During this year's election campaign, prime minister Tony Blair repeated his promise that by 2006 spending on healthcare in the UK would be brought up to the European average. He told Newsnight interviewer Jeremy Paxman in June that '...by the end of the second comprehensive spending review [2006-07], I wanted to reach the European average'.
Since Mr Blair made his original commitment on BBC's Breakfast With Frost in January 2000, there has been debate over what exactly he meant. Most commentators agree that the figure for comparison should be a weighted average of European spend and, more important, based on the latest estimate of European health spend in 2006. The government prefers a simple average and seems wedded to what is currently being spent. It is no coincidence that this gives a lower target.
But regardless of the exact target figure, there would be universal agreement that the key consideration is how any extra money allocated to healthcare is spent. While it may be true (in the rather dismal economic sense) that problems cannot be completely solved by throwing money at them, extra resources can certainly improve things - if used in the right way.
The strong implication of the prime minister's aspiration to match other EU countries in terms of funding is that by doing so the NHS will also match other countries' levels of service, quality and, indeed, health outcomes. But despite the NHS plan's shopping-list mantra of more doctors, nurses, equipment and facilities (which just about covers everything), meeting an internationally derived spending target will not, on international evidence at least, necessarily lead to a similar increase in the things that really matter - health and quality of care.
The key question arising from all this is whether the rest of Europe enjoys better healthcare and health outcomes from the extra money spent on healthcare. Unfortunately, there is no simple answer. Though the prime minister's pledge must be based on some such notion, no convincing evidence has been advanced - by government or others.
To begin to answer this conundrum requires: a comparison of the composition of UK and EU spend on healthcare; a comparison of UK and EU outcomes (both in terms of outcome and process of care); an analysis of the link between spend and outcome.
How does UK health expenditure compare with the EU?
There has been a rise in the share of national wealth consumed by healthcare in both the EU and the UK since the 1960s. But a significant gap remains between the UK and the rest of the EU in terms of average proportion of GDP spent on healthcare.
Our calculations suggest that by the end of the next parliament, in 2006, the UK will remain significantly below the average for the rest of the EU (see figure 1), based on the latest Organisation of Economic Co-operation and Development health data set for 2000.
Only two EU countries spent less in terms of proportion of GDP on healthcare than the UK in 1998 (figure 2). This is partly explained by a lower level of private expenditure. Of total UK health spend, 84 per cent is publicly funded compared with 16 per cent privately. In the rest of the EU, the ratio is 76:24.
But as figure 2 shows, most EU countries also spend more from the public purse than the UK. Of those spending more on healthcare overall, Belgium, Denmark and Sweden rely on a similar high level of public funding.
Nevertheless, if the UK is to achieve the EU average, we are probably going to have to increase both public and private expenditure.
Bringing the UK up to the EU average If, in 2001, the UK spent the same as the rest of the EU is projected to spend, how much extra would this be?
Using the weighted average approach, the UK would have to spend an extra 2.8 per cent of GDP (£28bn). But using the government's simple average, the figure is around£18bn.
The question that must be addressed is: what is the right breakdown of expenditure to meet the health needs of the UK population?
While it is possible to identify differences in the level and composition of the financial inputs to EU countries' healthcare systems, this reveals nothing about the way money is spent. This is not to say that there is no relationship between sources of finance, the type of resources purchased and, eventually, health outputs and outcomes. But this chain of relationships is extremely hard to piece together.
Spending habits So where does the money go? This year, the total spend on healthcare in the UK is estimated to be£69.5bn, of which£58.5bn (84 per cent) is from public funds and£11bn (16 per cent) from private sources.
The various ways of looking at how the money is spent essentially amount to two types of breakdown: by inputs to, and by outputs from the healthcare system.
Looking at inputs to healthcare in the UK in 1998-99, 65 per cent of public expenditure on hospital and community health services was spent on labour, and 35 per cent on consumable items such as drugs, medical equipment and fuel.
