Britain’s National Institute for Clinical Excellence (NICE) has drawn controversy for using cost-effectiveness information to help determine coverage for drugs and medical devices. Until recently, the United Kingdom was the only country that took cost into account, but other countries, including Austria, Brazil, Columbia, and Thailand, now pay close attention to NICE as they strive to control health care costs. We asked Anthony J. Culyer, PhD, chair of the Research and Development Committee of NICE, to discuss the agency’s strengths and address criticisms.

What is the role of technology assessment in the U.K. health care system?

Broadly speaking, the U.K. has two parallel but somewhat intertwined health care systems. One is financed through private insurance premiums under which insurees can receive reimbursed care either in private institutions or in public ones. The other, called the National Health Service (NHS), is financed through tax-premiums; care is received either in public institutions or private ones contracted to the NHS.

Both systems generally claim to offer only effective health care – interventions for which there is credible evidence that they work. While evidence is not itself a sufficient basis for decisions, there is considerable commitment to the idea of evidence-informed decision – making, and a good deal of evidence that is available comes from scientific studies. Both sectors take NICE’s advice very seriously.

The terrible legacy of thalidomide still casts its shadow. Before the release of thalidomide in the late 1950s, adequate tests had not been performed to assess its safety, with catastrophic results for the deformed children of women who had taken thalidomide during their pregnancies in order to combat morning sickness.

How does NICE fit into technology assessment?

Being by far the larger of the two parallel sectors, the NHS takes the lead in seeking to address the question of what works. NICE is the principal agency within the NHS for giving guidance about this. NICE is not a research organization. It is, however, a major consumer of research outputs.

NICE’s technology assessment procedures are clearly laid out, are publicly accessible, and were established and have been subsequently revised only after extensive discussion and consultation with all stakeholders, including industry. The general aim is to maximize the health benefits from the care the NHS provides – whether preventive, diagnostic, curative or palliative, and whether in primary, secondary or tertiary care.

The cost-effectiveness criterion NICE uses is that only technologies expected to generate a quality-adusted year of life at a cost less than a threshold of £20,000 to 30,000 (about $26,000 to $36,000) will be recommended. Technologies that do not prolong life but that enhance its quality (as, for example, in palliative care) are judged by the same effectiveness criterion. Costlier interventions need exceptional grounds for being recommended.

What kind of authority does NICE have?

NICE cannot ban anything. It issues guidance, in the form of both clinical and public health guidelines, and on the use of technologies like drugs and medical devices. The guidance specifies the technologies in question, their dosages and frequency of use, the stages of a disease at which their use is most appropriate, counter-indications, and the patient groups for which they are likely to be effective.

Local health care purchasers and providers must make any technology recommended by NICE available when it is required by a local physician. In this sense, NICE enables rather than commands – only physicians have powers to command.

To date NICE has focused on new technologies but, in response to intense parliamentary pressure, it is currently exploring the possibility of issuing “disinvestment” guidance, that is, recommendations that particular practices should be stopped or used more selectively.

What is the role of QALYs in decision-making?

NICE recommends the use of a version of the Quality-Adjusted Life-Year as its principal outcome measure (EQ-5D), partly to enable NICE advisory committees to make consistent comparisons between the many possible procedures that could be included in the “benefits basket” and partly to encourage researchers to use that outcome wherever appropriate. The QALY is not itself, a criterion. It is only the denominator of the incremental cost-effectiveness ratio.

The usual practice is to compare the QALY difference between two technologies with their cost difference, and if the cost difference divided by the QALY difference is less than the threshold ratio mentioned earlier, the most cost-effective technology is recommended. The higher a specific incremental cost-per-QALY sits in the threshold range, and certainly when it exceeds it, the greater the power of other factors needs to be to result in positive guidance.

In many cases, however, the QALY is not available or is an inadequate measure in a specific context. The presence of patients and caregivers on advisory committees is one way that the appropriateness of the QALY is assessed in each situation and for NICE to apply patient-informed other judgements in reaching its decisions.

The cost-per-QALY is at best a guide to the probable effectiveness of a procedure. It is an aid to judgment, not a determinant, always requiring the consideration of contextual factors.

Is NICE a threat to physicians’ freedom to practice? Is it a threat to industry?

NICE’s guidance to professionals is just that: guidance.

The general expectation is that most professionals will follow the finest advice and guidance that can be mustered to support their work. This seems the right approach to me — the best way to encourage best practice is to provide the best information.

With passage of time – and the likely creation of NICE look-alikes around the world – an important new determinant of research patterns in industry seems likely to emerge. For the first time, strong indications will exist about the types of research-based products entire systems are willing to pay for and which will therefore generate returns for innovators. If this generates, as it should, incentives to invent new products that are cost-effective, then that will be one significant strand in the universal striving for both better health and cost containment.

Are recent attacks on some NICE decisions evidence of some deep unrest?

This is hard to answer. Whereas industrial sponsors and patient organizations have specific concerns, NICE has to take a broader view, representing the interests of all patients.

It is inevitable that on occasion there will be clashes. It is also inevitable that the value judgments that determine, among other things, the value to be placed on small extensions of not very satisfactory life, are contestable. NICE needs to be responsive to the insured public’s views without becoming a slave to every energetic protest from a sectional interest.

There is plainly a political balancing act to be performed. It is worth remembering that a NICE failure to recommend a technology is not a ban – and local decision-makers and clinicians may choose differently. There remains, however, considerable public resistance to the significant differences in what is available according to where you live in what is regarded as a national service. The balance to be struck between local autonomy and national standards is a further inevitable point of tension.

Are any important reforms in the NICE program likely?

There is continuing pressure for NICE to speed up its decision process – coupled awkwardly with a conflicting imperative to make decisions in a consultative fashion. There is also pressure to extend NICE’s work into disinvestment guidance.

There are issues about how best to make decisions when the evidence is absent, weak, contested, or biased.

There are many issues surrounding NICE’s application of technology assessment to public health measures – especially those involving non-health care agencies such as schools and municipal authorities, and measures whose principal aim is greater equity in population health. NICE is making progress with the application of technology assessment in surgical procedures, but this has, to date, fallen short of full-blown cost-effectiveness studies, not least because of the frequently poor clinical evidence base.

Another concern is how best to incorporate matters of fairness into NICE processes. Should people with rare diseases get special consideration? What about children? Or those near death? What about those with a lifetime of deprivation and disease for whom the smallest improvement ought perhaps to count for more than similar gains received by more fortunate folks?

A further ongoing concern is the effectiveness of existing knowledge translation and exchange methods. While the provision of excellent information to professionals is of critical importance, nothing of value happens if those for whom it is developed have neither the time nor the inclination to take notice of it. It’s NICE work if you can get it! NICE is necessary – but scarcely sufficient.

A great feature of having a NICE is that these issues get posed in a public forum and can be addressed in the interests of the whole community and with the whole community’s participation.

Anthony J. Culyer, PhD, is the Ontario Research Chair in Health Policy & System Design at the University of Toronto and professor of economics at the University of York, England. Ajc1​7​@​york.​ac.​uk; 011‑1904-433762.