Editor’s note: Eric Thomas, MD, MPH, is the Director of the University of Texas Houston Memorial Hermann Center for Healthcare Quality and Safety and is Associate Dean for Healthcare Quality. We spoke to him about zero harm and patient safety. 

Sarah Mossburg: Could you tell us a little bit about yourself and your current role.

Eric Thomas: I am a Professor of Medicine at the McGovern Medical School at the University of Texas Health Science Center in Houston. I am a general internist and do adult primary care about 20% of my time. I do research on patient safety about half of my time, and the rest of my time is in teaching and leadership of our outpatient quality and safety programs. Most of my professional time has been spent doing patient safety research for about 30 years now, and I have studied a wide variety of topics in the area of patient safety. I am also the Co-Editor-in-Chief of BMJ Quality and Safety.

Sarah Mossburg: Please start our conversation today by explaining the concept of zero harm. What is zero harm when it is used in literature?

Eric Thomas: That is a great question because I think it's not well-defined. There are two definitions of zero harm generally used. Some people use zero harm as a goal, as in, “Our goal as a healthcare organization is to have zero preventable harm.” Other people use it more as an abstract concept, something that we aspire to—zero harm, but not necessarily using it as a measurable, quantifiable goal. Or to be more precise, some people use zero harm as a value, and others use it as a goal. For example, “Do no harm” has been a professional value in the medical profession forever. But a goal is something that is specific, measurable, actionable, and tends to be time-bound. People, including me, sometimes conflate those two ways of thinking about or defining zero harm. This is an important distinction; I personally believe it's not helpful to have zero harm as a goal. Whereas it is, of course, desirable and mandatory to affirm a value not to harm patients.

Sarah Mossburg: What types of harm does “zero harm” apply to, either as a value or a goal?

Eric Thomas: Most people would agree that either as a value or a goal, we're talking about harms that are thought to be preventable. There are certainly harms in a very literal sense that are an intentional part of healthcare. It is harmful to have surgery, it hurts. You could use the term “harm” in that context. Surgery could be considered an intentional harm, but it may be necessary. There are also harms that are unintentional and unpreventable. For example, if I give you a medication that you never had before, and you have an allergic reaction to it, we can call it a harm. It was not intended, like surgery might be. And it was not preventable, because you never used the medication before. And let us say there’s no way to test for it. I think when people talk about zero harm, they are not talking about those things, not talking about intentional harm in the course of appropriate therapy, and not talking about things that are unpreventable, or unavoidable, like an allergic reaction that you couldn’t predict. We are talking about preventable harms. So, that would be a lot of things like surgical site infections, certain medication errors, central line-associated bloodstream infections; there are a variety of preventable harms that we already have the knowledge to prevent.

Sarah Mossburg: The term “zero harm” is commonly used in safety these days. As you have mentioned, it is often espoused in healthcare systems as a value. A colleague recently mentioned to me that they had zero harm pins. You have written about some of the challenges related to a goal of zero harm. What are some of your thoughts on zero harm as a goal?

Eric Thomas: First of all, I'd acknowledge that we are used to seeing stretch goals, or idealized goals. Jim Collins, a popular business book author, identified some companies with goals that he called Big Hairy Audacious Goals. People in the management literature might call them idealized goals or stretch goals. For example, John F. Kennedy said that we're going to send a man to the moon and return him safely before the decade is out. That was an amazing goal. We have this sense that if we set lofty goals, they will inspire us to meet them. I think that is true; if you set a goal that is specific, you can actually measure it, and it's actionable, meaning there is a method to achieve it. But in health care right now I would argue we don't have the means to have zero preventable harm. 

