Placebos

Wired magazine has a very interesting piece about how a larger number of modern FDA clinical trials of new drugs are failing against the placebo tests. If this is just because the drugs are rubbish, that would be one thing. But there's some evidence that the placebo effect seems to be getting stronger - old drugs that passed the placebo tests when they were approved now don't seem to show much of an effect.

This is not an ideal test, of course. It's also quite possible that this is simply reversion to the mean - the old drugs never really had an effect, passed the initial tests just by luck, and now are back to their normal level of ineffectiveness.

The article does describe the time-series changes in the placebo-only effect, though:

But if these same drugs were vetted now, the FDA might not approve some of them. Two comprehensive analyses of antidepressant trials have uncovered a dramatic increase in placebo response since the 1980s. One estimated that the so-called effect size (a measure of statistical significance) in placebo groups had nearly doubled over that time.

Let's assume for the purposes of argument that the placebo effect really has increased over time.

The reaction of the current FDA guidelines is to reject more drugs, as they don't seem to have any significant effect over sham medication. In terms of understanding the science behind the new medicine, this is a reasonable response - if you're not doing any better than sugar, your research isn't actually valuable.

But here's an alternative response that would make scientists aghast, but seems reasonable in terms of standard public health - if placebos are getting stronger, why doesn't the FDA just approve more placebos? They're doing the same thing, but costing virtually nothing! Hell, it's not like we need to spend billions of dollars on fundamental research to produce milk-and-sugar pills, and if the milk-and-sugar combination seems to produce better mental health at a cost of 0.2 cents per pill, let's start cranking it out!

The placebo effect is a real effect, but one that modern medicine is largely reluctant to embrace, because it makes their whole industry look like a joke. It says that a lot of the time they aren't doing better than witch doctors, and moreover that said witch doctors were actually having a real effect, despite not understanding anything (usually lacking any other effect than placebos, most of the time). This of course is deeply wounding to the pride of medical researchers.

But sod the medical researchers! If you believe these studies, increasingly valuable treatments from placebos are being denied to people, even though they'd cost pennies to produce.

Obviously there's some problems trying to do this under the current regime. The FDA couldn't tell people which medicines were placebos, otherwise they presumably wouldn't work. But it couldn't lump them together with the medicines that actually beat the placebos, without unfairly (and counter-productively) switching people away from more effective to less effective medication. If they only approved some placebos, this would create huge rents for the companies that they approved, unless they approved everything.

The simplest solution to all of this would be to simply abolish the FDA. I'm on board with this, but most people aren't. A more moderate (and politically feasible) alternative would be to restrict the FDA's mandate to simply testing for drug safety, not drug efficacy. Then we'd be able to capture the placebo effect to the fullest.

Science's reach must exceed its grasp, or else what is basic research for.

Scorning the placebo effect says that we don't trust anything we don't already grasp.

I say that if you can be pretty sure that what you're reaching is safe, reach away.