The manufacturer of an Alzheimer's drug has made a formal complaint to the NHS ombudsman accusing the National Institute for Health and Clinical Excellence of secrecy for refusing to make public the mathematical model it used in the decision to restrict use of the drug.

The manufacturer of an Alzheimer's drug has made a formal complaint to the NHS ombudsman accusing the National Institute for Health and Clinical Excellence of secrecy for refusing to make public the mathematical model it used in the decision to restrict use of the drug.

In a controversial decision in January, NICE reversed a previous ruling and recommended that Aricept only be prescribed to patients with moderate Alzheimer's. It said it was not cost-effective in patients with a mild form of the disease.

Manufacturer Eisai, along with patients group the Alzheimer's Society and the Royal College of Psychiatrists, have made a formal appeal against the ruling. The appeals were heard over Thursday and Friday of last week and a decision is expected soon.

In order to mount a comprehensive appeal, Eisai says it has repeatedly asked NICE for full details of the mathematical model it used to come to its decision on cost and effectiveness, going as far as making a request under the Freedom of Information Act.

But 18 months after the initial request, and with the information commissioner still considering NICE's reason for non-disclosure, Eisai last week complained to the NHS ombudsman, saying the commissioner failed to issue a decision in a timely fashion.

Eisai managing director Dr Paul Hooper said: 'It's a disgrace that NICE are using economic models which they refuse to make available for scrutiny by others.'

He said NICE's refusal to disclose the model goes against its own guidelines 'and the principles of natural justice'. 'We hoped that the information commissioner would understand the importance of this request and make a timely judgement, but sadly this has not happened.'

And Alzheimer's Society research director Clive Ballard asked: 'How can NICE facilitate open and honest consultation when they conceal the detail of their decision in a model they won't release? Without a workable model we don't know to what degree they have taken into account our concerns, or if they have done so correctly.

'This is especially of concern when independent review bodies around the world and the majority of the stakeholders are all finding that the drugs are cost effective.'

Eisai said it had been told by NICE that it could not disclose the model as it had been provided by a third party - Southampton University's health technology appraisals centre - and that disclosure could constitute a breach of confidence.

In a statement, NICE chief executive Andrew Dillon said: 'All issues relating to this appraisal are now subject to the NICE appeals process. The institute is not in a position to make any further comments until the outcome of the appeal has been made public.'

A spokesperson for the information commissioner said: 'This case is complex and detailed. We are examining a large amount of information which has been submitted by both NICE and the complainant.

'The case involves certain legal and pharmaceutical issues which we must examine very carefully to ensure the right decision is reached.

'We also have a number of cases outstanding, meaning it is taking us longer to deal with some complaints than we would like.'