We should be striving toward zero preventable harm, but there's just not the knowledge to prevent all of the preventable harms that we have right now. Healthcare systems are generally very stretched and taxed, and providers are burned out. Healthcare systems do not make millions and millions of dollars in profit that they can reinvest into this kind of a goal like some other private companies can, so it is not clear that we really have the means to do it. The other problem that happens when you have a goal that is hard to meet is that people on the front lines start to game the system. So, if the C-suite says our goal is zero preventable harm and we're going to measure that and hold you accountable for it, then at the unit level, the nurse manager, the medical director, the clinicians will say, “Okay, if you want zero harm, then you're going to get zero harms. We are not going to report the preventable harm, or we're going to redefine the harms in a way to say, well, that wasn't really preventable.” So you get this gaming behavior that is obviously detrimental to learning and to really improving care. There'll be a tendency toward gaming if these goals are associated with rewards and punishments, whether financial incentives or rankings. Daniel Pink wrote a book called Drive that talks about how external rewards and punishments, or even just external rewards, reduce people's intrinsic motivation to do the right thing. To summarize, I would say that goals like zero harm can incentivize people to game the system. They can reduce intrinsic motivation, and it's not clear that we even have the ability to do it right now.

Another problem I have with zero harm is that setting idealized goals is not how organizations become excellent. They become good by focusing on improvement and getting better. In 2016, Richard Bohmer wrote an article for the New England Journal of Medicine about healthcare organizations having success in quality and safety. He says they “constantly make small-scale changes to their structures and processes, over long periods.” And then he remarks, “Major change emerges from aggregation of marginal gains.”¹ So it is these small things that you focus on daily that create excellence and improvement over time, not focusing on this goal of zero.

Sarah Mossburg: Rather than that overarching Big Hairy Audacious Goal, it's better to have marginal improvements that add up over time. And in the long run, these improvements get you there, is that right?

Eric Thomas: Exactly. I can give you a non-healthcare illustration. A couple of the swimmers who competed in the recent Olympic games were coached by Eddie Reese. He has been the coach of the University of Texas men's swim team since 1979, and his teams have won 15 national championships, more than any swim coach ever. They win the most championships, but the goal at the beginning of each year is not to win the championship. The goal is for each person to get better and for them to care about each other as a team. When you have that laser focus on improvement and supporting each other, you end up being the best. It is not by saying, “Oh, we're going to win or else.” 

Sarah Mossburg: Some people who espouse the benefits of idealized goals feel that lofty goals provide inspiration to staff. Are there any situations where clinicians might find that type of goal motivating?

Eric Thomas: In my experience, clinicians see zero harm goals as insincere because they realize that they are not attainable. When they see leadership touting zero harm to the public, they say, “You know, this isn't possible.” From that perspective, it is demoralizing, as opposed to inspiring.

Sarah Mossburg: Like you said at the beginning, zero harm is an important value for health care. How do we clarify for staff that it’s something we are espousing as a value versus as a measurable goal?

Eric Thomas: I think it's what we were talking about before. You work towards it by having measurable goals that you know are taking you towards your value. It's doing things like decreasing the rates of certain events in a clinical area and giving the staff the support to do that. Then celebrating when it happens and spreading that good news around the organization, spreading the best practices throughout the organization. There are some things that we can keep at zero. Those should be measured, monitored, and celebrated. So, you have intensive care units that go months or years without any ventilator-associated pneumonias or have no central line bloodstream infections. I would argue that you might have zero harm goals for specific events, where you have the knowledge and the ability to prevent them. Then that is inspiring to people, to know that they can accomplish zero harm. But it is demoralizing to talk about goals that are not reachable. It is also messy to communicate. It is very hard for leaders to communicate that the goal is zero. “I know that it's not really zero because people are still harmed by unavoidable things” is a complicated message to use for inspiration.

Sarah Mossburg: The proposed CMS Patient Safety Structural Measure includes a specification that the organization's strategic plan publicly shares a commitment to zero preventable harm. Given your thoughts in this conversation, I am wondering what your reaction is to CMS putting a public commitment to zero harm into a proposed quality measure?

