Brave hospital quality chiefs, a few hundred strong, sat before The Joint Commission president last week as he meted out some deflating news: They've got to do a better job of preventing death and injury in patient care.
In fact, said TJC President Mark Chassin, MD, U.S. airlines, military aircraft carrier flight decks, and even amusement parks—with their potentially dangerous fun rides—do a better job of recognizing and reporting process problems long before they cause harm. The healthcare industry needs to do that too.
Chassin made his remarks at the Institute for Healthcare Improvement 24th National Forum in Orlando, arguably the nation's amusement park capital. And during his talk, he urged hospitals and doctors to work harder to move their systems to "High Reliability," a phrase he used about 50 times within the span of an hour.
Consider these numbers, he urged:
The airline industry in the United States between 1990 and 2001 averaged 129 deaths per year, out of 9.3 million flights a year. Or 13.9 deaths per million flights. "In the 1990s, US air carriers were the safest air carriers on the planet by a large margin."
But in the next decade, 2000-2010, airlines became much safer, with only 18 deaths per year out of 10.6 million flights per year, for a death rate of 1.7 per million.
That decline, greater than eight-fold, was accomplished "in an industry that was already the safest on the planet. A characteristic of high reliability [is that] they are never happy with where they are today. They are always pushing the envelope for safety."
Safety in U.S. healthcare has been a different story, he said.
In "To Err Is Human," (the Institute of Medicine's 1999 report, of which Chassin was one of the authors) there were 44,000 to 98,000 deaths in hospitals due to errors a year, out of 34 million hospitalizations. Or a death rate of 3,800 deaths per million, Chassin said.
Compare this to 1.74 deaths per million in the same time period for the U.S. airline industry, he said.
"Hospital care is 750 to 1,000 times less safe than air travel in the United States," he said.
"And if you don't like that estimate, here's another," he said. "The Harvard Medical Practice Study (of 1991) showed that 1% of hospital patients were injured due to errors judged... to be negligent. That's a death or injury rate of 10,000 per million. The U.S. airline comparative number is 341 people dead or injured in 95.2 million flights, U.S. hospital care is 3,000 times less safe than U.S. air travel, or 3.6 per million. By this measure, hospital care is almost 3,000 times less safe than air travel."
Chassin acknowledged that the incentive for airlines to be safer is a bit different. He didn't say it, but everyone understands the ignominy an airline faces when 350 people fall to their deaths in one moment. Healthcare safety deaths are not publicized in any way that is nearly as impactful.
"But nevertheless, through a series of very clear steps, the [airline industry] got to where they are now, 1.7 deaths per million flights. And if we don't commit to zero, we're not going to know how close we're getting and we'll be satisfied with small incremental progress every year.
"We can't be satisfied with 5% better anymore," Chassin said.
Aircraft carrier flight deck and amusement park safety operations have "a bunch of things in common," Chassin continued. "They have very effective process improvement tools that allow them to create nearly perfect safety processes."
And, he said, "their culture encourages, obligates, and expects every person who works in the organization, from the most senior to the most junior, to be looking carefully at every process they touch every day [asking] 'Is this exactly the same as when it was performing perfectly yesterday?'
And they do this "before they create situations in which amusement park visitors might be harmed, when they're easy to fix. They're identified, recognized early, [and] they are reported, to the organization, [which] uses those effective process improvement tools to fix the process and that improvement is reported back to the folks who reported it in the first place. And that's how those organizations stay safe."
In healthcare, "we're too often facing a patient who's already been harmed, and then we try to work back with adverse event investigations, root cause analyses, asking why did it happen. And then try to figure out how we can fix our defenses so it doesn't happen again. But that's not how these organizations stay safe."
There are major efforts that must be strengthened and reinforced, he said. And the science of high reliability must play a more important role.
First, healthcare workers must trust each other from management down to housekeeping.
Trust means healthcare workers need to know they will be protected from "slings and arrows from their peers" when they step up and say there's a problem over here that needs to be fixed right away. And they need to trust that management "won't ignore or blame" the worker for reporting it.
Second, leadership from the board of directors and the CEO, down to the nursing ranks "all must be coalesced around the goal of zero harm to patients, and zero quality failures."
Third, providers must move to a "blame-free culture," but not so far that the employees with track records or patterns of behavior or errors are not held accountable.
"It requires assessing errors and patterns of behavior uniformly," Chassin said. "It requires having a process for eliminating intimidating behaviors... no matter who the person is, the most senior or the most junior—housekeepers, physicians—we need to apply the same process for evaluation and the same progression to disciplinary action," and it must be transparent.
"If you don't do this, you impair the trust component."
Chassin noted that problems of hand hygiene, wrong-site surgery, hand-off communication, surgical site infections, sepsis mortality, insulin safety, falls, and preventable re-hospitalizations persist, not for lacking of trying to get better.
But often in healthcare, the solution isn't a cookie cutter one that can be applied in a stamp-like form, with a checklist or protocol, at every hospital and on every floor, or with every type of patient.
What safety experts are increasingly realizing is that the problem might be the same, but the causes at one organization are quite different than the causes at another.
For example, he said, in an experiment with hand hygiene compliance at eight hospitals, experts realized there were 10 very different reasons for non-compliance, and those reasons varied at each facility.
At one hospital, the sinks and hand gel dispensers were in inconvenient locations. At another, there was no convenient place for workers to put down their items while they wash.
At another hospital, the safety culture did not stress hand hygiene at all levels. At another, there was the incorrect perception that hand hygiene is not needed if gloves are worn. At another, workers frequently simply forget to wash their hands. And so on.
When hospitals realize that they must individualize their own process obstacles, and work from there, they can move forward to that progressively less-elusive goal: higher reliability, Chassin said.
At the end of any given day, a conscientious person looks back to assess what went well, and what didn't. And of course, there are excuses and rationalizations. I know I do.
I forgot. Didn't know. Got distracted. Misinterpreted. Presumed. Or, I just wasn't concentrating.
In healthcare, however, we have to find ways to check those errors, and make it easier to prevent than to cause them.