We reported in our August 2007 update on the first court challenge to the guidance by NICE (established by the National Institute for Clinical Excellence (Establishment and Constitution) Order 1999), over what drugs should be available for prescription on the National Health Service (NHS), and on how the High Court had largely rejected the challenge to NICE’s recommendation that three Alzheimer’s drugs should not be prescribed on the NHS for mild cases of Alzheimer’s disease, but should be prescribed for moderately severe cases of the disease only. On 1 May 2008 the English Court of Appeal reversed a significant aspect of the High Court judgment, relating to the procedural fairness of NICE’s assessment system, and the nature of the access to be given to consultees of the economic model used to assess the cost effectiveness of the drugs. Although it was not asked to reverse the determination originally made by NICE, the effect of the judgment is to mandate the release to the consultees of information that would enable them to make further representations to NICE with a view to its making a further determination about the circumstances in which the drugs in issue should be prescribed.

The Court of Appeal held that NICE had acted unfairly by only making available to consultees (namely the companies whose drugs were the subject of the assessment), a read-only version of the economic model, in the form of an Excel spreadsheet, rather than the fully executable model that had been requested and that would have allowed changes to be made to the inputs or assumptions on which the model was based in order to test its robustness or reliability. NICE had sought to justify its refusal to provide the fully executable model on two grounds – the first that it had been provided to it by a third party on terms of confidentiality which precluded its wider dissemination, and the second that its use would result in extra work and delay. The Court of Appeal rejected both reasons. The economic model had been commissioned for the purposes of NICE’s appraisal process and paid for, the confidentiality provisions in the commissioning agreement were in general terms only and did not restrict the use or disclosure of the model, and in any case any disclosure of the fully executable model by NICE to consultees could be made subject to standard undertakings of confidentiality. As for the second objection, a possible extra 2 or 3 months in the context of a 2.5 year appraisal process, as had been the case here, did not weigh heavily in the balance in deciding whether procedural fairness required release of the fully executable version of the model to consultees.