Acknowledgments  
The authors would like to acknowledge funding received from the Medical Research Council/National
Institute for Health Research through its Methodology Research Programme (Award G0901498/1).   

We acknowledge the contribution of the following individuals who participated in various aspects of this
research: Ling-Hsiang Chuang, Craig Currie, Jamie Garside, Simon Gilbody, Charlotte Haylock, Sarah
Jenkins-Jones, Tim Kendall David Parkin and John Parkinson.   

We are also grateful to those colleagues who attended our research workshop in May 2010 and offered
comments on our preliminary methods.  We would especially like to thank Peter Littlejohns for his
support in advocating this topic to the Methodology Research Programme.   

This is an internal report which will be formally reported in Health Technology Assessment as a peer-reviewed
journal.

All views expressed here, and any errors, are entirely the responsibility of the authors.  


Scientific Summary   vi
  
Chapter 1: Introduction   1  

1.1   Policy context  1   
1.2   Estimating the cost-effectiveness threshold  1
1.3   Aims and objectives  1
1.4   Report structure  2


Chapter 2: Policy Context and Conceptual Framework  3

2.1   Introduction  3
2.2   What should the NICE threshold represent?  3
  2.2.1  The threshold as a measure of opportunity cost  3
  2.2.2  The threshold as the consumption value of health  5
2.3  Estimating the threshold  5
  2.3.1  NICE’s threshold range  5
  2.3.2  The basis for empirical work  6
  2.3.3  Studying displacement locally  7
  2.3.4  What evidence is needed?  7
2.4  An introduction to study methods  8
  2.4.1  Past work  8
  2.4.2  Further econometric analysis  9
  2.4.3  Moving from life-years to quality-adjusted life-years gained  9
2.5  Conclusions   10


Chapter 3: The link between NHS spending, mortality and the   11
    cost of a life year

3.1   Introduction  11
3.2  Previous studies  12
3.3  Modelling framework  13
3.4  Data      15
  3.4.1    Programme budgeting in England  15
  3.4.2    Health outcome data  18
  3.4.3    Other variables  21
3.5  Approach to model estimation  22
  3.5.1    IV estimation  23