• health economics

Health economics has developed on a truly massive scale in the past 20 years. As the pressure on health budgets has grown, so too has the “priesthood” seen by politicians and funding agencies as well equipped to answer key questions about what health care should and should not be provided. In practice, the tools and answers of health economists are severely limited, by methodology, politics, and ethics. But while not answering key questions, economic analysis can shed light on the questions and give those making the decisions a much clearer idea of the choices they face.

Analysis to help choice of treatment, for governments and insurers, usually falls into one of three categories: cost minimisation; cost effectiveness; and cost utility. However, definitions vary in practical use and many authors have now merged some of these categories into cost effectiveness or cost benefit studies.

A cost minimisation study is the simplest, focusing on comparing treatments with an (assumed or demonstrated) identical outcome. For example, an early UK study compared bottled oxygen with use of an oxygen concentrator. Both provide oxygen with similar (assumed) effects so the economic choice is the therapy that minimises costs.

Costs in most theoretical discussions in economics include costs to all, specifically the value of what is given up to achieve something. There is much theoretical discussion about the use of average or marginal cost, the latter covering the actual change in costs rather than the average across a range of activity. Some things can be free, for example, if there is spare capacity in a system (a spare bed or a vacant and staffed theatre session), and marginal costing tries to take account of this. But marginal costs are often very context specific and there is widespread use of average costs in the practical literature. Also, while costs to citizens and families are theoretically important, practical studies often concentrate on the costs to the health care system alone.

Where outcomes have been measured in some standard units, it is possible to develop a cost effectiveness study. This looks at the cost of achieving each unit of outcome (for example, patient-year free of symptoms, case cured) using each treatment. Here the emphasis is on the incremental cost of achieving an extra unit of outcome by switching from one treatment to another. But cost effectiveness studies of this kind beg the question of whether treatment should be provided. That is, they do not ask the question “Is the health gain worth the cost?”, but only “How much more outcome (in standard units) can we get for our money?”. For example, a case cured may offer little gain relative to a case uncured if a disease is self-limiting and causes little long term health damage.

When outcomes are measured in more detail, typically by assessing the length and quality of life gains achieved by alternative treatments, we have a cost utility study. Utility was originally a definition in economics for the benefits of consumption (of goods and services, not lung disease). Terms such as welfare or wellbeing can be substituted for utility without too much loss of meaning and it is clearly only a short step from welfare to health.

In cost utility studies, changes in health are ideally measured using specialised instruments to assess the changes in individual capacity and the value attached to each element. A range of other techniques have been used to try to elicit values from what usually remain hypothetical choices. It is clearly very difficult to decide whether you would rather have ten years with moderate pain or eight years with no pain until you have had experience of the pain for some time. But equally, we are often assessing potential treatments in advance so the views of what the healthy would spend to avoid something may be more relevant than the views of sufferers, after the event.

In this issue, Ungar and Santos¹ have examined the literature reporting health economics studies in paediatrics. They find a much smaller literature than for adults, reflecting, in their view, the greater complexity of research on children. There is no doubt that children raise particular problems, but these may be of degree rather than distinction. Of course, children may struggle to value the alternative outcomes from the treatments available explicitly. But this may be less important, in policy terms, if decisions on treatment are made by their parents or other adults. Other factors limiting economic research on children include the reduced level of research by pharmaceutical companies on the smaller market for children’s therapy and perhaps the greater concentration on research in chronic diseases, where expenditure is high and some outcomes rather marginal. There may also be a tendency for children’s health care to gain priority due to the age of those suffering, rather than explicit evaluation of outcomes, so that there is less technical scrutiny of paediatrics (though this explanation is speculative).

Interested readers should see Drummond and colleagues² (also available in other languages), and may wish to consider the Distance Learning Course in Health Economics at the University of York (http://www.york.ac.uk/res/herc/distance.htm).