Health Economics: 3 - Financial Resource Allocation

This section discusses the way in which National Health Service (NHS) funding is allocated from the Department of Health (DH) to Primary Care Trusts (PCTs) in England. It does not cover any other means by which financial resources are allocated,
for example from parliament to the NHS or from PCTs to providers. It also does not cover the NHS financial resource allocation mechanisms in Scotland, Wales or Northern Ireland, which are different to that of England, although some principles remain
the same. Full details of NHS allocation mechanisms are available for all four countries online:

England:

www.dh.gov.uk/en/Managingyourorganisation/Financeandplanning/Allocations/index.htm

Scotland:

www.nrac.scot.nhs.uk/allocation.htm

Wales:

www.wales.nhs.uk/sites3/page.cfm?pid=11612&orgid=452

Northern Ireland:

http://www.dhsspsni.gov.uk/index/stats_research/stats-resource.htm

The DH allocates financial resources directly to PCTs for them to spend on health care. The allocation for each PCT is a share of the resources available nationally, which is decided during the Treasury’s annual Spending Review. Each PCT’s
share is determined by a process that includes a ‘needs-based’ formula. This formula calculates what funding PCTs should receive so that they are all able to commission the same level of services for a given level of need.

The government has allocated resources to different areas of England using a formula since 1971, and has used a needs-based formula since 1976. It is worth noting that as well as being a means of distributing resources, it is also a means of
redistributing resources, since allocations to different areas are not equal with respect to need. The resource allocation process does not aim to achieve such equality every year; rather it seeks to get closer to equality. Historic inequalities
between areas that the NHS inherited when it was founded in 1948 remain, though of course they are not as large as they were and have continued to diminish over time.

The resource allocation process starts with a calculation, using the national resource allocation formula described below, of what the PCT should receive given the size of its population, the age distribution of its population, any additional need
factors and unavoidable variations in the cost of providing services. This is referred to as the target allocation. This is then compared with the PCT’s current funding, called the recurrent baseline, to see if the PCT is above target
or below target. Depending on the outcome of the spending review and therefore how much is available for the NHS in total, the next year’s allocation is calculated from the recurrent baseline. All PCTs will get a percentage increase, though PCTs
that are below target may get an additional percentage amount. The extent to which under-target PCTs get extra funding, and therefore inequalities are reduced, will depend on a number of factors, including how much of the additional funding available
has to be given to all PCTs because of national priority spending commitments. The precise details of these changes are called the pace of change policy. They may include aims such as ensuring that PCTs do not go further above or below their
targets, or that there will be a minimum increase for all.

The targets are derived from the capitation formula, in a way that will be explained below. The formula itself is essentially a weighted capitation formula. PCTs get a certain amount for each member of their population. However that amount may
vary depending on the characteristics of that population and of the area in which the PCT is based.

The formula is an aggregate of four different components, each of which has its own formula and weights. The contributions of these four components are themselves weighted within the overall formula to reflect their importance as a component of
overall spending. They are, with their weights, as follows:

Each formula has the same structure, but has different weights. They start with the resident population in the PCT area and adjust this for three factors, which are the age distribution, other need factors and unavoidable costs:

The age distribution

The rationale for this is that different ages have different needs and therefore demands for health care. In particular, if there are many older people then needs and demands will be high.

For the HCHS formula, the age adjustment is based on the national levels of service use and of cost by people divided into seven age groups. For the prescribing formula, it is based on prescription costs for different age and sex groups, plus an
adjustment for the number of temporary residents in a PCT. For the PMS formula, it is based on the GP workload for different age and sex groups, taking into account the different locations in which GP consultations take place.

The HIV/AIDS component is rather different; it actually has two formulas. One is intended to reflect treatment costs, and is based on the estimated prevalence of HIV infected persons in a PCT. The other is intended to reflect prevention costs, and
is based on the size of the at-risk population, taken to be the number of 15 to 44 year olds in each PCT.

Other need factors

The rationale for this is that in addition to age, socioeconomic factors may affect the need and demand for health care.

For the HCHS component, services are divided into Acute & Maternity and Mental Health, and there is a separate formula for each. These are then combined according to weights reflecting shares in national expenditure – 85% for Acute &
Maternity and 15% for Mental Health. In the Acute and Maternity model, the following need indicators are included:

• Standardised mortality ratio (SMR) less than 75 years old.
• Proportion of low birth weight babies born.
• Standardised birth ratio.
• ID 2000 education domain scores. This refers to the Government’s Index of Deprivation (ID), using the version of this produced in 2000. The domain scores are indicators derived from a number of variables taken from different sources including
  the national census. The education domain includes variables such as the number of adults with no qualifications and performance by primary school children.
• Proportion of aged 75+ years living alone.
• ID 2000 income domain scores. The variables in this domain are essentially the numbers of people who live in households that claim state benefits.
• Nervous system morbidity index.
• Circulatory morbidity index.
• Musculoskeletal morbidity index.

Other components include different sets of need variables and weights.

Unavoidable costs

The rationale for this is that even if the same amount of funding is given for equal need, this may not enable PCTs to purchase as much care for their population if the costs of providing care are higher in their geographical area. This therefore
attempts to even out the different purchasing power in different PCTs. It is referred to as ‘unavoidable’ costs because not all differences in costs are out of the control of PCTs. If costs are higher because of inefficiency, then there is no
justification for compensating for them; in fact this gives an incentive to be inefficient.

In the HCHS, unavoidable costs are mainly dealt with by the Market Forces Factor (MFF), which combines three separate indices: a staff index based on geographical differences in wages in the private sector; a land index based on valuations of the
NHS estate; and a buildings index based on average costs of new buildings. These indices are calculated for PCTs based on the MFFs for the Trusts from whom they commission, weighted by the amount of activity that takes place in them. The three
indices are combined into using national average expenditure shares of staff (67%), buildings (5%) and land (1%), with the remaining 27% of running costs being assumed not to vary geographically. The HCHS formula also has an emergency ambulance cost
adjustment (EACA) to take account of geographical variations in the cost of delivering emergency ambulance services.

The PMS component has a slightly different MFF, but uses the same principles, based on GP and practice staff pay variations and land and building costs in primary care. The prescribing and HIV/AIDS components do not have an MFF.

The index numbers that are derived from this weighted capitation formula are, of course, expressed in terms of populations, not funding. However, the index numbers are then used to calculate target allocations using information on the total
resources that are available for spending on the NHS nationally. There are also two special supplements to the formula, an English Language Difficulties Adjustment and a Growth Area Adjustment, which are then added as monetary adjustments to the
targets.

This resource allocation process has been subject to much debate. One view is that it has been one of the most long-lasting and consistently successful policy initiatives that UK governments have ever produced in terms of distributing and
redistributing public funds. It has certainly reduced geographical inequalities and, just as importantly, may have prevented greater inequalities from developing. However, the formula in particular has been heavily criticised, not always simply by
those who believe that they lose out because of it, on the technical grounds that the data within the formula are not always appropriate for the purpose for which they are used and that the weights are not based on sound statistical principles.
Unfortunately, because data are never perfect, it is probably the case that it is not possible to create a perfect formula, so that there will always be grounds for criticism. Unless a viable alternative to the formula approach is developed, it is
likely that it will continue, and continue to attract criticism.

© David Parkin 2009