In message <[log in to unmask]>, Ruth Livingstone
<[log in to unmask]> writes
>Buried in the bulletin of
> Effective Health Care, Feb 98, vol 4 no 1, ISSN: 0965-0288
>from the NHS Centre for Reviews and Dissemination,
>University of York, page 7
>
>"Further important trial results are awaited which appear
>likely to extend the range of indications for use of statins.
>The next trial to report will be the Air Force/Texas Coronary
>Atherosclerosis prevention Study of lovastatin in 6605 people
>(15% women) with no evidence of coronary heart disease and
> with average blood cholesterol levels. This trial was stopped
>early after finding a 36% reduction in the combined fatal and
>non-fatal CHS endpoint."
>
>So, this trial was stopped because it was deemed unethical
>to continue NOT treating people with statins.
>Implications are - statins for everyone !!!!
>
Yup - well that fits in with other statin trials which show similar
relative risk reductions. The point is, what is the absolute risk
reduction? If a population has a prevalence of CHD events of say 20%,
then a 36% relative risk reduction means there will be an absolute risk
reduction of (quick and inaccurate mental arithmetic) 7%. The number
needed to treat (NNT) is 1/ARR=100/7=14. Quite useful. But if the
background risk is 1% (say in a young, fit population which doesn't
smoke) then the absolute risk reduction will be 0.36%, which is peanuts.
The number needed to treat here is 100/0.36=c300 - pretty useless, or at
least expensive. I don't know the cost of lovastatin, but say it's
UKP250 a year, and the risk reduction is expressed as events in a year
(I'm assuming these figures), then the cost per event treated is around
ukp250x300= UKP75,000 for your low risk population, but only
UKP250x14=UKP3,500.
It depends how rich your country is as to what level of risk you can
afford to treat.
Cheers
Toby
--
Toby Lipman 7, Collingwood Terrace, Jesmond, Newcastle upon Tyne. Tel
0191-2811060 (home), 0191-2437000 (surgery)
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|