Spoilt for choice

This content relates to the following topics:

Patient choice sounds great in theory, but we have to take care to avoid rushing into an untried and uncertain solution to the problems of the NHS, warns John Appleby.

I am not an opponent of choice: I am more than happy to live in a capitalist democracy and to exercise my rights to choose a government through the ballot box and to choose the things I consume through the market.

But I am also aware of the limits and failings of choice. I can vote, but the party I vote for doesn't always get into power. And I can go to any one of hundreds of car mechanics – but did my car really need all its shock absorbers replaced?

A key question, therefore, is not the simplistic one – choice: good or bad? Rather, it is whether the costs – and not just the financial ones – justify the benefits. Another way of framing this question is whether patient choice is not just an effective, but a cost-effective, way of achieving the benefits.

The latter, according to Julian Le Grand's recent Public Finance article – 'A better class of choice', March 31-April 6 – include, in theory at least, more equitable access to quality health care services across social groups and an 'embedded' incentive mechanism to pressure providers to improve the efficiency and quality of their services.

But we need to be even more specific. Too often in debates about patient choice, what is being chosen and the necessary limits to choice, are left indeterminate. While NHS patients do not face the financial consequences of their choices, the NHS does: it has a finite budget. So, regardless of the demands of patients, a choice of an unlicensed high-cost drug of dubious effectiveness is not the sort of choice that should be on offer.

Indeed, the current policy on choice at the point of GP referral is in many ways highly restrictive. It is a choice – from a limited set of hospitals – of outpatient department. It is not a choice of surgeon, or of treatment, or of any of the myriad possible choices that could be imagined along a patient’s pathway of care. Further, hospitals that look likely to miss their waiting times targets due to their popularity among patients will be removed from the choice 'menu'.

Le Grand builds a case for choice based partly on a description of a failing NHS, with long waiting times, haughty professionals and dastardly monopolistic practices. But it is too easy to slip into hyperbole, exaggerating the decrepitness of the current situation, when you have a new policy solution to promote.

This is an over-reaction to what Karl Popper termed 'solutioneering': the jumping to a solution without properly setting out the problem the solution is meant to address. Perhaps we need a new term to describe this: 'problemeering'?

It is, of course, not true to say that there has never been choice in the NHS – its very creation opened up access to health care for millions of people who were previously unable to afford even the basics. And many exercised a right to be referred to the consultant of their choice (equally, consultants could exercise their right not to accept a patient). And once in the system, good medical practice has always been to set out treatment options – including the choice of no treatment.

The system was not and is not perfect. There is a strong grain of truth in Le Grand's assertions on the failings of the NHS. But whether patient choice as currently conceived is the panacea for these is not as straightforward as Le Grand suggests.

What we need is some evidence of what might happen when patients really choose. On the impact of choice on equity of access to high-quality care, a recent and soon-to-be-published experimental study by the King's Fund, Rand (Europe) and City University suggests that in making a choice at the point of referral, different groups value the key factors that influence their choices differently.

For example, we found that those with formal educational qualifications placed a higher weight on the impact of treatment on their health than those without qualifications. If this leads to the better educated choosing better hospitals (as with schools), it is hard to see how the less well educated (and poorer) will benefit.

At the hospital end, competition and Patient Choice might bring about the dramatic improvements in NHS productivity and efficiency that Le Grand states. But, irrespective of choice, payment by results with a national average cost tariff might turn out to be a much bigger incentive for above-tariff hospitals to reduce costs.

A welcome change that choice might bring is a greater impetus to produce the information patients need to make choices – in particular on health outcomes. Again, irrespective of patient choice, such data would vastly improve PCTs' commissioning role.

Incentives are, of course, important in improving performance in the NHS, but it makes little sense to emphasise the value of just one incentive mechanism. What impact it might have depends critically on other measures – the payment system, the right sort of information in the system and so on – which themselves are likely to produce many of the benefits Le Grand attributes solely to choice.