Intended for healthcare professionals

Rapid response to:

Primary Care

Coronary heart disease prevention: insights from modelling incremental cost effectiveness

BMJ 2003; 327 doi: (Published 27 November 2003) Cite this as: BMJ 2003;327:1264

Rapid Response:

Setting Risk Thresholds For Coronary Event Prevention

EDITOR - Tom Marshall advocates a stepped approach to therapy for
preventing coronary events, introducing more expensive therapies at
progressively higher levels of risk (1). This approach contrasts with
those who argue that these treatments should all be used very widely, or
even universally (2, 3).

Extending Marshall’s incremental cost effectiveness analysis to a
greater range of risk values shows some interesting effects. The
relationship between the 5 year coronary event risk (x) and the
incremental cost per event avoided (y) is given by: where (a)
is the incremental cost per event reported for 10% five year risk in Table
2 of Marshall’s analysis (in the case of aspirin this formula is a
simplification which slightly underestimates the true cost, as it
disregards adverse events). The curves for the five drugs are shown in
the figure.

The graph shows that the incremental cost is extremely high for small
values of risk. The relationship illustrates two points. Firstly, those
who argue for the universal use of multiple therapies to prevent coronary
events must be willing to countenance enormous costs for each event
avoided in low risk individuals. Secondly, the cost for the more
expensive agents decreases only slowly.

The full curve shows that the implications are more extreme than
Marshall himself suggests. If avoiding an event is valued by society at
some pre-specified cost, then this can be used to find the degree of
absolute coronary risk at which it is acceptable to use a given drug.

For example, Marshall suggests using statins at 30% or greater five
year risk of a coronary event, with an incremental cost of £40,800. If
£40,800 is the value placed upon avoiding a coronary event (indicated by
the horizontal line on the graph), then this implies that it is reasonable
to use aspirin at a level of less than one percent risk, bendrofluazide at
a level of 2.9% and enalapril at 8.3%. If £40,800 is an acceptable cost
to avoid a coronary event, then it could arguably be reasonable to use
aspirin universally, while clopidogrel would never be used.

As more agents become available for preventing coronary events, there
will be continuing controversy about their cost-effectiveness. It is time
the debate regarding cardiovascular risk moves forward from simple
assumptions using relative benefits to cost effectiveness studies, such as
Marshall’s, that include absolute risk, patient utilities and the cost of
treatment (4).


(1) Marshall T. Coronary heart disease prevention: insights from
modeling incremental cost effectiveness BMJ 2003;327:1264-8

(2) Law MR, Wald NJ. Risk factor thresholds: their existence under
scrutiny. BMJ 2002;324:1570-6

(3) Wald NJ, Law MR. A strategy to reduce cardiovascular disease by
more than 80% BMJ 2003 326:1419-23

(4) Montgomery AA, Fahey T, Ben-Shlomo Y, Harding J. The influence of
absolute cardiovascular risk, patient utilities, and costs on the decision
to treat hyptertension: a Markov decision analysis J Hypertension 2003


Competing interests:
None declared

Competing interests: No competing interests

15 December 2003
Tom Love
Research Fellow, Health Informatics Centre
Tom Fahey
University of Dundee DD2 4BY