From an output perspective, around 54 per cent of this expenditure was in the acute sector, 12 per cent on mental health, 5 per cent on learning disabilities, 10 per cent on elderly people, 5 per cent on maternity and 14 per cent on other care. These allocations partly reflect the relative costs of providing care in different sectors, but also different priorities (mostly arrived at in an ad hoc fashion over the years).
These are only crude breakdowns of total spend, but even at this level it is hard to make comparisons with our European neighbours.
How does the EU spend its healthcare resources?
Unfortunately, available data is simply not good enough to enable proper comparison, but there is work going on to allow this. For example, the World Bank and the OECD are promulgating ways of producing healthcare accounts.
But we are far from achieving consistency across the different countries. The government's commitment to extra spending is based on a belief that absolute increases are good, rather than on any detailed examination of how the extra money would or should be spent.
Of course, the NHS plan has stated that the UK should spend more on staff - particularly doctors and nurses.
And, comparatively, there appear to be strong reasons for doing so. Total healthcare employment as a percentage of total employment in the UK is low: 4.6 per cent compared with 7 per cent in France and 6.3 per cent in Germany.
And estimates of the numbers of doctors and nurses per 1,000 population in the EU in 1996 (figure 3) also show that the UK lags behind many other countries (though it should be noted that it is difficult to ensure complete consistency of data across countries).
Converting these rates to absolute differences emphasises the yawning clinical staff gap between the UK and the rest of Europe. To achieve German levels of staffing, for example, the UK would require an additional 100,000 doctors and 300,000 nurses.
It is, of course, possible to spend more on staff without increasing inputs merely by increasing input prices (the wages of doctors and nurses). Indeed, there is some evidence that this is already happening.
The outcome of the national beds inquiry also showed that the UK needed more beds. Again, the UK compares unfavourably with other European countries on the number of acute beds per head of population: around 2.4 compared with 4.5 in France and 7.3 in Germany.
There are also considerable differences in the availability of equipment such as CT scanners and MRI units in different European countries. Unfortunately, comparable figures for the UK are not available.
But even though extra spending translates into more basic resources such as beds and staff, is it possible to draw any firm conclusions about the impact such resources have on healthcare processes and outcomes?
Comparing process and outcomes As with many other comparative issues, comparing the process element of healthcare is deeply problematic - not least because international data is so poor. While, for example, the UK is at the forefront in measuring waiting lists and times (despite the recent National Audit Office report), most other EU countries either fail to record this sort of information at a national level or, in the case of Sweden, consider it a confidential part of the patient record.
Health outcomes are more easily compared, however. Does the EU enjoy better health outcomes than the UK? And is there any link with healthcare spending? The data in the table provides a mixed picture.
The UK ranks low on spending, but not as badly on some measures of population health as such comparatively low spending would perhaps suggest. Equally, Germany, for example, does not seem to perform as well as might be expected given the size of its healthcare investment.
Such apparent anomalies should not be surprising. Nor, for example, should the high-level finding that the number of doctors per 1,000 population appears to have no connection with infant mortality rates.
This sort of evidence should not be misinterpreted to imply that, from a health point of view, there is no point investing in more healthcare staff (or, for that matter, in allocating more money to healthcare in general). Rather, it highlights the complicated relationships between inputs and population health outcomes. It also highlights the need to think about investment in non-healthcare sectors in order to effect changes in people's health.
The government's commitment to spend more on healthcare in the UK is welcome.
But this needs to be based on more than a knee-jerk reaction to 'spend as much as our European neighbours'.
While the NHS plan goes some way towards indicating how any extra money might be spent, it is too broad a brush (more nurses, more doctors) to give any confidence that systematic connections have been made between the extra inputs purchased, the resulting increase in health sector activity, and improved outcomes for the patient.
John Appleby is director of health systems at the King's Fund. Sean Boyle is visiting research associate at the London School of Economics.
The government is committed to raise spending on healthcare in the UK up to the European average by 2006.
There is debate about the exact target figure, but agreement that the key consideration is how the extra money will be spent.
If the UK is to achieve the EU average, it will probably have to increase private as well as public spending.
To match German levels of staffing, the UK would need an extra 100,000 doctors and 300,000 nurses.
There has been little analysis of how the extra expenditure will improve outcomes.