Surgeon and writer Atul Gawande’s latest book is a simple, compelling story about how the author, commissioned by the World Health Organization to improve the safety of surgery around the world with a minimal budget, did so with a 19-item checklist. Of course, Gawande didn’t work alone; the checklist was the result of an international effort of doctors and researchers. And the effort isn’t over. While every country that has implemented the checklist has seen a significant decline in complications and deaths caused by human error (not to mention millions of dollars saved on complications-related procedures and malpractice lawsuits), many governments (including the United States) and hospitals have yet to mandate it.
Part of the resistance may be the humble checklist’s reputation as a crutch for the forgetful, incompetent, or especially obsessive-compulsive. It is these perceptions that the author tries to overturn in The Checklist Manifesto. He does so handily in the first half of the book, showing through interviews with professionals how checklists are used to undeniable advantage in construction, aviation, and finance. The idea is that with projects of sufficient (and, these days, always growing) complexity, it’s simply not possible for humans to remember everything, all the time. Surgeons today have more training than ever before, yet still sometimes forget to take all their tools out of patients before sewing them up.
Another point of resistance may be that people in positions of power don’t like being told what to do–whether it’s by a checklist handed down from above, or by a subordinate reminding them that they’ve missed a step. But the point of a checklist isn’t to dictate how important decisions should be made. The point is to offload the simple stuff (“Is this the right patient?”) so people can focus on the nuanced, complicated, attention-absorbing stuff.
- Checklists come in two varieties: DO-CONFIRM and READ-DO, having to do with whether you first check the list or perform the action. DO-CONFIRM lists work best for projects where you want to allow autonomy but still make sure you’ve got everything covered. READ-DO checklists seem more suited to procedures where the order of steps is critical, or emergency situations where people may be comforted by being told exactly what to do.
- Checklists need clearly defined pause points (e.g. before making the first incision) where you stop and actually check things off.
- They must strike a balance between brevity and effectiveness. Leave things off and they may get overlooked. Include too many items and the checklist can be burdensome. On their surgical checklist, for example, Gawande’s team eliminated checks related to fire-prevention (“Are the oxygen lines sealed tightly?”) because surgical fires, while terrible, are exceedingly rare and several different things must be checked. Performing surgery on the wrong patient is rare too, but since this is a quick detail to check, it made the cut. The author admits, “there was nothing particularly scientific or even consistent about the decision-making process” for what gets put on the checklist. But experience can help since…
- Like any designed thing, checklists must be tested in the real world to see if they’re sufficiently clear, simple to follow, and actually helpful. One emergency procedure checklist for airplane pilots includes the line “FLY THE AIRPLANE!”, which probably seemed unnecessary to the writers until testing their checklist under “real” emergency conditions in the flight simulator.
On a final note, one of the most interesting items on the WHO’s surgical checklist is for everyone involved in the operation to introduce themselves by stating their name and job. I’d never have considered this important, but studies show a correlation between the success of operations and whether or not those involved knew each other’s names. On some level this is to be expected: when every second counts, “John, hand me the defibrillator” is going to produce less confusion and a swifter response than “Hey you, hand me the defibrillator.” But less directly, knowing each other’s names creates an atmosphere of teamwork, and the increased collegiality makes people less hesitant to speak up when something seems not quite right.