Beggars belief

Published: 10/04/2003, Volume II3, No. 5850 Page 28 29 30 31

The new funding formulae for primary care trusts are intended to close the northsouth divide.

But, says John Hacking, progress will be so slow that reductions in health inequalities will take more than 20 years to achieve.

The new funding formulae for primary care trusts give substantially more money to PCTs in deprived areas.

1So health secretary Alan Milburn has, in theory, delivered on his promise.

However, it seems to be a case of 'the Lord giveth and the Lord taketh away'. The rate at which PCTs will move to their new target allocations is so slow as to largely negate the theoretical redistribution, and so hold back potential reductions in health inequalities.

A year ago Mr Milburn, in an interview on BBC2's North of Westminster, said: 'One of the things I want to achieve is to make sure that when We have got a big pot of cash, it goes to areas that most need it. And the areas that most need it are those areas that have got the biggest health problems.'

Mr Milburn is to be applauded for bringing in a more redistributive set of allocation formulae which move money to more needy areas. But unfortunately for the areas 'that most need it', the redistribution at the current rate will take over 20 years to complete. This is disappointing when the timescale could readily be reduced to five years without any radical changes.

The three new needs formulae were produced by a team of academics from various parts of the UK.

They cover acute and maternity (including community) services, mental health and prescribing. The method of derivation of the formulae is, as before, utilisation modelling using small areas, but new features include the use of components of the index of multiple deprivation, and also morbidity indices and welfare benefits data. Another novel feature is that the formulae attempt to take account of unmet need.

2But there are always two components to an allocation targeting process. The first is to research and set the target, and that tends to be the component that is publicised. The second is the decision on the speed at which the recipients - currently PCTs - have their allocation moved closer to target. This second component is usually not clearly publicised. It is rather like the small print in an insurance document - very important but easily overlooked.

Because historic allocations are rarely close to new targets, all PCTs are either under or over target - approximately half in each category. PCTs that are over target receive more than their 'fair share' as determined by the formulae, and they can be seen as 'winners'. Conversely, under-target PCTs can be seen as 'losers'. The total amount of money by which winners are over-target is mathematically balanced by an equal total amount by which losers are below target. The total imbalance is the sum of these two totals which, for 2003-04, is£2,171m - or£7.1m or 4.5 per cent per PCT.

Obviously if losers are to move up closer to target they require an above-average increase in allocation, and this must be balanced by an equivalent movement of winners down towards target by having a below-average rise in allocation. Because a lot of the imbalance is caused by the new needs formulae, winners are usually good health areas and losers are usually poor ones.

This speed at which allocations move closer to target fair shares goes by the technical name of 'pace of change' or 'rate of convergence'. Unlike the derivation of the formulae themselves which, after the decision to commission the work and the issuing of a broad remit, is a technical and statistical process, the setting of the pace of change inevitably constitutes a practical and political judgement.

One of the most important factors which feed into this judgement is the overall year-on-year increase at the disposal of the secretary of state. If the increase is small in real terms - ie after allowing for inflation - this will constrain the pace of change because it can be seen as exerting too much financial pressure on 'winners' to have their increase much below average. Consequently, in this situation it is perhaps understandable - even if not universally approved - that the political and managerial priority is to limit damage to winners, and therefore both winners and losers will move only slowly towards their targets.

The irony of the current slow pace of change is that it coincides with a series of large year-on-year increases in real terms. The total increases are 8.8 per cent for 2003-04, 9.5 per cent for 2004-05 and 9.1 per cent for 2005-06. These are high relative to inflation and would allow movement of PCTs to target several per cent per year, while still giving everyone a significant real increase.

But the minimum increases for any PCT have been set very high and very close to the average.

These minima - or 'floors' - are 8.3 per cent, 8.9 per cent and 8.5 per cent for 2003-04, 2004-05 and 2005-06. These are only 0.5 per cent, 0.6 per cent and 0.6 per cent below the respective average increases. These high minima greatly constrain the overall rate of convergence of PCTs to targets because they entail small reductions in distance from target, not only for the winners who have these high minima, but also for losers whose overall reduction must balance that of winners. The vast majority of PCT, 295 out of 304, get increases within a very narrow range - 8.3 per cent to 9.3 per cent, which is only 0.5 per cent away from the average of 8.8 per cent. This pattern is also typical of the latter two years.

