Risk assessment

We are having a rain event here in Boston. Were it a snow event, the hair hatted newsbots would be proclaiming Armageddon. It's been bad for a long time, but in recent years they have actually ramped up their obsession with inclement weather, which I would not have thought possible. An inch of snow is enough for them to devote the entire local newscast to it -- of course there is nothing to say so they send some clown to stand on an eroding barrier beach, with no hat of course because he wouldn't want to waste his $100 coif, and announce with the snow piling up on his ears that there are waves and the eroding barrier beach might be eroding some more. The resulting mass hysteria has the sheeple clogging the supermarket parking lot with SUVs while they strip the shelves of milk and bread. It is idiotic beyond all understanding.

Anyway, back on topic. The FDA issued two alerts today, one concerning the diabetes drug Avandia, the other concerning two HIV drugs, saquinavir and ritonavir, which are often prescribed together.

Avandia has been in the news lately, with a Senate report alleging that the manufacturer Glaxo failed to inform the FDA of everything it new about cardiac risks associated with the drug. For the record, Glaxo denies the charge and says the facts are not all in yet on the risks of Avandia as a major trial is still underway. The FDA release only says yeah, we're waiting on results of the ongoing RECORD trial and we'll talk about it all in July.

I'm not going to predict how all this will turn out, but it does add more weight to the proposition that most new drugs should not get an "all or nothing" approval. In a case like Avandia, where there are alternatives available, it might make more sense to give a preliminary approval, which would allow only relatively small scale marketing with surveillance of patients who receive the drug, to get stronger reassurance that there are no long term safety issues that wouldn't have shown up in the original clinical trials. Critics assert that Avandia causes 500 heart attacks and 300 heart failures monthly. This may or may not turn out to be true but it's a chance we didn't have to take if we had a policy of gradual and conditional new drug approval under ordinary circumstances.

The saquinavir and ritonavir issue is a little different. These are so-called protease inhibitors and yes, they do have side effects, potentially serious ones, but they also do an excellent job of suppressing HIV replication in conjunction with one or more other classes of drugs and that's worth putting up with some risks. There are many patients who absolutely need them. When people face serious and ultimately life threatening illness and they don't have better alternatives, then a more expedited path to widespread use makes sense. (The new problem that has been discovered, as it turns out, is somewhat speculative and only effects some patients who are at risk for cardiac arrest.) It seems to me that these are distinctions we ought to be able to make even though. It would slow down some drug company profits but it would make more sense from a population health point of view.