Skip to content

This content is more than five years old


Ministers, not NHS England, should decide on the affordability of cost-effective new treatments

The NHS, in England at least, can no longer afford to fund cutting-edge technologies to the extent that it has over the past couple of decades – even when those new treatments have been judged to be cost effective.

That is the message from the recent consultation launched by NHS England and the National Institute for Health and Care Excellence that might, in shorthand be dubbed, ‘NICE, affordability, and the NHS.’

It is just one more example of the funding pressures that the NHS is under. And potentially it has big implications for patients. But the message is being delivered by un-elected NHS organisations when it should be ministers – the budget setters – who first of all acknowledge that is the reality and then take responsibility for the decisions that flow from it.

What are the proposals?

The proposals in the consultation are in three parts.

First, NICE plans to introduce a ‘fast track’ appraisal process for new technologies and treatments (though we are talking here chiefly about pharmaceuticals) that cost £10,000 or less for a quality-adjusted life year, otherwise known as a QALY, to allow quicker access to them.

The cost per QALY is the tool that NICE uses to judge the cost effectiveness of new technologies. To put it simply, it is the cost of each additional year of life gained by a treatment or procedure not just in terms of extra months or years of life, but in the quality of that life: defined, for example, by freedom from or reduction in pain, or the ability to do basic activities of daily living such as feeding oneself and being mobile, or having a decent state of mental health.

Currently, for its standard technology appraisals, NICE uses a threshold of £20,000 to £30,000 per QALY. Treatments that come in at or below the threshold are usually approved. Those that come in above it tend not to be judged cost-effective, and so are not recommended for adoption by the NHS. But it is important to understand that the threshold is not a fixed point, and the QALY is an aid to judgement about cost effectiveness, not in itself the judgement. NICE has on occasion approved treatments that cost more than £30,000 per QALY. In addition, since 2009, there has been a higher threshold of £50,000 per QALY for end-of-life care, which in practice has applied chiefly, although again not exclusively, to cancer drugs.

Part two of the proposals, although this is not the order in which NICE and NHS England presents them, is that for the first time there should be a new approach to treatments for very rare conditions. These can be immensely expensive. The proposal is that treatments costing up to £100,000 per QALY should be funded automatically from routine commissioning budgets, and that treatments with a higher cost will be considered through NHS England’s process for prioritising other highly specialised technologies. This is a means of providing faster access to treatments for very rare conditions that come in at or below the £100,000 threshold. The rest of this paper does not address that part of the proposal.

The third part of the consultation – and the key one as far as this commentary is concerned – is the proposal that special arrangements be put in place to manage the impact on the NHS budget of these new technologies. That a new ‘budget impact threshold’ of £20 million should be set.

When NICE estimates that a new cost-effective technology is going to cost the National Health Service £20 million or more in any of its first three years, NICE will ‘signal the need for a commercial agreement between the company and NHS England’. That will initiate a negotiation between NHS England and the company to seek a lower cost for the new treatment.

In effect this is likely to mean a (commercial in confidence) price discount, although the lower cost could be achieved in a number of ways. If the negotiation gets the cost to below £20 million, NICE will recommend adoption under the standard NICE and NHS rules – namely, that commissioners and providers have to provide the new treatment within 90 days.

If, however, such an agreement cannot be reached, NHS England will be able to ask NICE to allow longer than the standard 90 days for adoption. In other words, NICE will recommend how quickly or slowly the new treatment can be adopted ‘without causing disruption to other services’.

There is already a process in place, known as a patient access scheme, under which the Department of Health negotiates a lower cost for a new treatment or procedure so that it meets NICE’s cost-effectiveness criteria. There are some 80 of these in existence, and they involve a variety of approaches – anything from a straight discount to the company providing a product free if it does not work in an individual patient, or the NHS paying for initial courses of treatment after which the company subsidises them or provides them free, or the other way around.

