Guest Post: Relying on Culture or Technology to Improve Patient Safety?
Today's post is from my friend Naida Grunden, author of the outstanding book The Pittsburgh Way to Efficient Healthcare: Improving Patient Care Using Toyota Based Methods.
In 2003, in a meeting room in Pittsburgh, then-Treasury Secretary Paul O'Neill led a discussion about the fledgling work at the local VA hospital to eliminate MRSA on a pilot unit, using the tenets of the Toyota Production System. The program was a joint venture between the VA, the CDC, and the small, nonprofit Pittsburgh Regional Health Initiative, which O'Neill then chaired.
That day, half a dozen guests lined the room — physicians and health leaders from Denmark. They listened attentively as O'Neill lectured on the importance of hand hygiene, the dismal baseline compliance observed on this unit, and ways in which Toyota methods could improve it. There was a LOT of discussion of hand hygiene that day.
At one point, an American physician in the room turned to the Danish guests and asked them what they thought of this discussion.
Bewildered, one of the Danes said, “This is not a discussion we would have in Denmark.”
The Danish physician then told the story of the new nurse that nobody liked. The nurses n the hospital break room gossiped about her, how her technique was poor, how the patients didn't like her very much. But by far the most damning bit of gossip — “the piece of information that actually got her fired — was, “She didn't even wash her hands when she went in the room to see that patient!”
Culture. What Lucian Leape defines as, “The way we do things around here.” That rugged American individualism doesn't always serve us well on those occasions when we need to be doing the same thing, the same way, every time.
So it was interesting to read Tina Rosenberg's take on using technology to encourage or enforce hand hygiene requirements in American hospitals. My overall response was, “Meh.”
Those of us who study Toyota-based approaches like Lean recognize that, while technology can eliminate whole classes of error, it can also introduce whole new classes of error if not designed thoughtfully. (Bell and Orzen's Shingo Prize-winning Lean IT: Enabling and Sustaining Your Lean Transformation provides a great discussion about what high-tech solutions can and cannot do.
At least it can be said, with Lean, we try the low-cost, low-tech first.
In my book, The Pittsburgh Way to Efficient Healthcare, I describe the efforts of MRSA reduction that started at the Pittsburgh VA, focusing initially on hand hygiene. There, starting in 2002, Alcoa-trained engineer, Peter Perreiah, dove deep into the culture on the unit to discover why staff members couldn't always observe proper hand hygiene. (Remember, when it comes to MRSA, staff members are also at risk of becoming colonized, and bringing it home to their families.) The nurses said they didn't have time.
So Perreiah set about using Lean principles like 5S, quick shift change, and reliable supply chains to free up their time. They began solving small problems at the front line and the bedside, making it easier to do things right the first time (like making sure gloves were always available). And making it harder to do things wrong (like using hot pink stethoscopes in the isolation rooms, creating a visual cue that they must STAY in the isolation room).
(story continues below ad)
Significantly, focusing on process, they created a way for patients to ALWAYS be tested on admission, transfer and discharge from the unit. That way, patients who had MRSA, or were colonized with it, could be put in isolation rooms where infection-reduction precautions could be taken with every encounter. It also created data that let them know whether a patient had been colonized or infected during his stay. Without this information, it was impossible to know who had come to the hospital with MRSA, or who had acquired it in the hospital. This information was an absolute baseline requirement for measuring the current state and any improvement.
Thinking this screening was a good idea was one thing: creating a reliable process for getting it done every time was another entirely. It took more than signs and admonitions, but it took constant vigilance and institutional understanding that this was the new “way we do things around here.”
The result of the layering of improvements over that first year was an 85% decline in MRSA at the Pittsburgh VA pilot unit. In other words, Lean process improvement had worked.
The VA also made a conscious effort not only to look at problems but to look at and analyze the stunning successes too. Think of it as doing 5 whys – not just on problems, but on sparkling successes. How'd they do that? What were their processes? And can it work here, too?
Fast forward 9 years. Since that tentative beginning in one unit in 2002, the VA generalized the MRSA reduction work over 150 hospitals, with 100% testing on admission-transfer-discharge and 100% hand hygiene as standard work. The result was a 62% decline in MRSA–a result just published in the New England Journal of Medicine and noted in the New York Times.
Although a contradictory study was also published in NEJM, it looks to me as if it was in a different hospital environment, where systematizing and standardizing this kind of testing would have been nearly impossible. For better or worse, the VA healthcare system is the closest thing the United States has to Britain's National Health Service (NHS) — where hospitals are owned by the government and physicians are government employees. Such a closed system does make the systematic adoption of 100% testing on admission-transfer-discharge more feasible and generalizable.
So congratulations to the VA Pittsburgh, to Drs. Jain and Muder who have overseen this work from the very start to the Lean thinkers and others who have had the perseverance to make testing and hand hygiene “the way we do things around here.”
And to those Danish physicians, we can only say we're still at it, trying to get American hospital culture to the point where failing to wash hands is cause for snip-snipping in the break room.
What do you think? Please scroll down (or click) to post a comment. Or please share the post with your thoughts on LinkedIn.
Don't want to miss a post or podcast? Subscribe to get notified about posts via email daily or weekly.
- Guest Post: The Time is Now for Lean-Led Hospital Design - March 7, 2013
- Guest Post: Relying on Culture or Technology to Improve Patient Safety? - May 11, 2011
- Guest Post: Captain Sully Talks about Lean and Healthcare Safety - March 25, 2010
Why don’t hospitals prepare as well for surgery as you will see astronauts preparing for missions, or soldiers for war?
I think the ‘process’ for creating doctors may need improvement.
Taking the brightest kids; sending them to med school to learn chaos and learn by rote (ie without guides and systems); telling them “you’re the best” ie don’t listen to anyone tell you what to do; promoting ‘heroic’ cases of medicine on television – unusual cases solved with heroic surgery; paying them vast sums of money irrespective of quality of outcome; having brotherhoods of surgeons who all agree to keep quiet about each other’s quality problems and not publish them: is this supposed to be a guarantee of good healthcare?
Most docs don’t fit into the above characterisation, but unfortunately enough do to tarnish the rest. We laugh at shows like Scrubs because there’s something in it.
I’ve had everything thrown at me in the past – ‘cookbook medicine’, ‘we are not a factory’, ‘what would you know, you’re not a [insert specialty]’.
The genie is out of the bottle, but it would be good to imagine a world where medicos were trained as if they had an important, but functional role as part of an acute or chronic care system, working with other equally worthy team members to deliver whole of episode or whole of life care together.
Our old friend “Human Error” has returned – this story about a train crash investigation is atrocious because apparently the only thing keeping two trains from colliding was whether the line controller ‘remembered’ – wasn’t that engineered out in the 1830s?
The article ends with:
It’s always better to try to be PROACTIVE rather than just reacting after the fact. Fixing processes is better than just blaming an individual, but I’m also surprised the system allowed so much human error.
We are all human; we are fallible. We must design systems accordingly, especially in healthcare.