Fairness requires fuller disclosure of model to drug companies

From The TimesMay 7, 2008

Fairness requires fuller disclosure of model
Regina (Eisai Ltd) v National Institute for Health and Clinical Excellence in the Court of Appeal
Court of Appeal

Published May 7 2008

Regina (Eisai Ltd) v National Institute for Health and Clinical Excellence

Before Lord Justice Tuckey, Lord Justice Jacob and Lord Justice Richards

Judgment May 1, 2008

Procedural fairness required the National Institute for Health and Clinical Excellence to release a fully executable version of an economic model to those consulted in the course of an appraisal process and not simply a read-only version. To do otherwise would place drug companies at a significant disadvantage in challenging the reliability of those models.

The Court of Appeal so held in a reserved judgment, allowing an appeal by Eisai Ltd against Mrs Justice Dobbs ([2007] EWHC 1941 (Admin)) who, inter alia, refused its application for judicial review of guidance issued by the National Institute for Health and Clinical Excellence (NICE).

The guidance, issued in 2006, recommended the use of Aricept for NHS patients only with a moderately severe form of Alzhemier’s disease, rather than for those with a mild to moderately severe form for whose use the drugs had previously been recommended.

In its consultation process for the purpose of appraising the clinical benefits and cost-effectiveness of Aricept, NICE made available a read-only version of an economic model, in the form of an Excel spreadsheet, which was used to assess the drug’s cost-effectiveness.

Eisai requested, but was refused, a fully executable version of the model. Eisai’s case was that the nonprovision of that version rendered the consultation process unfair and the decision to issue the guidance unlawful.

Mr David Pannick, QC and Mr Tom de la Mare for Eisai; Mr Nigel Giffin, QC and Mr Daniel Stilitz for NICE; Mr Brian Kennelly for Shire Pharmaceuticals Ltd, as first intervener; Mr Ivan Hare by written submissions for the Association of the British Pharmaceutical Industry, as second intervener.

LORD JUSTICE RICHARDS said that there had been exceptions to NICE’s policy of not releasing the fully executable versions of economic models but that no such exception had been made on this occasion.

NICE’s stance was that there had already been sufficient disclosure and they were not able to go further because of confidentiality, and the delay to the appraisal process that would be caused by releasing the fully executable version of the model.

It was not in dispute that NICE was subject to general principles of procedural fairness in relation to the appraisal process and, in particular, that it had to act fairly in the consultation exercise. Whether or not consultation was a legal exercise, once embarked upon it had to be carried out fairly: R v North East Devon Health Authority, Ex parte Coughlan ([2001] 1 QB 213).

The mere fact that information was significant did not mean that fairness necessarily required its disclosure to those consulted. It was better not to follow the approach of Mrs Justice Dobbs of asking whether Eisai was denied access to significant information. Nevertheless, the degree of significance of the undisclosed material was a highly material factor.

The authorities established that what fairness required depended on the context and the particular circumstances. It was not possible to extract from them a principle that fairness required disclosure of advice obtained from an outside adviser.

There was no express or implied duty of confidentiality restricting the use or disclosure of the model. It would be very surprising if a model commissioned and paid for by the Secretary of State for Health for the purpose of NICE’s appraisal process were subject to obligations of confidentiality preventing disclosure of the fully executable version to those consulted.

The concern raised by NICE as to the adverse practical consequences caused by the provision of the fully executable version of the model was a serious one to which weight should be given.

However, the additional time that would be added to the appraisal process had to be viewed in the context of an already lengthy process. Neither the additional time nor the additional cost to NICE should weigh heavily in the balance in deciding whether fairness required release of the fully executable version.

If fairness otherwise required it, the court should be very slow to allow administrative decisions of that kind to stand in the way of its release.

Notwithstanding NICE’s position to the contrary, procedural fairness required release of the fully executable version of the model. Those consulted would, otherwise, be placed at a significant disadvantage in challenging its reliability.

The appeal would be allowed subject to submissions on the appropriate form of relief.

Lord Justice Jacob and Lord Justice Tuckey agreed.

Solicitors: Arnold & Porter LLP; Beachcroft LLP; Ashurst LLP; Legal Director, ABPI.

Drug companies win Alzheimer’s appeal against watchdog

The Times (“Drug companies win Alzheimer’s appeal against watchdog“, May 2, 2008) reports:

Tens of thousands of Alzheimer’s sufferers and their families had their hopes raised yesterday as two drug companies won a landmark victory in the Court of Appeal.

The court ruled that the powerful body that controls the prescription of new drugs must give up its most precious secrets — how it measures the benefits that novel treatments bring.

The ruling is the first case that NICE, the National Institute for Health and Clinical Excellence, has lost in court. It means that in future it will have to be completely transparent in the way it reaches its decisions, revealing the inner workings of the computer models it uses to measure value for money.

Drugs are approved if they cost the NHS less than about £30,000 per quality-adjusted life year. That means for every £30,000 spent prescribing them, the benefit enjoyed by patients must add up to the equivalent of a single patient living an extra year of good-quality life.

