Wednesday, June 8, 2016

Hey, number crunchers! If "the Heimlich" has saved 100,000 lives, why are the stats flat for US choking deaths? [UPDATED]

UPDATE, 11/10/16: Via Management of Airway Obstruction by Michael R. Sayre MD, Contemporary Cardiology: Cardiopulmonary Resuscitation, Humana Press (2005)
Despite widespread education on the use of the Heimlich maneuver and other techniques for treatment of acute airway obstruction, the (US) death rate remains stable.

Here's an interesting question for statistics experts.

How many people have been saved from choking to death by the Heimlich maneuver? 

Via At 96, Dr. Heimlich Uses His Own Maneuver on Choking Victim by Christine Hauser, New York Times, May 27, 2016:
In 1974, (Dr. Henry Heimlich) developed the method that compresses the lungs, causing a flow of air that carries the stuck object out of the airway and then the mouth. More than 100,000 people owe their lives to the technique, by some estimates.
I've come across that "100,000 lives saved" number many times. I've also seen 50,000.

To the best of my knowledge, both numbers originated with the Heimlich Institute, a Cincinnati nonprofit whose sole activity appears to be running Heimlich Heroes, a program that teaches young people how to perform the Heimlich maneuver.

Via the Heimlich Heroes website
Via the Heimlich Heroes website

Ever since my wife and I started researching my father's unusual career, I've wondered if there was a way to determine how many choking deaths have been prevented by his namesake treatment.

My math skills are challenged when I have to balance my checkbook, but the only approach that made sense to me was to look at choking death statistics.

That is, if the deaths of 100,000 or 50,000 choking victims had been prevented, presumably there would be a statistical decrease in the number of choking deaths since my father introduced the treatment.

And to adjust for population growth, it seemed to me that the only useful numbers would be choking deaths per capita.

The National Safety Council (NSC) compiles statistics on various unintentional injury-related deaths: motor vehicles, falls, drowning, fires, firearms, poisonings, etc.

According to the organization's 2016 Injury Facts, the NSC has been compiling choking death statistics from 1943 to the present.

That's three decades before my father introduced the treatment and four decades since, which seems like a reasonable and perhaps useful date range.

I first reported about the NSC's choking death statistics in an October 21, 2013 item about a problematic audio documentary about my father produced by Radiolab, a syndicated NPR program.

At the time, here's what Radiolab reporter/producer Pat Walters (who has since left the program) e-mailed me:
I checked my NSC stats, and it looks like I was wrong. I’d had an intern run the numbers for me initially, with the intention of checking them later, which I always do. Here’s what I’ve found: I only have data up to 2009 (from the 2011 report), which I believe you said you have, too. I’m waiting on the 2012 report from the com people at NSC so I can avoid paying 90 bucks for it. But anyway, according to that data and my back of the envelope calculations using population estimates from the US Census, in 1973 (pre-Heimlich manuever [sic]), choking was listed as the cause for 1.42 deaths per 100,000 people in the US. In 2009? The rate was 1.49 per 100,000.

So, if anything, the rate has gone up a bit. But just a bit. Not even significant, if you ask me. My take on this is that, essentially, almost nothing has changed.

But Mr. Walters isn't a statistics expert, so I'm hoping this item finds its way onto the screens of some capable number crunchers who might be able to shed some light on the subject.

Page down for the NSC's most recent (2016) per capita choking deaths per 100,000 population. Click here to download a copy.

I'm going to forward this to Nate Silver's crew at, but if you know any other stats experts who might want to chime in, I'd welcome any thoughtful analysis for a future item.

Click here for my contact info.

This item has been revised for simplicity.