The difference under the new system is that even where NICE judges a treatment to be cost effective, if it goes over the budget impact threshold and the cost cannot be lowered, NICE will be asked to advise the NHS on how it should be phased in. In other words, despite a new technology being judged cost effective, some patients at least will have to wait to gain access. How many and for how long will clearly vary from case to case depending on the scale of budgetary impact.

In other words, for the first time since 1999, when NICE began operating, the NHS is creating a system that reflects the fact that it cannot afford full and immediate implementation of a new technology that NICE judges to be cost effective.

Why is this happening now?

Because of the money, is the short answer. NHS England, which has set the ‘budget impact threshold’ of £20 million, judges that if it cannot negotiate costs below that, it cannot afford to fully fund new technologies immediately_._

The reason for that conclusion is that the pharmaceutical bill has been rising considerably faster than the overall increase in NHS expenditure.

Since 2010, and on current projections of NHS expenditure, spending per head over the decade to 2020 is rising at less than 1 per cent per head in real terms, and a growing percentage of that extra cash is taken up by increases in prescribing costs – more than 30 per cent in the past two years. This might be acceptable if other parts of the service were not already struggling with their finances, but, of course, they are.

Percentage of Department of Health cash growth accounted for by increases in prescribing costs

Percentage of Department of Health cash growth accounted for by increases in prescribing costs

Source: King’s Fund analysis; NHS Digital 'Prescribing Costs in Hospitals and the Community, England, 2015/16'; HM Treasury, Public Expenditure Statistical Analyses 2016

To be fair, these figures may slightly overstate the case as they do not include the rebates due under the Pharmaceutical Price Regulation Scheme (PPRS) – the agreement between the UK government and the UK pharmaceutical industry that applies some control over the prices of, and spending on, in-patent medicines. The current version of this scheme includes a system of rebates but the scale of these is not in the public domain.

Furthermore, most of the recent growth in the drugs bill is concentrated in the hospital sector and very largely in specialised services – for which NHS England is directly responsible and on which many of NICE’s technology appraisals impact.

Since 2012/13, hospital prescribing costs have risen steadily – way above any percentage cash increase in NHS spending – while the rise in the cost of general practice prescribing actually fell in 2012/13, before rising by much smaller percentages.

Estimated prescribing cost from each sector, 2010/112015/16

Estimated prescribing cost from each sector, 2010/11–2015/16

Source: NHS Digital

So, from where the Department of Health sits – and it is notably absent from this consultation – and from where NHS England sits, there is clearly an affordability issue.

Is affordability a new problem?

On one level yes, and on another no.

‘No’ in the sense that spending in any health care system, however funded, has a limit. ‘No’ in the sense that in the years leading up to the creation of NICE the NHS was – as it has been at various times in its history – chronically short of money. In 1995, the chair of the then NHS providers’ organisation went public to say that the NHS needed to be reduced to a ‘safety net’ for the old and weak, given the pressures of an ageing and growing population and medical advance (see A terrible beauty). In 1996, the ‘rationing agenda group’, which included leading health economists, consultants, junior doctors and political funders, convened by an editor of the British Medical Journal and with its funding underpinned by The King’s Fund, declared that ‘rationing in health care is inevitable.’ At the time a number of the equivalents of clinical commissioning groups were striking treatments off the list of procedures for which they would pay.

NICE, which became operational in 1999, finessed that argument. Rather than trying to construct a list of things for which the NHS should and should not pay, or people whom it would and would not fund (the weak and elderly versus the healthier and richer, for example), the creation of NICE reframed the argument into one which said that the NHS should seek to pay only for treatments that are cost effective.

Also ‘no’ because some health economists would argue that in a pure world the budget sets the threshold and vice versa. If the budget is very low, new treatments have to be very cost effective to be affordable. In the case of the NHS, for example, health economists at York have calculated, through some fairly heroic manipulation of NHS data, that the average cost per QALY of NHS treatments is around £13,000 and thus that NICE’s standard threshold of £20,000 to £30,000 is too high. It is, by implication, squeezing out other cost-effective treatments. But the range of cost per QALY is enormous, from a few pounds for an antibiotic that prevents an infection becoming life-threatening to many hundreds of thousands of pounds per QALY for the treatment of some rare diseases.

