Saturday, August 04, 2012

Medicine and Anti-Fragility

Black Swan author Nassim Taleb in his highly-anticipated upcoming book uses the term anti-fragility to describe systems that are the opposite of fragile. He coins the term after a lengthy search of the world’s languages convinced him that there is no other word to describe something that is the opposite of fragility. The obvious “robustness” doesn’t capture the sense of how some systems not only survive, but thrive when exposed to attempts to break them.

He’s been leaking chapters regularly on his web site, and a few days ago he released Chapter 21, about medicine. He looks at iatrogenics, the study of medical mistakes, as evidence that a large amount (perhaps the major part) of what today’s doctors take for granted is as unproven as bloodletting:
  • Put ice packs on a fresh sprain.
  • Eat breakfast 
  • Lower the temperature of someone with fever

To him, statin drugs are a dangerous example of intervention “to get a grade to pass a school-like test”. Yes, it effectively lowers cholesterol, but so what? It’s like muzzling a baby to stop the crying: sure, it works but you haven’t addressed the root cause. 

He has similar concerns about antibiotics and disinfectants, insulin injections, and even toothpaste: each effectively treats a “problem” but do we really understand the implications?

I’m sympathetic to the whole perspective, especially how “experts” are often (usually) wrong, etc., but I think Taleb is in danger of being too correct. Like the old joke about the two economists who ignore a $20 bill laying on the ground (“it can’t be real, or somebody would have picked it up already”), finding ways to “improve” things is pretty much what humans are for. 

In the wrong hands, Taleb’s concerns might sound like a defense of the precautionary principle, the easy-to-refute idea popular among environmentalists, that a theoretical, even unproven potential harm is enough to justify restrictions on new technology or development.

You could apply Taleb’s reasoning to say that we shouldn’t make gardens because by artificially hedging trees or cutting grass, we are interfering with natural processes we don’t understand. Well, yeah. But if we don’t interfere with nature we don’t get a garden – and maybe I want the garden. Sure, maybe there will many unintended consequences (my fruit trees will attract annoying birds, my vegetables will attract deer, etc.) but I’ll deal with those problems as they come up.

I apply a compress to a fresh wound and take aspirin for a fever because maybe I want to treat the symptoms. Forget the long-run issues of the consequences of “unnatural” treatments: in the long run, we’ll all be dead.