NICE was adjudged to have acted unfairly in making an appraisal of the Alzheimer drug Aricept, which works by increasing levels of a brain chemical linked to memory and decision making. NICE had ruled that Aricept should not be prescribed on the NHS to patients with mild Alzheimer’s disease because the model failed to show that it provided good value.

But it refused to allow Eisai and Pfizer, who market the drug, full access to the model. The Court of Appeal yesterday ruled that refusal unlawful.

The judgment said that the two companies were disadvantaged in appealing against the guidance by NICE’s refusal to let them have a “fully executable” version of the economic model.

Had the companies had the full version, they could have tested it using a variety of assumptions and been in a better position to challenge the guidance. NICE must now make such a version available.

Lord Justice Richards, giving the ruling of the appeal judges, said NICE had supplied a spreadsheet of the economic model and had refused a request from Eisai for full details.

He allowed the appeal by Eisai/Pfizer, which will receive the full details and make new representations to NICE, which will then make a fresh appraisal of the drugs.

The ruling could influence many other appraisals made by NICE. For example, last year it issued guidance over drugs for osteoporosis that similarly relied on an economic model.

When the National Osteoporosis Society appealed against the guidance, it complained that it had never had access to the economic model, despite several requests.

“It always seemed to us that this public policy should have been subject to proper scrutiny,” said Nick Rijke, of the NOS. “It is a pity it took a court case to establish that.”

Economic models of this sort can be very sensitive to the precise details that are entered into them. The drug companies will want change the starting points and the assumptions built into the models to see if that produces a different answer.

Potentially, it opens up a large and controversial area of public policy to greater scrutiny. NICE has been criticised widely for its propensity to reject new drugs.

Companies disappointed by its rulings will in future be able to judge whether, for example, treating different groups of patients with a particular medicine might result in it being found more cost-effective, the Association of the British Pharmaceutical Industry said yesterday.

“This judgment provides further momentum behind the drive to make NICE processes more transparent,” said Richard Barker, director-general of the association.

NICE could appeal to the House of Lords and seek a reversal of the ruling.

Andrew Dillon, its chief executive, said: “We will be considering very carefully the findings and the implications for the time it takes us to provide advice to patients and the NHS on the use of new treatments. The ruling will increase the complexity of our drug appraisals in some cases and they may take longer as a result.” The ruling will not make Aricept available to new patients. It will simply enable Eisai and Pfizer to search for any flaws in NICE’s reasoning.

The drug acts against a key process of the disease. In Alzheimer’s, the damage is caused by the loss of brain cells that produce a transmitter, acetylcholine, that carries signals from cell to cell. When it has finished transmitting its messages, it is broken up by an enzyme, acetylcholinesterase. Aricept inhibits the action of this enzyme, thereby slowing progression of the disease.

Nick Burgin, managing director of Eisai, said: “We believe that this decision represents a victory for common sense. As soon as we have reviewed their cost-effectiveness calculations we will submit any new findings to NICE. We hope that this action will ultimately restore access to anti-dementia medicines for those patients at the mild stages of Alzheimer’s disease.”

John Young, managing director of Pfizer, said: “Contrary to NICE’s position that they follow a fully fair and transparent process, the Court of Appeal found that this is not the case.”

Neil Hunt, chief executive of the Alzheimer’s Society, said: “Today’s decision is a damming indictment of the fundamentally flawed process used by NICE to deny people with Alzheimer’s disease access to drug treatments.”

Battles to prescribe for chronic illnesses

— Aricept was the first drug to be licensed in Britain specifically for the treatment of Alzheimer’s disease. It was introduced in 1997

— People with Alzheimer’s have a shortage of the brain chemical acetylocholine, which nerve cells use to communicate with each other. Aricept prevents enzymes from breaking down acetylocholine

— The Alzheimer’s Society recommends Aricept for people with mild to moderate Alzheimer’s. It cautions that improvements in memory loss or delay may be temporary

— The National Institute for Health and Clinical Excellence (NICE) is an independent organisation responsible for providing guidance on good health and the prevention and treatment of ill health. NICE committees include healthcare professionals, patients, carers, academics and technical experts

— NICE governs whether a treatment will be made available on the NHS by weighing its cost against the benefit to patients

— For chronic conditions such as Alzheimer’s, the NHS advises costing analysts to consider “cumulative impact”. Guidelines say that “patients starting therapy on diagnosis of a chronic condition then remaining on treatment for the rest of their life could lead to a build-up of costs”

— In 2000 Ritalin and Equasym won approval from NICE for the treatment of attention deficit hyperactivity disorder in children over 6, despite fears of side effects and addiction

— The breast-cancer drug Herceptin was originally licensed only for treatment of advanced cancers. In 2006 it received approval for wider use after a high-profile campaign and the intervention of the Health Secretary, Patricia Hewitt

Sources: NHS; Alzheimer’s Society; Alzheimer’s Scotland; Times archives