Eric Thomas: I would encourage them, for the reasons we discussed, not to have the public zero harm commitment. I have looked at the other structural measures, and I think they are fantastic. I really like the movement towards structural measures as opposed to the discrete measures of outcomes, the traditional kind of quality metrics. I like many of the structural measures, like the one about a commitment to transparency with patients after harm or implementing communication and resolution programs. I think it's good for CMS to try to hold hospitals accountable for some of these things. But when it comes to this kind of zero harm goal, I would just repeat that I think it is not a good thing to do. A better structural measure could direct hospitals to set a goal for year-over-year improvement in the rates of preventable harm. It wouldn't be hard to change that structural measure from zero harm to improvements on reducing preventable harms 

Sarah Mossburg: Are there areas where you're in agreement with someone who holds the opposing viewpoint on zero harm?

Eric Thomas: I’m certainly in agreement that it would be wonderful if zero preventable harm were true, and I look forward to that maybe happening someday. Where I disagree is in the path toward that vision of zero. I would just say it's not by holding organizations accountable for zero as a goal, it's going to be through other types of goals.

Sarah Mossburg: You’ve talked about the importance of incremental goals that hospitals have the means to implement for year-over-year improvement. Are there other strategies to improve safety that you’d advocate?

Eric Thomas: Yes, there is a lot more to improving safety than just goal setting. I would focus on things like teamwork and communication, safety culture, and better engagement with patients, their families and partners in care. That is another important part of this conversation, if you are a leader of an organization, how could you look a family member or a patient in the face and say, “No, we do not have a goal of zero harm.” I mean, that is hard. It is much easier to tell families and the public that our goal is zero harm. It is easier to say that, but as I have told you, I do not think that is how you are really going to improve safety. 

Sarah Mossburg: It is difficult to say to someone, we know that we're going to have a number of harms, and our goal is to have slightly fewer harms year-over-year.

Eric Thomas: You know, in some cases that may be a hard conversation, but in other cases, it's not. Patients understand that healthcare is incredibly complex. It also depends on what you are talking about. This is another problem with zero harm: some things can be zero, and others cannot. There can be a goal of zero for things we know how to prevent, like central line-associated bloodstream infections. There are some processes in health care that we can map out in linear steps. We can make sure these processes happen the right way every time. And we should expect perfection in certain things. 

There are other things we do that are not predictable. I'm a primary care physician. If you come to me tomorrow with a week of abdominal pain, the list of possible causes is extremely long. It may be that we immediately figure it out, and you need to go to the emergency room and have your appendix taken out. Or it is just as likely that we are going to embark on an outpatient diagnostic evaluation. You might go to get blood tests. You might get a CT scan. Your symptoms are going to change in the meantime. We are going to communicate about it. We are going to narrow it down to a few things. We are going to try this or that medicine. You may see a specialist. You may not be able to see the specialist for a week. I am going to try to call and get you in. I mean, this sequence is not predictable. You cannot force that care pathway to happen a certain way every time. And you do not know what is going to happen. We're working together to try to do this safely, quickly, and efficiently, but it's not the kind of thing where you can say, “Zero things will go wrong,” because I don't know how it's going to happen every time. We can identify things that we have to do to try to make complex care safe, and we can know where the process may go wrong. And we can work together with communication and transparency about health information and decision-making. But it is still going to be hard to make that turn out perfectly every time.

Sarah Mossburg: Some proponents for zero harm state that the organizational setting and leadership culture can determine the effect of a zero harm goal. To your mind, is there an evolution in an organization's approach to safety, a place where it becomes appropriate for them to set these Big Hairy Audacious Goals?

Eric Thomas: There could be organizations that have the resources to try realistically to move toward that goal. And they could have leadership that leads in a way that is non-punitive and supportive. Those factors could help them avoid some of the negative unintended consequences of zero harm goals that I mentioned before. I think it is possible. It is rare that we have organizations staffed by leaders that can do that. It requires leadership not just at the top but also at the unit level and throughout an organization to pull that off.

Sarah Mossburg: Thanks for joining us today, Dr. Thomas. We enjoyed the conversation.

Eric Thomas: You are welcome. Thank you for having me. 

1. Bohmer RMJ. The hard work of health care transformation. N Engl J Med. 2016;375(8):709-711. [Available at]