Monday, November 22, 2010

Checklists ain't all

I am often approached by people who have read Atul Gawande's excellent work and who ask me if we use checklists to improve the quality of care in our hospital. We do, in order to reduce variability where appropriate, but stating the issue in terms of checklists is simplistic. This 2009 Lancet article by Charles Bosk, Peter Pronovost and others, entitled "Reality check for checklists," states it clearly:

The emphasis on checklists is a Hitchcockian “McGuffan”, a distraction from the plot that diverts attention from how safer care is really achieved. Safer care is achieved when all three—not just one—of the following are realised: summarise and simplify what to do; measure and provide feedback on outcomes; and improve culture by building expectations of performance standards into work processes. We propose that widespread deployment of checklists without an appreciation of how or why they work is a potential threat to patients' safety and to high-quality care.


Anonymous said...


Applying a tool without understanding what it does is a recipe not only for failure, but also a great way to convince yourself and others that the whole quality movement is bunk. One example is applying parts of lean without understanding the three pillars; it can lead to layoffs WITHOUT quality improvement, which is disastrous (and inept management).

Said differently, there's just no substitute for knowing what you're doing.

Epdave on a feeble handheld

Anonymous said...

Great article, thanks. Another quote from it helps bring your point home in a way all of us can understand:

"First, checklists are suited to solving specific kinds of problems, but not others. The success of checklists in preventing disasters during the takeoff and landing of commercial aircraft is often pointed out. But checklists are also used to track baggage for the airlines. On this task, checklists perform less admirably. Handling baggage that comes in different sizes and shapes, involves complex transfers, and is often in poor condition, is a more realistic analogy for use of checklists in achieving patients' safety than their use on takeoffs and landings."


Anonymous said...

Where is the pressure for clinicians in academic (i.e. non-employee status) hospitals to review and compare step-by-step individual practice behaviors? Many surgeons (and others) appear to work from an internalized and idiosyncratic checklist. How do they know if what they learned at med school is the best in terms of safety, efficiency, effectiveness, etc? Does a Chief of Surgery say to his department "OK. We've had some errors. Let's compare our work."?

Matthew said...

From Facebook:

Seems like a straw man argument to me. Who is arguing that there should be checklists *instead of* process improvements? Who is arguing that a checklists are a replacement for situations calling for clinical judgment? For that matter who is arguing that checklists are a *sufficient* condition for patient safety?

The bottom line is that smart people don't like to admit they make stupid mistakes, but stupid mistakes are the precisely the type of mistakes smart people are most likely to make.

Anonymous said...

Dear Anon 8:54,

Much of that review occurs in M&M conferences after something goes awry, but that kind of setting, as you have recognized, is not generally the best for sustained process improvement.

Elaine Schattner, M.D. said...

I see a lot of value in the checklists (see related ML post: But I agree that their scope is limited. The real issue is defining, within medicine, which kinds of practices are amenable to and improved by this sort of strategy, and which are not.

Unknown said...

This is a great thread, and very timely. A few observations:

As noted, any tool is only good if people know how to use it. That means training, standardized training, for ALL clinicians. There are known best practices for using a checklist, and they're not obvious, so they must be taught.

High-reliability organizations that use checklists, use them with discipline. This includes concepts like, "We never start this procedure without the checklist," and "Skipping and item on the checklist, or calling an item complete when it isn't, is sacrilege." I don't know if this culture can be imported, but it's helpful to expose your organization to professionals who live in such a culture until it is cooked into your own. Dr. Gawande has recommended the same in several forums.

Half-hearted implementation is worse than doing nothing, because it sets up an expectation that we're reliably doing the "must do's", when we may or may not be actually doing them. I have yet to audit a single organization that is "doing" the WHO Surgical Safety Checklist, where there aren't items being skipped. This is dangerous because instead of relying on pretty good human memory, we're relying on spotty compliance with a standardized process intended to replace human memory.

If you want placeholders for clinical judgment, as you should, have those who will use the checklist create it, trial it, and revise it. This will also make it efficient and effective.

Finally, on the issue of peer review, physicians should audit one another in clinical practice for compliance with known best practices. As Mr. Levy notes, this is very different from M&M. As a comparison, every commercial pilot gets such an audit, with his/her license on the line, every nine months at a minimum.

Susanne said...

From Facebook:

It isn't really a straw man. People grab on to the "whats" like checklists, in part because they are easy to measure, and vastly underappreciate the importance of the "hows," the collaborative processes that result in improvement tools.

Beverly said...

From Facebook:

Agree with Susanne. Matthew; the lay press, for one, presents this issue as a simple one of 'ask your doctor if s/he is using a checklist.'

Matthew said...

From Facebook:

So -- are you saying checklists should *not* be used because they don't solve *every* problem?

Not too long ago, I guy I knew went into a hospital for a hernia operation and ended up getting a colostomy he didn't need. It wasn't because the surgeon didn't know how to do the hernia operation. It wasn't because the surgical team didn't care enough. It was human error, specifically the kind that happens to people who are highly skilled and caring, but are working from a bad piece of information that could have been caught if somebody remembered to check.

Should we pooh-pooh measures that might have prevented this because they're *simple*? Does being difficult to measure somehow make a policy objective more worth pursuing?

Recently I was very sick with a bacterial skin infection which my PCP (a well known and respected doctor) misdiagnosed as shingles. He'd *insisted* I come into his office and not see some resident in the emergency room, and I believe that biased him toward a diagnosis I was unlikely to get at the emergency room. Like a good doctor, he knew my medical history well. The irony is that I'd have been better off with a resident. I've often noted that residents can give more helpful and informative advice, because their medical knowledge is still conscious rather than intuitive. In any case I didn't hold this against my doctor, because he's only human, and I *do* value his experience and intuition. But a system that made him think critically about his intuitive diagnosis might have saved me several days suffering with a temperature of 104 and some disfigurement. After a single dose of amoxicillin my temperature dropped to 99.

I'm not saying "don't improve your processes." I'm not saying "don't reform your organizational culture." In fact quite the opposite. *Do* reform your organizational culture, starting with the stubborn belief that only bad, poorly trained or uncaring people make mistakes, and the attitude that systematic improvements are only worth making if they are a panacea.

Unknown said...

I disagree Susanne...
With regard to the checklist - more specifically, the Pronovost checklist for central catheter insertion: The checklist is the 'how.' The 'what' is the desire to decrease iatrogenic infections; to reduce introduction of bacteria that cause infections; etc. Knowing the 'what' and having a simple procedure available that has proven the benefit of the 'how'is not contrary to lean or medical judgment...its just good common sense.

Anonymous said...

Robert R.A. Hoover, the famous pilot when asked how he flew so many different airplanes. "I always use the checklist".