As the NHS notches up 50 years this summer, Alan Maynard says it should not get any extra cash but should use its Pounds 42 billion budget better.

Every few months there is another lament about the National Health Service and someone describes it as being in "crisis". The remarkable thing about the NHS is how well it works despite its detractors, who tend to ignore that resources are always scarce, that rationing is ubiquitous and that death and decay are the human lot.

The NHS is paid for by taxes, is largely free at the point of use and is assailed by demands for more money. If we assume that the purpose of the service is to improve the health of the population and not merely to produce jobs and profits for doctors, nurses and supply industries such as the pharmaceutical industry, then the main policy issue is how to target Pounds 42 billion of public money to treatments that can be shown to be cost-effective.

Much health care has no scientific basis. It is provided in hope rather than in certainty by carers who often find it difficult to diagnose an illness, let alone identify a treatment that has been shown to work. Eighty per cent of health care expenditure goes on therapies of unknown cost-effectiveness, 10 per cent goes on care that damages us and 10 per cent on care that improves our health. The problem is that we do not know which therapies are in the 10 and which in the 80 per cent categories.

³ÉÈËVRÊÓƵ

Despite advocacy by radicals such as A. L. Cochrane 25 years ago, medicine has not been proficient at testing medical treatments in well-designed, randomised, controlled trials. In recent years the Cochrane Collaboration has begun the enormous task of systematically reviewing evidence and encouraging doctors to practise "evidence-based medicine''.

At the same time, the demand for health care is apparently increasing. Usually we are told demand is increasing because of demography, technological "improvements'' and public expectations. When analysed, however, these factors can often be demonstrated to be exaggerated claims by medical lobbies trying to improve their lot. All too often such groups behave as if the health-care system's purpose is to provide them with social security rather than to improve population health.

³ÉÈËVRÊÓƵ

Some years ago a group called Healthcare 2000 produced a report, disseminated by its funders Glaxo Pharmaceuticals, that argued that the NHS was "in crisis'' due to the increasing gap between the demand for medical treatment and its supply. Assuming naively that the supply of health care was efficient and that part at least of the gap could not be met by improved practices, the group concluded that more cash was needed. As this could not be paid for from taxation, they argued for the use of private funding. This nice collection of non sequiturs was based on poor evidence about the demand for health-care.

The demand for care is driven by advocacy of new drugs, equipment and techniques, in the absence of evidence about their efficacy. The government should tackle this problem by requiring producers to demonstrate the cost-effectiveness of innovations before they are routinely purchased by the NHS. Such a policy is overdue but will, nevertheless, lead to claims by producers that lives are being threatened by delays in the use of their products. This assertion can be translated as a demand to use unproven products on vulnerable patients at public expense.

We are constantly subjected to producer propaganda, delivered by an uncritical media. BBC TV's Tomorrow's World portrays the "wonders'' of modern science often on the basis of poor data about the effectiveness of some treatments and with no attention to cost. Thus industry can test for prostate cancer but the test is inaccurate, and even if it is accurate, patients have to recognise that treatments for advanced tumours are often of little benefit in terms of longer, better lives. Despite such facts, industry is busy profiting from such tests.

The Healthcare 2000 group and others that regularly "pop up'' marketing "doom and gloom'' usually offer an incomplete analysis of expenditure needs and conclude that they can be met only by new (non-tax) funding methods. But alternative financing methods have different effects on behaviours and outcomes. Typically governments are interested in three health-care objectives: expenditure control, efficiency and equity. Most European health economists agree that the best way to maintain control over spending is to have cash-limited budgets funded from taxes.

Without cash limits and "single pipe'' tax finance, international experience shows that attempts to control one route of (for example, public) expenditure, leads to others (for example, private insurance and user charges) increasing. Thus fragmented funding arrangements, like those in the United States and those proposed by Healthcare 2000, facilitate cost inflation. Healthcare 2000's advocacy of charging NHS patients can be interpreted as an attempt to increase spending on health care and thus benefit health-care providers.

In the UK the demand for health care exceeds the supply and so implicit and explicit rules are used to ration access, although politicians refuse to accept that rationing exists in the NHS. Increasingly social scientists, and even clinicians, are demanding a public debate about rationing and the development of a consensus about the principles by which access to health care be decided. Clinicians make rationing decisions every day in the NHS but their choices are inconsistent. If rationing is to be explicit,what criteria should determine access to care and cure?

³ÉÈËVRÊÓƵ

One principle for determining access is "value for money'', that is, targeting resources at patients who get the greatest benefit per unit of cost. This principle causes discomfort for some clinicians trained to pursue the ethic of treating the patient in front of them. When we are that patient, we want them to treat us to the limits of effective therapies, regardless of cost.

But a decision to treat me is one not to treat you. Resources are finite and every treatment choice wipes out the possibility of treating another patient. Thus there is a conflict between the ethic of the doctor and the social ethic of the efficient use of resources. Doctors are reared in "fact factories'', learning soon-to-be-redundant knowledge, where their analytical skills are retarded and they are taught nothing about the economic consequences of their trade. It is long overdue that doctors be instructed in the facts of life of the NHS.

³ÉÈËVRÊÓƵ

Is there a public consensus that NHS resources should be targeted on those patients who can benefit most or is rationing more complex? Society may have social values that override efficiency considerations. Society may, for instance, wish to favour the poor or discriminate against the elderly.

Alan Williams, a York health economist who is over 70 years of age, argues for "fair innings'' equity. He argues that society may decide to transfer resources from the efficient treatment of the elderly to the inefficient treatment of, for instance, a chronically ill young person who has not had a "fair innings''. Does society share such values? How much money should be transferred from the old to the young? Could such a principle be translated into practice? A public debate is needed to determine rationing principles. Or perhaps the British prefer rationing by post code because decision-makers do not agree about "the facts'' and use different principles of resource allocation? Politicians claim they do not want such inequities - in which case they need a public debate about rationing medical treatment.

A single mother with four children on an estate in Hull smokes to stay sane and is advised by the NHS to give up cigarettes to avoid cancer. Her decision, sanity now and shorter life later, is rational. To what extent should nice middle-class folk impose their values on poor rational folk whose lot can be nasty and brutish but ameliorated by "poor'' habits that maintain short-term sanity?

Health promotion may be cost-effective but how far should such policies be pushed when the poor's circumstances are the major cause of their behaviour? This question illustrates nicely the issues surrounding inequality. A child born today in leafy Surrey will live seven years longer than a child born in a poor area of Merseyside. Such health inequalities can be reduced by evidence-based interventions but it may be more effective to reduce the deprivation associated with poor housing, unemployment and bad education.

The NHS lament is always with us. The service should use its Pounds 42 billion budget better: the case for more money, in the face of continuing variations in practice and efficiency, remains weak. Additional investment to improve health should be spent outside the NHS to reduce inequality and give the poor reasons for investing in their long-term health.

Alan Maynard is professor of health economics at the University of York.

³ÉÈËVRÊÓƵ