Education And Debate

Personal paper: how to get the best health outcome for a given amount of money

BMJ 1997; 315 doi: (Published 05 July 1997) Cite this as: BMJ 1997;315:47
  1. Matthew Sutton, research fellowa (
  1. a Centre for Health Economics, University of York, York YO1 5DD
  • Accepted 16 January 1997


Increasingly, healthcare programmes are evaluated in terms of both their costs and outcomes. However, existing evaluation designs, in which equal numbers of participants are compared, remain a hangover from the time when outcomes were the sole concern. Outcome maximisation studies, in which costs are equalised, are an alternative way of comparing the costs and consequences of health interventions. They may be feasible in some instances and are attractive because they highlight two of the main messages of health economics.

Firstly, economists emphasise that their concern for the costs of interventions is due to the implications of opportunity costs—that is, resources used for one intervention should be compared with the potential benefits achievable by the next best option that is forgone because the resources for health care are finite.1 2 3 In an outcome maximisation design the opportunity costs of high cost interventions would be highlighted more clearly in the study results.

Secondly, the marginal costs of implementing programmes do not necessarily equal average costs, in which marginal costs are the change in costs resulting from a small change in the level of provision.4 By evaluating alternative technologies at comparable levels of expenditure, outcome maximisation studies would take account of the fact that programme costs often do not rise proportionally with the number of patients treated.

Summary points

Health care evaluations used to focus solely on outcomes, but increasingly they have compared the costs of competing interventions

Popular study designs in which equal numbers of people are subjected to each regimen at different levels of total cost do not reflect this innovation

An outcome maximisation design, in which the costs allocated to each programme are equal, might be appropriate

This design has two main advantages: (a) the opportunity costs (or next best use) of more resource intensive treatments are readily expressed …

View Full Text

Sign in

Log in through your institution