Wednesday, March 10, 2010

Lean retreat, day 3: The Pig Game (learning about standardized work)

Modern healthcare is complex, and managing complexity is no laughing matter: it can be life and death. (More on this below.) As always this leaves us with the question, what can we do that would make any difference? That's where Lean comes in.

One aspect of Lean is standardized work - where "standardized" means a shared and continuously improved standard approach. Physicians are given leeway to use their judgment, but all in the enterprise must agree that the right way is to agree on a method, all use it, all contribute their ideas for improvement. And track what works.

That's how we know, for instance, that the central line infection rate has dropped so dramatically, as I mentioned the other day.

Well, to teach us about standardized work, today we did a classic Lean exercise: the Pig Game. Everyone is given a "tic tac toe" grid, and instructions on what to draw. In round 1 everyone's pig comes out vastly different, though all had the same spoken instruction. In rounds 2 and 3 .... well, I won't spoil it, but I'll say that the same group of humans produced far more uniform results when the work was defined more effectively. :)

This morning we heard from Steven J. Spear, author of Chasing the Rabbit. For homework we read chapters 2 and 3, which made a compelling case that in today's enterprises complexity is the rule and it's the thing that must be managed: no longer is it sufficient to be good at what each of us does - we must manage, together, the uncontrollably complex interaction between our areas of knowledge.

And yes, when I say "we," I'm including patients. I understand more clearly than ever why patients must be engaged in their care, in and out of the hospital. (Actually that's only true if they want the best possible outcome from their treatment. I do.)

The book relates a heart-rending story of a woman who was accidentally given the wrong fluid and died. I won't try to depict the circumstances because anything short of the whole story would give you the wrong impression; I'll just say that when I read it I cried, not just for the patient but for the nurse who did it. (Apparently the wrong tiny vial, looking almost identical, was in the cart.)

It was heart-rending, as I say; then he documents how very similar failures in process and policy led to NASA's Challenger disaster.

The very first business transformation course I ever took, many years ago, was from a company named Innovation Associates. The course leaders were Charlie Siefert (coincidentally a fellow member of my college glee club) and Peter Senge, who has since become famous for the "Learning Organization" concept, about enterprises whose central competence is to learn new competences. In that course we were taught two fundamentals:
  • How to envison a future, unconstrained by current reality. Very right-brain.
  • System dynamics: an understanding of how complex systems work. Totally left-brain.
It was a full weekend, back and forth between the cerebral hemispheres. The climax was apocalyptic: on the last afternoon an exercise led us, unsuspecting, to a conclusion that's horrid to any engineer: you can't figure it all out. You have to think it out as best you can, start operating, and see how it goes, adjusting on the fly.

And since in a complex system problems are likely to arise no matter how much you plan, you need to be really honest about how it's going. It's truly inspiring to work with a group that's dedicated to improvement like this. (And producing great results.)

And for the raw-business data junkies out there, here's today's Results Tidbit:

Yesterday I described how the Beth Israel Deaconess business transformation team worked with the orthopedics department. One outcome was a total rework of how hip replacement surgery was done. Not only did they improve quality, they can now schedule three total hip replacements per day, not two.

Yeah, folks, that's a 50% increase in business and better quality.

No comments:

Post a Comment

Your comments will be posted after they are reviewed by the moderator.