Van Gool & Bridges
who quit smoking. A common ‘natural unit’ of output is chosen to allow a single
dimensional maximisation. In this paper the output measure is CHD deaths prevented
and the objective will be to highlight those interventions which maximise the number
deaths prevented*.
In this project, the evidence is derived from the international literature. One of the
pitfalls of a CEA that is reliant on this type of evidence is that there are certain gaps
which prevent us from transforming the effectiveness evidence into a single outcome
measure. Also, there may insufficient information to formulate the cost function.
Section 5 will discuss this issue further.
Coronary Heart Disease (CHD)
In 1995, the NSW Health Department published a document entitled ‘Coronary Heart
Disease: NSW Goals and Targets and Strategies for Health Gain’. This followed an
earlier paper by the Federal Government that identified cardiovascular disease as one of
four national health priorities2. In 1997, a working party was established in SWSAHS
to develop a Health Improvement Plan for the management of CHD in the area.
CHD, also referred to as ischaemic heart disease, is the most common form of
cardiovascular disease (CVD). CHD occurs when one or more of the coronary arteries
become blocked, or partially blocked, by solid deposits or scar tissue. The international
classification of diseases (ICD-9) codes for CHD are ICD 410-414.
1 Putting this in mathematical language, consider that a possible combination of relevant strategies (health care
procedures) is given by the vector x = {x∣, x2,..., xn} ∈ X, where X represents all such possible strategy combinations.
Also, the single outcome measure is given by the scalar Y. Epidemiological evidence is used to transform the
strategy vector into an outcome measure. So we can consider a correspondence which relates interventions to the
outcome.
E : X —> Y is the evidence correspondence, E(x).
C : X —> C is the cost correspondence mapping from the strategy space into dollars.
It is given by C(x) = p∣x1 + p2x2 +...+ pnxn.
Now we can describe the CEA method as:
max {r=^ω∣ c(χ)≤c) ɪɪi
xεX
OΓ
min {C(x) ≤ CI Y(x) ≥ Y} [2]
xεX
As health economists we normally consider a binding budget constraint, rather than a binding outcome target. Thus
we will be using the first mathematical method as a theoretical construct for our project.
2 ‘Better Health for All Australians: National Goals, Targets and Strategies for Better Health Outcomes into the Next
Century’ (1994).
Chere Project Report 11- November 1999