The running battle over which National Health Service patients with dementia should have access to the dementia drug donepezil (Aricept) is to progress to the courts. Last year, guidance from the National Institute for Health and Clinical Excellence (NICE) restricted use of the drug to patients with moderate and severe Alzheimer's disease, thereby denying its use for 60% of patients with Alzheimer's disease who have mild dementia.1

NICE's decision was based on modelling Aricept's clinical and cost effectiveness through a contract with Southampton University's Health Technology Centre. NICE makes its own internal work accessible to the drug industry so that its processes are open to critical appraisal. Furthermore, it requires industry to supply all its evidence in an “executable” form, so that differing assumptions can be modelled. However, NICE's appraisal guide clearly states that the contracted work of external academic assessment groups, such as the Southampton University's Health Technology Centre, will be available just in a “read only” form in which different modelling assumptions cannot be re-run.2 NICE argues that this is essential to protect the intellectual property rights of assessment groups. This lack of transparency has never been challenged before, and …