At the pace of change set for the next three years, it will take 21 years to achieve 90 per cent of convergence. If the minimum increases were set lower at 1 per cent below average, 90 per cent convergence could be achieved in 10 years, or for 1.5 per cent below average seven years, or for 2 per cent below average, five years.

These hypothetical minimum increases are not very different from the average and they still yield substantial real increases for all PCTs, (eg even 2 per cent below average gives increases of 6.8 per cent, 7.5 per cent and 7.1 per cent for the next three years).Yet they allow a much faster convergence to the equitable fair shares. In particular this would be much fairer for the losing PCTs, and more in keeping with the government priority of reducing health inequalities. The great majority of underfunding for losing PCTs could be remedied in five years from now for a modest decrease in the minimum increase to 2 per cent below average.As it is they will have to wait over 20 years.

The 'headline' percentage increases over three years in the original press release are misleading as an indicator of the rate of convergence. The figures show a range of 28.1 per cent to 42.6, per cent which looks superficially to indicate a reasonable rate of convergence. Two factors distort the picture. This range is extended due to the inclusion of non-recurrent allocations for waitinglist capacity and high labour costs. Also, consideration of a range as a measure of distribution in this case is very deceptive.

Only a small number of highly under-target PCTs have increases much above average in any one year.

There are in fact only 12 PCTs with increases between 35 per cent and 42.6 per cent, another six with increases from 34 to 35 per cent and five with increases from 33 to 34 per cent. The great majority of PCTs, 281 out of 304, have three-year increases in the narrow range of 28 to 33 per cent, indicating a very slow rate of convergence.

It is hard to imagine that the NHS Confederation would be happy with this. It recently gave evidence to the Treasury select committee, which is looking at regional imbalance in public funding.

2There it stated that it would be anxious to see under-resourced areas move as quickly as possible to fair share.

Why, then, has the government decided to have such high minimum increases for over-target PCTs with the concomitant very slow pace of change? A succinct statement in the relevant DoH document gives the reason: 'so that PCTs have sufficient resources to deliver the improvements identified in the planning and priorities framework'.

3This framework includes improvements in access, service, outcomes and health inequalities. Clearly the government wants all PCTs to be given the chance to make improvements. This is understandable, but it needs to be balanced against the fact that the starting point varies between PCTs in terms of their funding 'fair share'.

According to the formulae, below-target PCTs will tend to have below-average access relative to their needs, and to have below-average health outcomes.

If health inequalities are to be reduced, such PCTs will have to improve at above average rate. These two factors of 'improvement for all' and 'faster improvement for the deprived' need to be balanced.

The current pace of change is skewed heavily towards the former, and it is not at all clear why.

A better balance would allow more progress towards the aim behind resource allocation of 'equal access for equal need' and the government's own priority of reducing health inequalities, while still allowing resources for improvements in all PCTs.

Table 1 (see over) illustrates the problem of the slow pace of change for PCTs (top 10 winners and losers). The gain or loss is that due to the distance from target expressed in terms of the composite targets and allocations. These figures contain an amount of about 7 per cent for GP costs which is currently controlled by the DoH, but this money is included in the method used by the NHS Executive to calculate distances from targets.

The 'years to target' column is an estimate of the time it would take to achieve the target, assuming the average pace of change set for the next three years continues. This may be an optimistic estimate because if the pace is slow in 'fat' years then there are bound to be leaner years in future when, if the political climate is still an extremely cautious one, the pace will be even slower. The last column, 'projected gain/loss', is the accumulated composite gain/loss over the years to target at the 2003-04 prices and total resource level. This is a conservative estimate since it takes no account of real increases in total resource after 2003-04.

In general, in table 1 the winners are wealthy areas with good health and the losers are poorer areas with relatively poor health. The gains to the winners are huge in both percentage and actual terms - particularly for the London PCTs.When this is combined with the estimate of years to targets, it can be seen that the cumulative gains are enormous.

