In this exclusive series, freelance writer Pamela Mills-Senn tells the story behind some landmark investigative journalism she published regarding Cincinnati’s infamous Dr. Henry Heimlich.  In this second installment, she describes what it was like realizing that Dr. Heimlich’s research was in error—particularly given the implications this error had for potentially causing harm to innocent people.

-----
(Read Part One:  ”Drowning in Funworld.")
-----

I didn’t receive duplicate research papers from both sides in every case. One such instance involved a study published in a June 1989 issue of Pediatrics that was conducted by Linda Quan, M.D., internationally known for her work in pediatric drowning. Heimlich had sent me just the first couple of pages of her study; Hazinski had not included this in her packet of materials.

Quan’s research was important to Heimlich because it was from this study (entitled “Ten-Year Study of Pediatric Drownings and Near-Drownings in King County, Washington: Lessons in Injury Prevention” ) that he derived a shocking figure—that the use of CPR to resuscitate children pulled from public pools resulted in a 42% fatality rate. Heimlich further stated in his own materials (in which this figure was used over and over again to demonstrate CPR’s ineffectiveness for all populations) that the children in Quan’s study had been given CPR by Red Cross-trained guards.

It was Quan’s study I held in my hands that early morning and it was this study that contained the red flag that had been flapping—completely overlooked—in my face all this time.

But before going further, here’s a little background on me.

I’ve earned two Bachelor’s degrees, one in psychology and one in anthropology. I completed two years of Master’s level coursework in anthropology, with a focus on medical anthropology and how culture affects compliance. I opted out of the thesis and instead, wound up at a small magazine publishing company doing research and marketing, an association that lasted about 14 years before I struck out on my own as a freelancer.

Both my majors were heavily research oriented. I spent six years as an undergrad learning the difference between well-constructed research and research that was not up to standard. Throughout the eight years I was at the university I took a variety of classes on research methodology, statistics and the like. I know how to tell good from bad, valid assumptions from invalid and so on, although I’m not an expert—certainly not at the level of say, someone with a PhD or an M.D. behind his or her name.

Looking at the Heimlich-supplied partial copy of Quan’s study, and then at his materials where he quotes the 42% death rate associated with the use of CPR, I realized that he had extrapolated from a small, self-contained regional population and translated the data derived from this population as being true across the board for everyone. Nowhere in the papers that he provided to me at that time, nor in the copies of his correspondence that he sent to various folks trying to win support for using the maneuver for drowning resuscitation, did he mention this was a regional study and therefore valid only for that population.

I was stunned that he would commit such a rank error, one that any beginning statistics or research student would be rapped across the knuckles for.

Here’s why you can’t do what he did. Say for example, you wanted to do a study on the average amount of time that people, unprotected by any sunscreen, could remain in the sun before burning. Now, say you included only redheads in this study, or only African Americans. How valid do you think it would be to extrapolate from these groups to the entire population at large? (Not to mention you would have to take varying levels of pigmentation into account along with a ton of other variables. You get the idea.)

Or, consider the recent revelations around heart attack symptoms. For years, researchers basically studied males and extrapolated from these studies a list of warning signs commonly believed to be true for everyone. But, as it turns out, the warning signs for women are quite different than those for men. If women had been included in these studies, or if researchers had questioned their assumptions that men and women are alike in this respect, a more scientifically accurate picture would have emerged.

What Heimlich did may seem like a small deal, but it wasn’t. It was enormous. My first reaction was to question myself; maybe I was wrong about this, maybe this extrapolation was valid. But I knew better. And if I knew better, why didn’t Heimlich? Could he have done this in error? Could this have been deliberate? Either way, it didn’t look good.

As I tried to wrap my brain around this, I waited for the workday to begin, too unnerved to sleep and for once grateful that my editor was hours ahead of me here in California. Finally, around 7:00 am, I placed the call.

“There seems to be a problem with how Dr. Heimlich is applying his data,” the first of countless times I would say this exact phrase.

I explained. He listened in silence. We were both thinking the same thing—oh shit. What could this mean for Ellis & Associates, who also relied on this figure?

He agreed we couldn’t move forward and gave me a few weeks more to dig around; weeks that turned into months—six to be exact.

Having only a partial copy of Quan’s study, and freaking out about the implications of what I thought I was seeing, I decided to call her. It was easy enough to find the hospital’s number where she worked through information. Maybe Quan would tell me that Heimlich’s use of her data was correct after all.

The message I left for her was the same as what I told my editor, “There seems to be a problem with how Dr. Heimlich is applying your data.”

The message I left for Hazinski was the same, “There seems to be a problem with how Dr. Heimlich is applying his data.”

And then I waited. And the phone calls started coming. And papers and documents and research started arriving. And then the bad dream began to take shape.

Next:  The Straw (Man) Argument

Read Part One:  Drowning in Funworld

Socializer!

del.icio.us

Furl It

Anonymous comments are allowed, but you can log in above to stamp your name and to avoid typing the anti-spam code.

If you are not familiar with our rules for leaving comments, click here!

1. So Far So Good... says:
   06 May 2006  at  04:31 pm | #

   I like this story but unless she’s writing this as she goes along, the unfolding drama is annoying.  Cool for the Enquirer but readers at the Beacon digest stuff like this along with their lunch, even Blue Gill.

   You can chalk it up to my interest, I suppose, but I’d rather you get to the point.  So far there hasn’t been enough grist for the mill out of these two postings to warrant this unfolding drama.  Where’s the beef?

