Dr Robert Wachter's opening remarks

I first became interested in patient safety in 1994, when I read Lucian Leape's paper ‘Error in medicine’.2 It introduced to me the concept that errors are mostly manifestations of dysfunctional systems, and most errors are committed by good, competent, caring people trying to do the right thing. That actually made a good deal of sense to me. I went on to read some of the foundational works in patient safety, notably a book by Jim Reason.3

Reason studied how errors happen in complex organisations; he studied space shuttle crashes, train derailments and Three Mile Island. When I first read about the Swiss Cheese Model,3 it was an epiphany to me—this idea that in complex organisations the root cause of most errors is that the protections the organisation has built are leaky, and on a terrible day, all the holes align and the error makes it through.

I work at UCSF Medical Center, and my colleagues—doctors, nurses, pharmacists—are spectacularly good. They are careful, they care deeply and they work hard—and yet we have errors frequently. So reading Reason's work made very real to me this idea that errors can't be all about bad people. It's mostly about good people working in dysfunctional systems.

The other reason the ‘No blame’ and ‘It's the system’ approach was so important is because in the early years of patient safety, in the late 1990s, it was vital to get doctors engaged in this work. At that time, if you said to a doctor, “We're going to talk about medical mistakes,” all they could think about was malpractice, and from then on, they were not listening. So the idea of focusing on no blame, it's the system, was an incredibly important point—both because it's mostly correct and because it was a politically astute way of getting started.

Several years ago, however, I began to recognise that we were invoking the idea of ‘No blame’ and ‘It's the system’ in almost every situation of harm or risky behaviour. One example is in the area of hand hygiene. In visits to many hospitals, I began to ask about hand hygiene rates. At those with less-than-stellar rates, I'd ask, “What are you trying to do about that?” And the response would be, “Well, we're trying to improve the system.” But as I looked around, it seemed that the system was pretty good—with gel dispensers everywhere, posters of clinical leaders washing their hands and computer screen savers that reinforced the message. I began to realise that the nature of the problem is a lack of accountability.4

A few years ago, I had the opportunity of giving the James Reason Lecture in the UK. As I was preparing the talk, I went back and re-read the book Managing the Risks of Organisational Accidents, Reason's foundational work from 1997, and I asked myself if the founder of the Swiss Cheese Model ever said anything about how to deal with transgressions. And in fact he had. Reason wrote, A no-blame culture is neither feasible nor desirable. A small proportion of human unsafe acts are egregious and warrant sanctions….severe ones in some cases. A blanket amnesty on all unsafe acts would lack credibility in the eyes of the workforce. More importantly, it would be seen to oppose natural justice. What is needed is a just culture, an atmosphere of trust in which people are encouraged, even rewarded, for providing essential safety-related information—but in which they are also clear about where the line must be drawn between acceptable and unacceptable behavior.3

When I asked hospital leaders what they do about people who refuse to follow reasonable safety rules, the reply was often “We have a ‘just culture’.” They would show me Reason's algorithm3 or David Marx's algorithm5 and it made pretty clear what you do in different situations. But, I asked, what if the head of CT surgery or neurosurgery says, “I don't like the idea of a time out, so I don't do that”? That question was answered with silence.

In fact, many organisations are not doing very much about that kind of a situation because it's really hard to address. But by not addressing it, we all become enablers of allowing people, and in some cases only certain people, to break key safety rules. The message it sends to an organisation and the entire field is quite profound.

Henry Kissinger once said, “Weakness is provocative,”6 and he was right. No blame is usually the correct response to most errors and unsafe acts. But by invoking it all the time, sometimes inappropriately, I think we are harming more patients and, quite likely, we are losing the trust and credibility of those who look to us for leadership in patient safety. Leaders who do nothing about clinicians who fail to clean hands, clinicians who bypass time-outs, or clinicians who engage in disruptive behaviour are being provocative—and it is time we stopped.

Dr Gregg Meyer's opening remarks

The resolution should be denied because it could very well be a danger to patient safety. The demand for accountability ignores our humanity. It runs counter to what we know from safety science. It's impractical, it does not work, it drives the wrong behaviours and it's a distraction.

To the first point: it ignores our humanity. We must recognise our humanity when it comes to making mistakes, to forgetting to do that hand hygiene, to skipping through that one part of the checklist on a particular morning. It is part of our human nature to err. We make mistakes, and when we are fearful, when we are worried that somebody is watching us and is going to punish us, we let that fear push us further over on the performance anxiety curve, and we make even more mistakes.

The proposition also ignores safety science. Advancing patient safety requires continuous learning. Being open about our failings is the first and most important step to getting that process going. Blame will create distrust and it will slow down the learning of the safety movement.

H.L. Mencken said, “For every complex problem, there's an answer that is clear, simple, and wrong”, and accountability is clear and simple and wrong. Patient safety is complex, and we need to treat it as such.

