Methods to estimate the NICE cost-effectiveness threshold

The decision to approve a health technology for use in the NHS requires an assessment of whether the heath that is expected to be gained from its use exceeds the health that is expected to be forgone elsewhere as other NHS activities are curtailed to accommodate the additional costs. NICE uses explicit methods to estimate the health expected to be gained from use of a technology and the expected additional costs to the NHS. However, it?s assessment of what might be forgone, the cost effectiveness threshold, is less well founded. A body like NICE cannot and does not necessarily need to know what specific services and treatments will be displaced in particular localities or who will actually forgo health. What is required, however, is an assessment of the health that is likely to be forgone on average across the NHS. The aim of this research is to develop methods of analysis that can be applied to data the NHS already collects (or could collected in the near future) to provide an estimate of the threshold which would be suitable for the type of decisions NICE makes. The research will build on work that has already be done, which examines data already collected about local NHS spending in 23 broad programmes of care. This work looks at the relationship between local spending on particular types of care and health outcomes. Additional work is needed to analyse all these programmes together and establish how particular types of spending changes with overall budget and how overall spending is related to overall outcomes. This work may allow changes in the threshold over time to be identified as well as the effects of decisions which could have a large impact on the NHS budget. Measures of improvements in health outcome need to capture increases in the length and quality of life. Additional work is needed to identify and use other sources of evidence which are published or routinely collected in the NHS that can be used to estimate the quality of life gains within the different programmes of care. Also, additional work is needed to better understand which services and treatments are added or cut back when spending changes. Therefore, we will work with local NHS organisations to identify and use other routine data and design new data collection which will give a more detailed picture of changes within key programmes of care.

Cost-effectiveness analysis used by NICE is essentially an assessment of whether the health expected to be gained from the use of a new medical technology exceeds the health likely to be forgone as other NHS activities are displaced to accommodate the additional costs of the new technology. The cost-effectiveness threshold represents an estimate of the health forgone as services are displaced and has implications for patients, local decision makers, Government and manufacturers. Currently the threshold used by NICE has little empirical basis.

The aim of this research is to develop and to demonstrate methods for threshold estimation which make best use of routinely available NHS data, allowing scrutiny by a range of stakeholders, improving accountability and predictability. Methods should also capture the effects on length and quality of life, estimate changes in the threshold over time and indicate the impact of ?non-marginal? changes. This research will focus on complementary methods which can make best use of those data that are already available, where there are already plans to make data available or where additional data could feasibly be made available at reasonable cost.

The research plans fall into the following 4 complementary areas of activity, all of which will be evaluated at a user impact workshop. Following a comprehensive literature review, econometric methods will be developed and applied to NHS programme budgeting data which allows examination of the relationship between local spending and health outcomes. Previous work in this area will be developed in a number of ways: i) simultaneous estimation across programmes will provide an overall threshold for the NHS; ii) examination of the relationship with local budgets will inform the impact of non marginal changes; and iii) the possibility of constructing a panel will allow an assessment of the feasibility of periodically re-estimating the threshold. Health outcomes need to be expressed in terms of quality as well as length of life. Therefore, the econometric analysis will be complemented with an analysis of quality of life within and between programmes of care using various sources of evidence including data currently routinely available, evidence likely to be come available in the future as well as other published evidence. Finally we will identify and evaluate other routinely available evidence of local investment and disinvestment to complement the more aggregate level analysis. We will also design and pilot the collection of additional data which are considered potentially the most useful.