Dirty Rotten Criminals

NPR ran a story today entitled “Catching Hospital Workers Dirty-Handed” in which they report:

“According to the CDC, health care-associated infections kill about 100,000 Americans a year, at a cost of billions of dollars.”

One solution being evaluated is the use of hi-tech wireless technology to spy on the hand-washing practices of individual doctors, nurses, and other caregivers.

The solutions we come up with depend on how we frame our questions. For example, in the NPR story they report that “The standard protocol in hospitals is for doctors and nurses to wash their hands on the way in to see a patient, and on the way out. But that doesn’t always happen — they get busy; they forget.”

If we frame the question in terms of who’s to blame…

• Workers who forget to wash
• Workers who are too busy to wash
• Workers who are too stupid to wash
• Workers who are hydrophobic
• Workers who want to spread death and disease

We end up with the answer that inspecting and policing workers is the best solution. Of course, when we institute a system for inspecting and policing people,  we also create a system of inspectors and enforcers against defectives and criminals.

Now think about this. That NPR report also says that, “Studies show that only about 40 percent of health care workers in the U.S. wash their hands as often as they should.”

No need for control charts here! Sixty percent non-sanitizers is a big, big, signal telling us that the problem is not one of individual performance. We are not dealing with outliers! If we understand the nature of a system and variation, we know immediately that we have a system that predictably and reliably produces 60% non-sanitized hands! We might be able to inspect out a small fraction of non-sanitized hands, at great expense, but we are foolish to expect a system that produces 60% non-sanitized hands to do what it does not do.

I'm sure you can come up with something better than this!

What if we frame the question differently? What if we ask, “By what means can we best assure that healthcare workers’ hands are free of disease bearing agents as those workers move from patient to patient?

Can we design a worker-sanitizing system that minimizes worker decision-making requirements regarding whether to sanitize or not to sanitize?

If we could accomplish this, wouldn’t  everybody win?