2. says:
   06 May 2006  at  07:27 pm | #

   I do like watching it unfold in stages.  While we’re not to the beef yet, we’re coming in around six pages.  If the whole thing is twelve, that might be a bit long for a lunchtime read, as you mention above.

   I still think this is a great peek at investigative journalism in action.

   I hope some of the so-called real journalists in town are taking a lesson from Mills-Senn!

3. anon says:
   06 May 2006  at  08:35 pm | #

   This is fascinating stuff. I’ve read some of the other articles about Dr. Heimlich, like his weird experiments infecting AIDS victims with malaria. Isn’t it obvious that there’s something very, very wrong with the guy?

4. says:
   07 May 2006  at  12:11 pm | #

   An article by Dr. Heimlich on his website—which, by the way, was not published in a medical journal, but by an organization called the “International Swimming Hall of Fame”—includes the same 42% statistic:

   A 42% death rate of those drowning in the presence of lifeguards should have caused concern about the method the lifeguards were being taught.

   Read Heimlich’s article and you’ll see he uses common—and deceptive—rhetorical tricks. As Mills-Senn points out, the 42% statistic is meaningless. Nevertheless, Heimlich uses it to erroneously conclude that CPR is ineffective. Then he uses that false conclusion to arrive at another false conclusion—that lifeguards are being improperly trained.

   Using that kind of logic, XYZ Paving Inc. might come up with this: “There are traffic accidents on I-75. This raises concerns that Ohio highways are incorrectly paved and other paving methods should be considered.” The company creates a problem based on a false premise. It’s a safe bet that XYZ Paving just happens to have a creative solution which their company would be happy to implement.

   Dr. Heimlich’s argument moves in the same direction. As it turns out, he’s got a better idea to rescue drowning victims: the Heimlich maneuver. Does he have any clinical evidence to support his theory? No. Do any drowning experts agree with his theory? He hasn’t provided any. Okay, then how did he arrive at his conclusion that the maneuver is better than CPR? Well, he says he’s got “many documented reports”:

   Many documented reports of drowning rescues by highly qualified professionals show that the Heimlich maneuver successfully revived drowning victims after CPR failed. These rescuers describe how water gushes from the mouths of victims with application of the Maneuver, then the victims recover. Among rescuers who have saved drowning victims with the Heimlich maneuver after CPR failed are Dr. Victor Esch, chief medical advisor to the Washington, D.C. fire department; Terry Watkins, Emergency Medical Technician and Chief of the Destin, Florida Fire Department; and Ron Watson, Vice President of the U.S. Lifesaving Association and Park Director of Jacksonville, Florida. In an official sheriff’s report to the mayor, Mr. Watson said, “Before you try to provide oxygen to the victim, it makes sense to clear water from the lungs.”

   In other articles, Dr. Heimlich has said that a 1980 Lima, Ohio case in which Dr. Edward A. Patrick saved a two-year-old of uncertain gender unequivocally proves his theory. (Heimlich and Patrick sometimes say it was a girl, sometimes they says it was a boy, but what’s a little detail here and there?)

   But let’s take Dr. Heimlich at his word. He says the Patrick, Esch, Watkins, and Watson rescue cases and other cases he has published prove his theory. (Never mind that long before their Heimlich drowning rescue cases, Patrick and Esch had longstanding relationships with Heimlich. Hey, maybe it’s just a coincidence that they were the alleged rescuers in his two primary case reports.) Heimlich claims to have built a better mousetrap and he says these cases prove it. He’d want to share the details in order to make his case, right? What legitimate scientist wouldn’t jump at every opportunity to show and defend his work?

   Recently, letters were sent to Heimlich by two physicians,Dr. Peter Wernicki, medical advisor to the US Lifesaving Association, and Dr. Robert Baratz, president of the National Council Against Health Fraud. Heimlich was asked to provide medical documentation to back up these his cases with real medical evidence, not just hearsay. Last year I wrote a similar letter to Deaconess Associations, parent organization of the Heimlich Institute.

   Heimlich and Deaconess will not—or cannot—provide any supporting evidence. Regarding Heimlich’s “many documented reports,” this may raise doubts about the quality of that documentation.

   Dr. Heimlich’s article, he also claims:

   Since 1995, the Heimlich Institute has received more reports of drowning saves with the Heimlich maneuver than choking saves.

   Dr. Heimlich, Phil Heimlich, Tony Woods, Barbara Lohr - if you’re reading this, I’d like to see all those reports of drowning saves. Perhaps some other reporters would, too.

5. anon says:
   07 May 2006  at  09:14 pm | #

   Sounds like Dr. Heimlich faked case reports. If so, it wouldn’t be the first time a doctor fabricated evidence, not by a long shot. But if someone as famous as he is committed case fraud and a reporter could prove it, that could be a big story.

   Dean or others, any idea what was in it for Heimlich? Money, fame?

6. A Few Other Blogs says:
   08 May 2006  at  12:12 pm | #

   After Reading Nikki and Nate’s Blog, (The last few entires) seems that the Cincinnati Beacon is the only one worth reading. That comment is not meant to be offensive. But I think we are all tired of teh Kabaka Oba saga. Why don’t you talk to them about giving it a rest? You seem to be friends with them given your comments about them. Their blogss are unbecoming. (more personal things than newsworthy)