Westminster is set to gain£1.75bn, and three other London boroughs to gain about£500m each. The timescale to equity is vast - from 53-24 years assuming the average pace of change between 200304 and 2005-06 is maintained.

The list of losers shows the other side of the coin.

Merseyside, which has one of the worst health records in England, has three PCTs in the bottom 10. Easington, at 20.2 per cent below target, is the worst off. Coincidentally, in the recently released census figures, Easington was revealed as the English local authority with the highest proportion of people reporting that their health was not good.

4Five of these bottom 10 PCTs show cumulative losses of over a quarter of£1bn.

Only three of the top 10 winners are from the north, but seven out of the top 10 losers are northern. Even outside the list of biggest losers, there are many other PCTs whose substantial losses would be much more tolerable if the pace of change were quicker.

One example of more modest losses from my own area is south Manchester PCT. The loss for 2003-04 is 3.55 per cent or£5.4m.A PCT board could presumably accept this as a temporary necessity during a period of readjustment to new targets.

However, the pace of change is so slow that the word 'temporary' changes to 'almost permanent'. The estimated time to target of 22 years gives a total accumulated loss of£64m. Is this acceptable in times of large real increases in total funding?

Table 2 is the equivalent data for strategic health authorities (top five winners and losers), with the estimated time to target projected using aggregates of PCT rates. At this level the gains and losses are huge, often of the order of£1bn for the projected total. It is clear that regionally overall, London is gaining at the expense of parts of several other regions - the Midlands, the north west and the north. The net annual gain to London is£304m and the projected gain is£4.4bn. This might point to the political difficulty Mr Milburn would experience in raising the pace of change and thereby correcting the over-funding of London at a faster pace.

However, the government has an implicit commitment to this movement. It therefore seems absurd that, for example, Westminster PCT, which spends 30 per cent more than its fair share, is given large annual increases of 8.3 per cent, 9 per cent and 8.6 per cent over the next three years.

When looking at the north-south picture, it is pertinent to continue the Mr Milburn quote we began with: 'There is no doubt when you look at the figures, which are very clear indeed, there is a big north-south divide on health, and that is something we have got to tackle.' For 2003-04 the north is 1.7 per cent under target and the south is 1.7 per cent above, giving a relative difference of 3.4 per cent, which is a difference of approximately£820m.

The slow pace of change reduces this difference to 3.2 per cent for 2004-05 and 3 per cent for 2005-06.

These percentages translate into monetary differences of£820m,£850m and£860m respectively (these actually increase because of the large year-onyear increases in overall allocation). This rate of change is not in keeping with the above quoted concern about the north-south divide.

Allocations which favour deprived areas are important in reducing the persistent - and in some cases, worsening - health inequalities in England.

The subject of health inequalities is very much a current government priority. The DoH has established a new health inequalities unit and the prime minister will chair a Cabinet meeting to oversee the production of a detailed national programme of action on inequalies.

To quote Mr Milburn again (November 2002): 'But our actions are as determined as our ambitions are bold: to do what no government has ever done - to improve the health of the country as a whole and to improve the health of the worst off at a faster rate still.'

5There is little doubt that the new allocation targets set by the new formulae are a commendable step forward towards improving the health of the worst off, but 'determined' or 'bold' are not words that come to mind when one sees the tortoise-like pace of change towards the new targets, and the consequent delay in reducing inequalities.

REFERENCES 1Secretary of state for health. Power and resources shift to NHS frontline.

Department of Health.

December 2002 2NHS Confederation.

Uncorrected evidence to the Treasury select committee.

December 2002.

3Secretary of state for health. Primary care trust revenue resource limits 2003-04, 2004-05 and 200506.HSC 2002-12 Annex

6 Dec. Department of Health, 2002

4Office of National Statistics. Census 2001. Key statistics for local authorities. February, 2003

5Secretary of state for health. Speech to Faculty of Public Health Medicine.

November 2002.

Key points

At the pace of change set for implementing the new PCT funding formula over the next three years, it will take more than 20 years to achieve equity.

If the minimum increases were set lower it would largely be achieved within five years.

The current state of affairs perpetuates the north-south health divide.

John Hacking is research officer, Manchester Joint Health Unit.