Drug Companies, Research, Marketing, and JUPITER

[Turtle]

Dr Steve, what are your thoughts about the JUPITER trial and CRP?

[Dr Steve]

Dr Steve, what are your thoughts about the JUPITER trial and CRP?

Here's the thing with these damned studies...the drug manufacturers love to tout the results when they look good for them, even though we're talking really, really small numbers here.

Let's talk about JUPITER for a minute. This was the Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin, funded, I believe, by the makers of Crestor. What they did was took otherwise normal people with pretty normal "Bad" cholesterol numbers, but high CRP numbers. The CRP (c-reactive protein) is thought to be a marker of whole-body inflammation. If you have an inflammatory process somewhere in your body, the CRP should be higher. Some thought has gone into the theory that inflammation is a mediator of diseases of all kinds, cancer and arterial diseases (like heart attacks) being high on the list.

So anyway, they gave people this drug, Crestor, that decreases "bad" cholesterol and lo and behold it decreased the CRP at the same time. AND there was a 44% decrease in the the incidence of heart disease in the Crestor group.

So what's the problem? Well, the total numbers of heart disease patients were so low overall, that indeed the Crestor group had 44% less cases, but in ABSOLUTE terms, you'd have to treat 95 people for TWO YEARS to prevent one case. Now, multiply that times 100,000,000 people and you've got something, but for the 94 people it didn't help, it's pretty disappointing.

So 44% reduction sounds pretty amazing. 1 case out of 95 over two years not so much. Guess which one they promote? :-)

Now, someone in my family is a drug rep, and I am not one of these "I hate drug companies" people. But statistical analysis is something that most people (including doctors) aren't good at. The American Academy of Family Physicians is doing something about this; they have instituted the STEPS program for drug evaluation. One of the parameters in this program is the "number needed to treat", which will help doctors and patients understand the potential benefits or pitfalls of a new drug.

your pal,


steve

[Turtle]

I would love to get more studies that show clearly number needed to treat as well as actual risk reduction as well as RRR.