Archive for April 19, 2017
Surgical checklists work–at scale
In an article in the Annals of Surgery this week, some of Dr. Atul Gawande’s team at Ariadne Labs demonstrated that widespread implementation of his surgical checklist lowers mortality on a massive scale: the entire state of South Carolina. Lead author Dr. Alex Haynes showed that not only does a standardized procedure work in particular settings, the benefit of the checklist can be implemented at scale.
Working with the South Carolina Hospital Association, they enrolled fourteen hospitals across the state to use the checklist as standard operating procedure, and then compared their surgical mortality over a three year period with the 44 hospitals that didn’t adopt the checklist. Mortality at the beginning of the trial was comparable between the two groups of hospitals, about 3.4-3.5%. After implementation of the checklist, the surgical mortality in the implementing hospitals dropped to 2.84%, while the control group mortality actually rose to 3.71%. This amounted to a relative reduction of 22%. Surgical mortality had been cut by almost a quarter, though the use of a simple questionnaire.
Why is this a big deal? So many times we find that something exciting happens in a small, controlled environment, but we are frustrated when we try to make it “just the way we do business”. What we don’t recognize is that while we consider the environment surrounding the experiment to be normal for us, significant conditions can be very different in other places at other times. For example, some of Jack Wennberg’s early experience in sharing variation among surgeons in New Hampshire was that once he showed them how differently they practiced, their variation narrowed almost automatically. For Wennberg, doctors taking action on data was normal.
But what Wennberg perhaps didn’t appreciate was the setting in which those conversations occurred. They occurred within communities in which the surgeons knew each other well, and trusted each other’s judgment. This was normal for them. In other settings without that community feel, many of us have been frustrated that changes sometimes take much longer to happen, or sometimes don’t happen at all.
So for Dr. Haynes to demonstrate that despite the likelihood that his sample of hospitals differed significantly in their characteristics, on average, using checklists saved lives. Patients did better even when the hospitals weren’t exactly like one another. And that is a significant and hopeful finding.
Want Healthcare LLC 