But the answer to the question ‘is affordability a new problem’ is ‘yes’ when looked at another way. Or perhaps, more accurately, it is now a pressing issue.

NICE was born into sunnier times. Shortly after it was established, Tony Blair pledged to get NHS spending up to the EU average as measured by shares of gross domestic product – a goal that by the late 2000s was broadly achieved. So while NICE’s recommendations undoubtedly added to NHS spending, they proved affordable within a significantly expanding budget that did not immediately require less cost-effective treatments to be sacrificed to pay for the new.

Since 2010, that position has changed. So affordability in that sense is a new problem.

For example, in 2015 NHS England delayed the funding of new treatments for hepatitis C, despite NICE’s recommendation that they were highly cost effective. Implementing that recommendation overnight would have added around £500 million to NHS spending in-year – a significant sum of money even in a £100 billion budget. So the introduction of these new treatments was phased, initially being concentrated on those in immediate need in those parts of the country best equipped to handle it.

So who is responsible for affordability?

This is a key question. When NICE was originally created, responsibility for the affordability of its recommendations was reserved for ministers.

But NICE did not exist until 1999. And since 1991, with the introduction of the purchaser/provider split, much of the responsibility for affordability has in fact lain with commissioners – in their current form with clinical commissioning groups, and with NHS England for specialised commissioning.

Even since 1999, not every new technology – new type of imaging, new surgical technique or even new pharmaceutical – has gone through NICE, although the drugs that are likely to be most costly to the NHS almost invariably do.

So commissioners, working with providers, have always had to decide how to afford the new – and thus, on the occasions when money was tight, what they will stop paying for to make room for that. Sometimes that has been done been very quietly, and sometimes more publicly when commissioners have adopted initiatives such as the 'Croydon' list of procedures for which they would not pay.

So it could be argued that the creation of NICE, and the requirement that its recommendations be funded, has in fact advantaged one group of treatments – notably very expensive pharmaceuticals – over other new treatments because, once NICE has pronounced them cost effective, they legally have to be made available on the NHS. So all this consultation is doing is redressing the balance slightly and putting affordability decisions on new treatments back where they belong – with the commissioners, and in this case NHS England.

But the point of creating NICE was to tackle the introduction of very costly new treatments by ensuring that they were cost effective. And the fact remains that the affordability of those decisions was reserved to ministers.

Proof of that came in 2009 when it was decided that a higher threshold – £50,000 per QALY – would be applied to end-of-life care. This was a government policy, implemented in Alan Johnson’s day as secretary of state for health. But as it would clearly add further to NHS costs, Sir Michael Rawlins, the chairman of NICE, had to write a formal letter to Alan Johnson outlining that fact. Unsurprisingly, given that this was government policy, Alan Johnson decided that this was in fact affordable, and wrote back agreeing (see A terrible beauty).

Similarly, in 2010, the coalition government established the Cancer Drugs Fund reflecting a deliberate decision to spend more per head on cancer patients than others, by approving treatments that NICE judged not to be cost effective. It was ministers who took the decision to find the money for that – to afford it, within the NHS budget.

Somewhere along the line, however, the responsibility for affordability in such cases appears to have slid across entirely to NHS England – without that ever being, as far as I can establish, announced by anyone. This is a large and significant change.

What is the consultation aiming to do?

The stated aim of the consultation is to speed up access to new technologies that are judged to be cost effective. So funding of treatments that fall below the new £100,000 per QALY threshold for very rare diseases would be automatic, and funding of treatments that cost less than £10,000 per QALY would be fast-tracked, although these will also be subject to price negotiation if the cost in any of the first three years is expected to exceed £20 million. Getting that price negotiation out of the way before NICE issues final guidance is also intended to speed up adoption of the treatments.

What is far from clear, however, is why the trigger of £20 million has been adopted as opposed to £10 million, or £100 million, or some other figure?

