle of allocating funding by means of a formula but that there were concerns with (a) the specification of the proposed formulae and (b) the timing for changing the current allocation arrangements.

NRAC recognises the concerns and agrees that more work is required to address some of these before recommendations are possible. The Committee will take forward development of a formula for ophthalmics and make recommendations about the further work needed for the dental and pharmaceutical services formulae.

5.5 Q. Why can’t you just create one allocation formula covering health and social services?

A. Social services are provided by Local Authorities and funded through Resource Transfers from the NHS Board to the Local Authority.

5.6 Q: Will GMS be included in the work of the Committee?

A: No, there is a separate process to maintain and review a formula for GMS. The formula for GMS is subject to UK-wide negotiation. However NRAC will keep abreast of developments on the Scottish element of the review and any implications for their own work.

5.7 Q: What work is being done on the Scottish Allocation Formula (SAF) within GMS?

A: The Scottish GMS allocation formula, SAF, is currently under review. All the components of the present formula are being looked at:

  • the age-sex weights and the deprivation measures of the workload component, which is based on the number of practice contacts;
  • the rurality component, which models the impact of density and sparsity on practice expenses; and
  • the staff Market Forces Factor, which models differences in staffing costs between NHS Boards.

The Review Group expects to report on its findings in the autumn of 2006, at the same time as the UK Formula Review Group.

6. Unmet Need

6.1 Q: What is meant by unmet need? How does this differ from a shortage of supply in services that results in some healthcare needs not being met?

A: The Arbuthnott Formula distributes resources on the basis of need, and assumes that the ‘use of services’ is a good indicator for need. However, research on this topic found that while people in severely deprived areas tend to use services more than those in non-deprived areas, this was less than would be expected given the higher levels of their healthcare needs. The Arbuthnott Formula could therefore, be underestimating the need for resources among those living in severely deprived areas. It is this issue of differential use of services for a given level of need that is defined by the concept of unmet need.

6.2 Q: Why were only 3 boards selected for Unmet Need pilots funding?

A: Greater Glasgow, Argyll & Clyde and Tayside were selected for these pilot projects because they have a high proportion of their population living in areas of severe deprivation (lowest decile of Arbuthnott Index). In Greater Glasgow, almost 40% of the population live in areas of severe deprivation. In Argyll & Clyde and Tayside this figure is just over 10%. All other Boards have less than 10% of their populations living in severely deprived areas.

6.3 Q: How much funding has been allocated to these pilots?

A: £15 million was made available to the selected Boards over a period of 2 years, from 2004/05 to 2005/06. Of this, Glasgow received £12 million and both Argyll & Clyde and Tayside received £1.5 million each.

It was initially intended that the outline plans would be approved by July 2004, to enable the pilots to be implemented during 2004-05. The pilot projects were due to be completed by the end of 2005-06 and overall conclusions reached about their effectiveness (and cost-effectiveness) by mid 2006-07. However, it took until November 2005 for all three bids to be approved. The timescales slipped partly due to unexpected difficulties in finding suitable projects for the pilots; difficulties with the interpretation of the pilots by Greater Glasgow ; and (latterly) the dissolution of Argyll & Clyde.

6.4 Q: How will the Formula be adjusted for unmet needs?

A: A workshop was held in January 2006 to provide an opportunity for the three Boards to share knowledge and experience from the pilots with each other. The Boards have been asked to provide an interim report on progress in September 2006 to feed into NRAC’s consideration of the issue of unmet need for the final report. However, the final evaluation stage of the pilots will not be completed until late 2007. That will be too late to feed into the work of NRAC. However, any relevant results from the pilots will be reported to NRAC during 2006-07 and will be incorporated in the committee’s final report, together with any recommendations for future work. Alongside this, further work on formulaic methods for adjusting for unmet need with the improved allocation formula will be considered.

7. Island Boards

7.1 Q: The island adjustment is unfair because it doesn’t apply to boards with island such as the former Argyll & Clyde or Ayrshire & Arran?

A: The greater cost of services associated with remoteness within boards such as the former Argyll and Clyde and Ayrshire and Arran is reflected in the remoteness indicator.

The remoteness indicator takes the average road kilometres per 1,000 population for the board as a whole, and therefore takes account of dispersed island population, though does not take account of sea miles between islands and mainland. The remoteness adjustment has been reviewed by researchers. See Technical Report E, prepared for consultation for more details.

7.2 Q: How is the remoteness adjustment for the 3 Island boards calculated?

A: All Boards receive a remoteness adjustment which is a combination of an adjustment for hospital services and community services.

For hospital - Orkney, Shetland and Western Isles all receive a uniform adjustment. This was based on the decision by the Steering group that all 3 Island Boards face the same scale of additional costs in providing hospital care. The rationale behind using a uniform adjustment was partly driven by the fact that there was no evidence to suggest that the differences in the extent to which remoteness influences their hospital costs was material enough to justify separate adjustments. In addition, the Island Boards were grouped together as a single observation in the analysis. This is simply because numbers for each island Board are quite small, and there were concerns about data quality.

For Community - Separate calculated adjustments were applied. The relevant community adjustments were calculated to be 12.5% for Orkney, 27% for Shetland and 27.1% for Western Isles. Separate adjustments were used for Community on the basis that these costs are influenced by the actual pattern of travel for community nurses which will differ between the island Boards.

When combining the hospital and community remoteness indicator, this produces slightly different values for the 3 Island Boards in the overall remoteness indicator.