IHI Vice President Carol Haraden and IHI President Emeritus and Senior Fellow Don Berwick recently sat down for a far-ranging discussion about the current state and future of patient safety.
HARADEN: In my recent travels, I’ve witnessed some people suffering what one might call “patient safety fatigue.” What do you think of that concept?
BERWICK: When we started on this safety journey and started becoming more aware of the problems in health care, why would we ever have thought that it would begin and then end?
Jim Reason — who first introduced the Swiss Cheese Model — once said to me that safety is a continually emerging property of a dynamic system. In other words, if you’re committed to safety, you are committed to a never-ending battle to find, mitigate, and intercept risks, and then find new risks and mitigate and intercept those. Anyone that started working on patient safety thinking it was going to start and then be over just had it wrong.
Patient safety fatigue? We’re certainly seeing it. People are saying, “Okay, let’s move on to the next one” or, “I’m done with that.” We don’t have the right to say we’re done or we’re moving on to something else. This is a burden of the pursuit of excellence and the work we will do for the rest of our lives and the lives that follow us.
To me, safety fatigue would be like breathing fatigue or heart beat fatigue. Sorry, guys, it’s the job.
HARADEN: One of the things that we’re seeing as a consequence of what I would call “project saturation” is that people think their care is safe. But when we use things like the IHI Global Trigger Tool and other approaches, we find that the absolute level of safety hasn’t changed. We know individual defects are important to drive out — like pressure ulcers and hospital-acquired infections, for example — but just focusing on the project level work is not changing safety. How does patient safety need to evolve?
BERWICK: I could be wrong, but I think we’re at an inflection point in the maturation of our work on safety. Not to oversimplify it, but you could say the first 15 years of work has been project by project. We find a hazard and mitigate the hazard — pressure ulcers, retained surgical objects, hospital acquired infections, surgical mishaps, etc. We were cutting our teeth. It’s very important work. I mean, if I’m the person you save from a pressure ulcer, God bless you. And let’s not stop it. New hazards will arise as technologies arise.
But now we’re at a new level of maturation: building systems of safety. That’s a different kind of challenge. It means creating an environment, a world, community, organization where the continual pursuit of risk reduction, hazard mitigation, safety in everything is embedded in everything one does. That’s a bigger enterprise. It’s systemic. It’s highly cultural. It has all of the problems we face in project by project safety — such as the need to be transparent about data, having information, and becoming a team — but it’s a bigger idea because it has to infuse our world.
HARADEN: What progress has the US made on addressing patient safety across the care continuum?
BERWICK: We became aware of safety as an issue and work on safety in hospitals, especially in acute components of hospitalization. Why? That’s where the money is. That’s where we had data. That’s where the cycle times are pretty short so you can work on fixing it and see what was working and what’s not. So a conversation about safety a decade ago was a conversation about hospitals.
Now, we have more information thanks to great researchers on safety hazards in other environments — long term care, ambulatory care, laboratories. The agenda has broadened. We are now trying to build a new health care world addressed to patients and families, communities’ needs over time and space, and the continuum of care. Whether it’s through the journey of chronic illness, whether it’s for episodes of need that are going to last 90 days, or whether it’s entire life spans, we need to work on safety there.
Why? Because we want to be excellent. So we need to be reliable, patent-centered, timely, respectful of diversity, efficient, reduce waste, and safe throughout the care continuum.
Are we? Really slowly. I can count on one hand maybe the number of truly ambulatory environments where safety is now a major concern.
HARADEN: What do you see as the biggest challenges in working with electronic health records (EHRs) today?
BERWICK: The EHR story is a good news, bad news story if I ever heard one. The good news is at last we live in a digital age. Information should be easily gathered, storable, retrievable, and accessible. We’re making progress. The majority of hospitals use them now. The majority of physicians are starting to.
Somewhere along the way, though, we lost something. There’s too much automation of the status quo approach and — even worse — the electronic record is sometimes being used as a component in the games we play around billing and compliance. I’m not sure if electronic health records have made care safer. For example, the way we’re cutting and pasting progress notes is an invitation to hazard.
We’re in a very important transitional phase. If we’re smart, if the electronic health record becomes an integrated, team-based, very patient-controlled, highly transparent tool for new care delivery, then it will become an asset instead of a liability. Right now, however, we don’t yet appear to have the political will to force the patient record to be all that it could be.
HARADEN: One of the things many of us [in the safety world] have talked about for a long time is trying to think upstream [about harm] versus always being reactive. Mitigation will always be a part of patient safety, but thinking about prevention — or safety prognosis — as a primary strategy, hasn’t taken hold.
BERWICK: I think we’ve approached safety largely as repair. You know, something went wrong, how can we mitigate it or prevent it next time? Mature approaches to safety don’t wait for the trouble and then fix it. They anticipate the trouble and the trouble never happens.
On the other hand, when we try to be anticipatory, we sometimes get stupid. We imagine everything that can happen and we mitigate everything that could happen and that’s not smart either. Indeed, you can completely put us in handcuffs with restrictions that come under the guise of making things safer, making work much harder, and when work gets harder guess what happens to safety: it gets worse.
What we really need could be called informed anticipation. There are tools for that — failure modes and effects analysis (FMEA), for example. We still don’t use it very much in health care, but it can help a lot.
HARADEN: Many organizations just rely on root cause analysis.
BERWICK: I totally understand where the focus on root cause analysis comes from: something bad happens, find out what caused it, and stop it. I think the primary focus on root cause analysis has on thinking, however, can sometimes be harmful. When you really study safety, you realize that a lot of properties of complex systems — the causal system for hazard — is a network of causes. It’s not one thing. Standing at the end of the event looking back up the retrospection scope, you can say, “Oh, the last thing that happened was the nurse was sleepy. Oh, let’s not have sleepy nurses.” But the thing that made something really untoward happen isn’t a thing, it’s a network of causes. Preventing harm requires thinking about very complex interactions and the hazards they create.
The other piece of prevention we need to build is a psychological issue: we need to notice and create an environment in which people can share what they notice. In the TSA world they say, “If you see something, say something.” Patient safety should be like that.