Saturday, December 04, 2010

No such thing as a random failure

#IHI Joseph Gavin, Jr., who died in November, was an aeronautic engineer who was intimately involved in the design of the first manned craft to land on the moon. He was also a key player in the rescue of the Apollo 13 astronauts. He was a remarkable fellow, and I had a chance to hear a tribute to him during this past week's meeting of the MIT Corporation. (He was in the class of 1941.)

One of the quotes ascribed to him during the presentation was, "There is no such thing as a random failure."

In this discussion board, a commenter says, with regard to that quote, "Amazing when you look at things now, that in the avionics industry of the time 'random' failures were acceptable! As he says, there is (almost) no such thing as a random failure... Everything has a cause, and in a safety critical system (or one-shot system like this), every failure cause has to be designed out..."

It strikes me that there is a parallel with medical care. I have discussed the problem of "These things happen" that often characterizes the delivery of care. I noted:

Several years ago, we had that attitude in our hospital with regard to certain types of medical outcomes. For example, we were content with our level of central line infections because we were below the national average. After all, these things happen. Then our chiefs of medicine and surgery said, "No, they don't have to happen. When they happen, people die. We are going to insist that we achieve zero central line infections." And then they got to work. As I have noted below, it is not an easy problem to solve, but it is worth the effort, and you can improve.

The science of process improvement can be applied to the delivery of care, as it has been to other high performance service and manufacturing industries. I hope my readers will excuse the degree to which I focus on this topic, but I cannot imagine a more important subject to cover.

This week, several thousand people will be attending the IHI Annual Forum to learn and trade information and stories along the theme of Taking Care. Maybe, if we learn well enough, we can say that "these things" no longer happen.

Here's Ethel Merman, to make the point in her own way. (If you cannot see the video, click here.)


Victoria said...

From Facebook:

Years ago, my dad was a safety engineer at Bell Canada. He was one of the first people into the Zero Defects idea and Defensive Driving (This was all before I even met you). They reduced danger and incidents -- I won't call them accidents -- significantly.

Some time later I was teaching a driving safety class and we had some interesting old Defensive Driving videos. One made the point "They're not car **accidents** -- if you know the casuse and choose to do it anyway, it's not an accident. They are car **collisions**" I've made that distinction ever since. So does the cop in Hot Fuzz [excellent dark humour -- see it.] Another video made the excellent point that *eighty to ninety percent* of auto collisions (and the following deaths and injuries and disabilities) are avoidable, simply by taking more time and space. Most people follow so closely that they are collisions waiting to happen.

People always give me a hard time about driving slowly, leaving a lot of space, being finicky about signalling -- but I and my old car are still here to witness that it works. People also give me a really hard time about triple-checking everything and going over checklists -- but when I plan and run the wilderness canoe trip they really appreciate the good food and comfortable campsite, not like the ones I let the other guys run when we had the tipped canoes, hypothermia, leaky tents, and one I returned from with a broken rib in a canoe for seven hours.

In other words, these ideas aren't new. Convincing the American medical establishment to wake up and pay attention to them, now that's new.

Tell me, is there any hope in Hades of getting the education establishment to pay any attention at all to outcomes? That's my problem.

wrinkledman said...

As a regular reader who sometimes comments on your political commentary, I usually just read about your hospital process improvement without comment. I am glad you harp on this. I'm with you. I can't think of a more appropriate response to preventable mistakes.

Dierdre M. said...

Interesting thought, but it bears some unpacking. Process improvement in healthcare is certainly a worthy goal and one that has benefited greatly from the attention focused on it by the IOM report. But saying that processes can be improved is not equivalent to saying that randomness can be engineered out of health care. Health care is fundamentally different from building airplanes. While both are complex, only one is living. As long as we are dealing with living organisms (human and not), their reactions and interactions to any intervention will have an element of unpredictability. For one thing, no one can predict the response of the organism to every intervention because we don't know everything there is to know about biology and biochemistry. For another, living organisms adapt, so human responses to bacteria and quasi-lifeforms such as viruses, can never be fully predicted. Consequently, there will always be unanticipated events in health care and some of these will be injurious, even fatal. Are these mistakes? It's tempting for those on the receiving end to think so.

People hunger for certainty, which is something that is impossible for physicians (or anyone else) to provide.

Anonymous said...


Your point is well taken. However, perhaps the point Paul is making is that, rather than first assuming an event is just part of that unpredictability ("these things happen"), instead we should first rigorously examine every event before relegating a small percentage of them to the random, unpreventable category.

I can tell you from my own training/practice experience, that has not been done in the past.

nonlocal MD