The proposition would not work. We've had decades of torts that were based on blame. Has the medical malpractice system made us safer over the last five or six decades?

We have evidence of how wrong things can go if we assign blame. We all celebrate our aviation safety reporting system. It's been emulated around the world, and the CALLBACK newsletter is something we all aspire to in terms of the way they share openly about those hazardous events. New Zealand was one of the countries that emulated it. They set up a system that was working well. But in 1991 they decided to divulge the identity of one of the reporters.7 In so doing, they shut the system down. No one reported anymore. Everyone was afraid. I would argue that the value of divulging the name of that pilot was not worth the 5 years it took to restart their safety reporting system.

The proposition is also impractical. Whenever we create such rules, there are three ways to approach them. The first way is to change the underlying process, to do it right, to make doing the right thing the easy thing. That takes time, energy and resources. But there are two other ways that people ensure they are complying with whatever rules have been set. The first is to game it, to check the box—perhaps without the seriousness of effort expected. The second is to create unsustainable efforts, to throw resources at the problem that we cannot keep there for the long run. Both of those are shortcuts, and driving to accountability will make us take those shortcuts.

It is also important to ask, “Where does it stop?” The accountabilities that Dr Wachter has proposed, such as hand hygiene and the preoperative checklist, seem quite defensible places where we should start with accountability, but what comes next and where does it stop?

For example, at my institution, a patient who had undergone surgery had a postoperative event that led to a significant bleed. The requirement for any blood transfusion was that two nurses had to sign off on the unit of blood with their full names and titles. When we had an accreditor come in for review, they found that we violated that policy 40 times in one particular case. There were just initials on the units. That was deliberate and it was repeated. The patient survived because those 40 times were for a single patient who was having a major bleeding event.

If those nurses were going to be held accountable for the fact that they did not sign their full names, would they have been as responsive in getting those units of blood into that patient as quickly as they needed to? There are going to be emergent exceptions to any of our rules, but where do you draw the line?

We need to think, and the thoughtful recognise that punishment is not the route to improve safety. It drives the wrong behaviours. Rules-based behaviours are one of the biggest sources of safety events in recent analysis, and what are we driving toward with this issue of accountability?8

We need to keep focusing on harm, but it is easier to punish than it is to put resources into making sure that we create a less harmful environment for our patients. When times are tight, reverting to punishment as a quick, expedient means of ensuring that we are moving forward on patient safety has a lot of attraction for our colleagues because it is a lot easier than educating, building systems and removing barriers.

Finally, the proposition is a distraction from the real barriers to patient safety. The real barriers aren't bad people. They are production pressure and they are the lack of systems. We need to spend our time on improving those rather than on counting checklists, policing and being Big Brother.

Lucian Leape said that the single greatest impediment to error prevention is that “we punish people for making mistakes”.9 It was right when he said it, and it is right now.

Questions and discussion

A poll of the Congress attendees after the opening arguments showed a slight move in opinion toward Dr Meyer's position (see figure 2). Yet the vast majority continued to agree with Dr Wachter's argument. In questions from the attendees, as well as direct point–counterpoint discussion between the debaters, the grey areas of each side emerged.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Audience poll after opening arguments.

Closing remarks: the future of patient safety

Ultimately, this debate was about how best to advance the field and continue to make healthcare safer. Dr Wachter said it is the job of leaders to figure out how best to deploy resources. Most organisations are already expending resources to monitor compliance; what is lacking is the feedback loop, letting people know that they are doing something wrong, giving them the opportunity to improve, and holding them accountable if they fail to do so. He suggested that the healthcare industry is capable of invoking these sorts of standards in appropriate ways and when the circumstances are right. He reiterated his belief that invoking no blame had been a good way to start our work in patient safety, but that today, as our field matures, there needs to be a balance of no blame and accountability.

Progress in patient safety has been slower than anyone would like, but Dr Meyer pointed out the need to focus on how far we have come. Bringing blame into the discussion, he said, would decelerate that progress, not move it forward. We still have much systems work to do. Moreover, a move to increase accountability would require additional resources to police policies and processes. A better use of scarce resources in healthcare would be to improve underlying systems, do root cause analyses, and look for factors that contribute to errors and safety lapses.

In conclusion, Dr Gandhi asked attendees to vote for a final time on the resolution, and a decided majority voted in favour of accountability (see figure 3).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Audience poll at the conclusion of the session.

This audience of health professionals and patient advocates clearly agreed there must be a move toward increased accountability for patient safety. This debate highlighted the benefits of imposing a system of accountability, as well as the challenges and risks if such a system is not done thoughtfully and with fairness. Going forward, the patient safety field will need to study and learn the best methods for implementing appropriate accountability to drive safety improvements while continuing to build a safety culture and encourage transparency about errors.