Investigating NICE's Cost Effectiveness Threshold

The purpose of this research is to determine what evidence appears to most influence the decisions of the UK’s National Institute for Health and Care Excellence (NICE) about medical technologies. All NICE decisions from 1999 through 2010 are being modelled to test hypotheses about the impact of a range of types of evidence, including cost effectiveness.

Since it was established in 1999, NICE has become increasingly explicit about how it uses evidence on cost-effectiveness in its decisions about health care technologies and, more recently, about its willingness to make tradeoffs between cost-effectiveness and other considerations. For example, NICE asserts that its decisions take account of a range of factors, including improvements in the health of patients at the end of life and disease severity. A number of testable hypotheses can be derived from the increasing transparency about these ‘social values’, the evolution of other aspects of NICE’s decision-making processes, and changes in contextual factors – such as the size of the NHS budget that NICE intends its cost-effectiveness threshold to reflect.

OHE is collaborating on this project with the University of Oxford and University of York. Data have been made available to the study team by HTAinSite, a user-friendly database of NICE decisions.

Investigators are Professor Nancy Devlin, OHE; Helen Dakin, University of Oxford; Professor Nigel Rice, University of York; Professor David Parkin, King's College, London; and Phill O’Neill, OHE.

While conclusions remain tentative, interim results suggest that cost effectiveness is by far the most dominant consideration in decision making, with almost no other variables having a statistically significant effect on decisions. A notable exception is whether or not the technology concerns the treatment of cancer; such technologies have a significantly higher probability of being recommended, all other things held constant.

The cost effectiveness threshold suggested by our models, at £40,000 per QALY gained, is higher than NICE’s stated ‘threshold range’ of £20-30,000, a finding also reported earlier by Devlin and Parkin (2004). There is no evidence that the threshold has shifted over time.

The project will be completed during 2011 and results reported in a paper to be submitted to OHE Research Papers and to a peer reviewed journal. Initial results, based on a preliminary data set, were reported in 2010 at the European Health Economics Conference and the ISPOR Conference and in early 2011 at the University of York’s Health Economics Study Group (HESG).

Devlin, N. and Parkin, D. 2004, Does NICE have a cost effectiveness threshold and what other factors influence its decisions? A binary choice analysis. Health Economics. 13(5), pp.437-452.