Data is not Optional

A man of genius makes no mistakes. His errors are volitional and are the portals to discovery. James Joyce


I just listened to a a Radiolab interview with Henry Heimlich.

It was quaint to hear Heimlich describe the invention of the Heimlich maneuver, testing it on an anesthetized dog using a chunk of meat tied to a string, jammed down the dogs throat. And his earlier invention, the Heimlich valve, developed for the battlefield to stop chest injuries from collapsing lungs.

Less quaint was Heimlich's advocacy of using his maneuver to treat asthma, cardiac arrest and drownings. And horrifying was his theory that infecting AIDS patients with malaria could cure them of AIDS. OK, the theory wasn't so horrifying, but his repeated human trials in Mexico and Africa certainly were.

Most interesting, however, is the lack of structure, science and regulation on any of these Heimlich 'discoveries/inventions'.

The Red Cross science advisory panel still debates the efficacy of slapping someone on the back (their primary recommended action) over the Heimlich maneuver. Their debate goes unresolved because there are no studies supporting either method.

With a little research, one discovers that the Heimlich valve (then called a flutter valve) was around in the Civil War. It seems Heimlich, in 1963, managed to find a flutter valve at a hardware store that didn't clog as easily. Employing an off the shelf solution with no data to back it up.

As for the Heimlich treatment for asthma, cardiac arrest and drownings - no FDA clearance on that!

And the human trials of malaria therapy for AIDS (and advocacy of it for cancer)? well, he left the US for that.

I celebrate the creativity of people like Heimlich, defend their right to be wrong and encourage their pursuit of atrocious ideas. But we still need to put rigor into our analysis, so that it could BE different.