The only available answer seems to be NICE’s calculation that in the year to June 2016, only 13 of the 62 technology appraisals it completed (20 per cent) would cost the NHS more than £20 million in any of the first three years of implementation. And while the cumulative cost of these technologies is not in the public domain, it is a reasonable guess that they are likely to be in the £300 million to £400 million range over three years.

Now the hope clearly is that in the price negotiation, the industry will be sufficiently flexible to find ways to lower the cost of implementation in the early years to bring it below the £20 million figure. And there are signs that the industry appreciates that health budgets are under pressure, not just in the UK but more generally, and is willing to explore more flexible pricing strategies.

So that would suggest that the number of cases where NHS England asks NICE to recommend a phased introduction of a cost-effective technology will be small. That, however, is completely unknowable at this stage. Furthermore, under these proposals neither NHS England nor NICE will be able to say ‘no’ to a technology judged to be cost effective. The treatment will be adopted, but in a slower and phased way than under the current standard 90-day adoption rule – although how slowly and how phased and affecting how many patients is again unknowable at this stage.

It is worth noting here that NICE already can make – and has made – exceptions to the 90-day adoption rule. But to date those have been where hospitals need to recruit or train additional staff, or obtain new equipment, or adapt buildings to implement a recommendation. In other words the delay has been for managerial reasons, not on the ground of affordability.

What is being proposed now is that – hepatitis C aside – for the first time for almost 20 years, some patients will be denied access, at least for a time, purely on the grounds of affordability to a new technology that has already been judged to be cost effective.

Affordability is clearly an issue, which, unless more money is provided, has to be addressed. These proposals are one way of doing it. But the reason the issue has arisen is because the money is so tight - and the decision about how much is spent on the NHS is a political decision, not a managerial or technical one.

If price negotiation fails to reduce the cost of these new technologies, and NHS England decides that they need to be phased in so that their cost does not cause disruption to other NHS services or break the budget, it should be the secretary of state who instructs NICE to devise a phasing-in scheme, rather than NHS England requesting it.

Ministers are responsible for the overall NHS budget. If the NHS can no longer fund new and cost-effective treatments, ministers should announce that decision case by case, and be held accountable for so doing.

Is this a sustainable solution?

It is also worth asking whether this is a sustainable solution. The answer is probably not. Phasing in a new cost-effective technology will delay the cost to the NHS, not remove it. In a health service facing extremely limited growth for the foreseeable future, these additional costs are likely to be met only by de-commissioning ineffective or less cost-effective treatments to free up funding for the new.

Finding entire services or therapies to decommission, however, has never proved easy, even when NICE has sought to do that (see Garner and Littlejohns). Which is not say that the existing budget cannot be spent more cost-effectively. The Academy of Royal Colleges recently launched the ‘Choosing wisely’ campaign, which identified 40 treatments that bring little or no benefit to patients, for example. Shared decision-making between patients and doctors to avoid over- or under- treatment can lead to a more cost-effective use of resources that is in the patient’s as well as the service’s interests. The Department of Health and NHS England have an extensive programme under way called ‘Get it right first time’, originally launched for orthopaedics but now being extended to 18 further specialisms, which appears to be producing genuine savings and spending NHS cash more cost effectively. Many other initiatives are under way.

So there is room for manoeuvre. But, even under these proposals, the extent to which the NHS will be able to provide new treatments will depend either on the current money being spent more effectively or on the government finding more of it for the NHS. And if the NHS cannot stay abreast of these developments, it should be for ministers to announce that, case by case.

This commentary is in part based on a Chatham House-style roundtable held at The King’s Fund on 22 December 2016 that was attended by representatives from NHS England, NICE, a number of pharmaceutical companies, the patient organisation National Voices, the Office of Health Economics and by John Appleby, former chief economist at The King’s Fund and now research director and chief economist at the Nuffield Trust. The Fund is immensely grateful to them for their time and their help, but the views expressed here are personal.