The publication of the Department of Health’s consultation paper on value based pricing and the ongoing consultation on the Cancer Drugs Fund plus each consultation’s accompanying impact assessment mean that it is now possible to see what is being proposed. This blog looks at the essentials of value based pricing; later blogs will deal with more particular issues.
According to the consultation paper, value based pricing has three aims:
- Better access for patients to drugs
- More innovative drugs
- Better value for the NHS (para 2.1).
Given that these may be incompatible, the consultation paper struggles to reconcile them. More precisely, each aim involves trade-offs with the others, assuming (as economists do) that all else remains unchanged. Better access for patients to high priced drugs currently refused by the National Institute for Health and Clinical Health (NICE) means lesser value for the NHS. If the NHS has to fund drugs that are less cost effective than at present, then with a fixed budget other treatments are squeezed out. If these displaced treatments are relatively more cost effective, overall health gain is reduced.
Value based pricing attempts to solve this by making the arguments for two kinds of premiums, one for “burden of illness,” the other for “innovation.” (The consultation paper uses the term “thresholds,” which can be translated via cost per quality adjusted life year (QALY) to drug prices that are higher by a “premium” amount.) This consultation needs to be considered alongside that on the Cancer Drugs Fund and their accompanying impact assessments, each of which include details of the option appraisals carried out by the relevant civil servants. The publication of these impact assessments is a recent development aimed at ensuring that policy is evidence based. However, in these instances, the impact assessments reveal the lack of evidence for the policies.
The consultation paper defines “burden of illness” (rather loosely) as follows:
“Under value based pricing there would be higher thresholds for diseases with higher ‘burden of illness.’ The most important factors contributing to the measurement of ‘burden of illness’ would be the severity of the condition and the level of unmet need.”
“Severity could reflect the health status without the new treatment and also if the condition leads to premature death or serious morbidity. It could be assessed in terms of the existing QALY unit of health benefit – which can be used to quantify the outstanding health loss.”
“Unmet need could reflect the degree to which there are existing treatments. A condition for which there is no effective treatment and where there is, therefore, significant unmet need, could be characterised by a high QALY loss and deemed to exhibit a high ‘burden of illness’. Conversely, conditions that are already well served with effective treatments would be scored at a lower level on this measure – even if the untreated condition was itself severe and life-threatening.”
Defined in this way, “burden of illness,” by including both severity and unmet need, offers scope to add a premium for severe conditions for which no treatment exists. This could be used to rationalise both NICE’s end of life guidelines and the Cancer Drugs Fund (indeed it states that “cancer patients … can be classed as disabled under the Disability Discrimination Act,” VBP Impact Statement, p.40).
The value based pricing impact assessment considers that such a premium could be justified if society’s valuation of benefits differed from those currently valued in QALYs. This could occur for three reasons: severity bias (public favours treating the more severe), equity bias (public considers some groups more or less equal as for instance with the equal innings argument), and an identifiability bias (priority for patients who have been identified). Its review of the literature found limited support for the first but insufficient support to operationalise it, and no support for the other two. Further, “Given its essentially arbitrary nature it is doubtful whether policy ought to acknowledge the identifiable victim effect.” The impact assessment for the Cancer Drugs Fund came to similar conclusions.
Innovation involves another new concept, the “therapeutic innovation and improvement” measure. This is “the scale of the clinical benefit provided by the medicine, beyond current best practice – known as the therapeutic innovation and improvement measure. Its units have not been determined buy may be based on incremental QALYs provided by the treatment, beyond existing best practice.” (Impact Assessment, page 18)
This measure will generate a second type of premium price that the NHS would pay for drugs offering a relatively large health gain. It is similar to that proposed by Ian Kennedy, which led to NICE’s “innovation pass” (the demise of which was announced in the Cancer Drugs Fund consultation). The difference seems to be that the proposed new premium will be formalised.
So far, so clever. But what about the third objective of value to the NHS? If the NHS must pay more through the innovation premium, who will bear the cost? The consultation paper struggles with this. Three solutions are mentioned, but only in passing: substantial modelling to ensure all three objectives are met (para 4.30), lower prices than otherwise for some drugs that are not innovative (para 7.3), and the possibility that the incentives for innovation will lead to “significant clinical breakthroughs” (para 4.25). None of these is convincing as presented, but the value based pricing impact assessment considers the arguments in detail.
The impact assessment states that the costs of value based pricing – higher prices both for products currently purchased by the NHS and for those to be purchased under the new regime – will lead to net losses for NHS patients. The direct benefit will be increased profits to companies enabled to these sell drugs to the NHS. The key issue thus becomes one of whether these raised company profits will lead in time via R&D to new products.
In brief, although the reasoning is dense, the impact assessment argues that this is highly unlikely, for one simple reason. Pharmaceutical R&D is global, and the UK is only 3.5% of the global market. Higher prices by the UK will have a small effect on companies’ profits and hence on R&D. And any benefits from new products will be shared with the rest of the world.
“For example, a 100% increase in the prices paid by the UK for medicines under patent might be expected to increase the returns to R&D by as little as ~3.5%, the fraction of global revenues resulting from the UK. This would lead to a very small increase in the number of new medicines discovered by R&D – while the UK would sustain a large cost – unlikely to be justified by the resulting benefits to its own citizens” (p 15).
What can be concluded?
Firstly, what is proposed is not value based pricing as advocated by the Office of Fair Trading or any other reputable body. Instead what is proposed are two types of premiums to award severity and innovation. Alternative titles might be “severity and innovation pricing,” “interest group pricing,” or perhaps “rhetorical pricing.”
Secondly, the proposed premiums are not new but rather aim to formalise and extend existing arrangements. The Cancer Drugs Fund and NICE’s End of Life criteria are in effect premiums for severity. NICE’s innovation pass, which resulted from the Kennedy Report provided an innovation premium. What is new is that the criteria for these premiums are to be formalised. But not yet, the impact assessment makes it clear that thinking has only just begun.
Thirdly, the arguments for awarding premiums to severity and innovation do not stand up, as is shown by their detailed reviews in the impact assessments for both the Cancer Drugs Fund and value based pricing. No conclusive evidence supports the view that the public values severity but research would be helpful on this. The argument that the innovation premium in the UK will change global pharmaceutical R&D is shown to be absurd. The argument that lower prices might be paid to less innovative drugs is raised but not developed.
Fourthly, these proposals have serious implications for NICE – which will be the subject of a separate blog.
Finally, these proposals look likely to worsen matters for the NHS and its patients. Some interest groups (cancer and selected other patient groups, the pharmaceutical industry) will gain, but the cost will be borne by those other patients who consequently cannot be treated.
James Raftery is a health economist with several decades experience of the NHS. He is Professor of Health Technology Assessment at Southampton University. A keen ‘NICE-watcher’, he has provided economic input to technical assessment reports for NICE but has never been a member of any of its committees. The opinions expressed here are his personal views. Comments welcomed.