Kaiser Permanente sponsored a special event in DC today – Charlie Rose interviewed Dr. Atul Gawande about patient safety in front of an audience of physicians. Dr. Gawande is a young surgeon at Harvard’s Dana Farber Cancer Institute, has written two books about performance improvement, and is a regular contributor to the New Yorker magazine. I had heard many positive things about Atul, but had never met him in person. I was pleasantly impressed.
Atul strikes me as a genuinely humble person. He shifted uncomfortably in his chair as Charlie Rose cited a long list of his impressive accomplishments, including writing for the New Yorker. Atul responded:
I’m not sure how my writing became so popular. I took one fiction-writing class in college because I liked a girl who was taking the class. I got a “C” in the class but married the girl.
He went on to explain that because his son was born with a heart defect (absent aortic arch) he knew what it felt like to be on the patient side of the surgical conversation. He told the audience that at times he felt uncomfortable knowing which surgeons would be operating on his son, because he had trained with them as a resident, and remembered their peer antics.
Atul explained that patient safety is becoming a more and more complicated proposition as science continues to uncover additional treatment options.
If you had a heart attack in the 1950’s, you’d be given some morphine and put on bed rest. If you survived 6 weeks it was a miracle. Today not only do we have 10 different ways to prevent heart attacks, but we have many different treatment options, including stents, clot busters, heart surgery, and medical management. The degree of challenge in applying the ultimate best treatment option for any particular patient is becoming difficult. This puts us at risk for “failures” that didn’t exist in the past.
In an environment of increasing healthcare complexity, how do physicians make sure that care is as safe as possible? Atul suggests that we need to go back to basics. Simple checklists have demonstrated incredible value in reducing central line infections and surgical error rates. He cited a checklist initiative started by Dr. Peter Pronovost that resulted in reduction of central line infections of 33%. This did not require investment in advanced antibacterial technology, and it cost almost nothing to implement.
Atul argued that death rates from roadside bombs decreased from 25% (in the Gulf War) to 10% (in the Iraq war) primarily because of the implementation of check lists. Military personnel were not regularly wearing their Kevlar vests until it was mandated and enforced. This one change in process has saved countless lives, with little increase in cost and no new technology.
I asked Atul if he believed that (beyond check lists) pay for performance (P4P) measures would be useful in improving quality of care. He responded that he had not been terribly impressed with the improvements in outcomes from P4P initiatives in the area of congestive heart failure. He said that because there are over 13,000 different diseases and conditions, it would be incredibly difficult to apply P4P to each of those. He said that most providers would find a way to meet the targets – and that overall P4P just lowers the bar for care.
Non-punitive measures such as check lists and applying what we already know will go a lot farther than P4P in improving patient safety and quality of care.
Atul also touted the importance of transparency in improving patient safety and quality (I could imagine my friend Paul Levy cheering in the background). In the most touching moment of the interview, Atul reflected:
As a surgeon I have a 3% error rate. In other words, every year my work harms about 10-12 patients more than it helps. In about half of those cases I know that I could have done something differently. I remember the names of every patient I killed or permanently disabled. It drives me to try harder to reduce errors and strive for perfection.
Atul argued that hospitals’ resistance to transparency is not primarily driven by a fear of lawsuits, but by a fear of the implications of transparency. If errors are found and publicized, then that means you have to change processes to make sure they don’t happen again. Therein lies the real challenge: knowing what to do and how to act on safety violations is not always easy.
Charlie Rose asked Atul the million dollar question at the end of the interview, “How do we fix healthcare?” His response was well-reasoned:
First we must accept that any attempt to fix healthcare will fail. That’s why I believe that we should try implementing Obama’s plan in a narrow segment of the population, say for children under 18, or for laid off autoworkers, or for veterans returning from Iraq. We must apply universal coverage to this subgroup and then watch how it fails. We can then learn from the mistakes and improve the system before applying it to America as a whole. There is no perfect, 2000 page healthcare solution for America. I learned that when I was working with Hillary Clinton in 1992. Instead of trying to fix our system all at once, we should start small and start now. That’s the best way to learn from our mistakes.