Actually, that 66,600 number is not ‘lives saved’. It is, as Dr. Rich correctly labeled it, preventing “major cardiovascular events over a 2 year period” (defined by the JUPITOR investigators as “myocardial infarction, stroke, arterial revascularization, hospitalization for unstable angina, or death from cardiovascular causes”).

Table 3 of the JUPITOR trial publication breaks down specific event rates by type. This list includes some composites, but also gives data for MI, CVA and death separately.

I don’t have the statistical chops to actually do a cost-benefit analysis, but it should be kept in mind that death costs ‘the system’ essentially nothing (unless we do a bunch of procedures, etc. that prove to be futile, which still is less expensive in the long run if the patient dies). On the other hand, the survival of an MI or CVA is quite an expensive proposition.

I only mention this to point out that cost-benefit analyses, even when performed with perfect statistical methods, can be quite misleading. We try to get around this with the QALY (quality adjusted life year). The threshold used to determine whether an intervention is worthwhile is often the cost of 1 year’s worth of hemodialysis, which is at best somewhat arbitrary.

If I had my way, instead of doing all of these statistical back flips trying to determine how much an average 1 or 2 years of some else’s live is going to cost ‘the system’, we would be trying to translate these numbers into the probability that a given intervention will help the patient sitting in front of his/her doctor. Of course, that would imply/necessitate individual decision making based on an individual’s cost/benefit analysis, but since the ‘cost’ to the individual is distributed across at least some of the population by health ‘insurance’, the system falls apart and we’re back to where we started.

I agree that will be very interesting. The more obvious the implicit QALY threshold becomes, the more likely we will actually be able to discuss open rationing.

Rich

Thanks, Dr. Gaulte, but as you know I am very humble, and so I will settle for just plain brilliant.

Rich

My public choice theory generated cynicism aside,your essay regarding Jupiter and the various species of criticism of it is at least brilliant.

I agree completely – in an overt rationing environment, a full cost-benefit analysis would be performed, and the decision would be made based on hard evidence and pre-announced cost or QALY-threshold guidelines.

Taking your numbers – 7.4 million people on Crestor for two years is $120 * 24 * 7.7m = $21.3 Billion.

If over that period we have saved 66,000 people, that’s $322,909 per life saved.

An overt body would look at that and make a straightforward decision. But I suspect that even a covert body would do the same math – and if Crestor were $12 a month, or $32,290 per life saved, they’d probably approve it, even though their public statements would say nothing about cost. I’ve no idea what AstraZeneca’s internal ROI needs are for Crestor though – though I will bet that it’s a lot higher than $12, which is generic-level pricing.

It will be interesting to see to what extent pharma and device companies try to “test” the system in order to learn the secret rationing numbers, and if an implicit QALY threshold develops over time.