In Conversation With... Pascale Carayon, PhD and Nicole Werner, PhD
Editor’s note: Dr. Pascale Carayon, PhD, is a professor emerita in the Department of Industrial and Systems Engineering and the founding director of the Wisconsin Institute for Healthcare Systems Engineering (WIHSE). Dr. Nicole Werner, PhD, is an associate professor in the Department of Health and Wellness Design at the Indiana University School of Public Health-Bloomington. We spoke with both of them about the role of human factors engineering has in improving healthcare delivery and its role in patient safety.
Sarah Mossburg: It is nice to have both of you here with us today. I think the best place to start would be if you would tell us a little bit about yourselves and your current roles.
Pascale Carayon: I’m a professor in the Department of Industrial and Systems Engineering at the University of Wisconsin in Madison, and by training I’m an engineer. I’ve got my engineering degree from France and a PhD in industrial engineering from the University of Wisconsin in Madison, and my expertise has been in human factors engineering for my entire career. For more than 20 years, I’ve worked on a range of projects in healthcare and patient safety, primarily on the healthcare delivery side. My focus has been in primary care, the hospital, ICUs [intensive care units], ED [emergency department], care coordination, looking at healthcare professionals, and over the years, more looking at the patient and care partner side.
Nicole Werner: I’m an associate professor in the Department of Health and Wellness Design at Indiana University and School of Public Health. Before that, I was at the University of Wisconsin-Madison, in the Department of Industrial and Systems Engineering. I’m a human factors engineer, and I received my doctoral degree in applied cognitive psychology. I have been really passionate about applying human factors to improve healthcare quality and safety since I was an undergraduate. While I was an intern at Johns Hopkins Hospital and worked with the human factors engineer, I realized that what I was learning in the classroom about human factors engineering and cognitive psychology could be applied to actually save people’s lives.
Sarah Mossburg: You both mentioned human factors engineering as part of your expertise. Could you give us a brief explanation of what that is, for readers who are less familiar with the topic?
Pascale Carayon: So, we use to term human factors engineering, or HFE. In our research the “E” stands for engineering, but the “E” sometimes is used for ergonomics. Different countries have different traditions about the terms that they use, but the International Ergonomics Association has stated many years ago that it was the same discipline. In HFE, we engineer systems so that people can do their job in order to improve performance, but also in order to also improve safety and well-being. We have kind of that dual objective of a typical industrial engineering objective to improve quality and productivity, efficiency, and so on, but also an objective that you would find in other disciplines, such as psychology, sociology, HR [human resources], to improve conditions for people. We are a unique discipline that sits at the intersection of many disciplines. It includes people like Nicole and I, who come from an industrial engineering background, and the discipline of psychology, cognitive psychology, and industrial and organizational psychology.
We are interested in all kinds of issues about how you design a technology. So, when you hear about usability of an electronic health record or EHR, that’s us. We’re also interested in designing processes and the physical environment. Anything that people touch on or interact with. And so, a core concept is really understanding people and the rest of the system and the interactions, all the tasks, all the activities, that people engage with; working in a certain physical environment, in an organization, a team, and so on.
Sarah Mossburg: Could you give us some examples of the application of HFE in healthcare settings?
Pascale Carayon: I mentioned technology, usability of health IT, of the EHR, of computerized provider order entry, and of medical devices. Those could have huge patient safety implications. Human factors engineers can identify problems with technology design, as well as provide solutions for how to improve those technologies. Issues of the physical environment are something we don’t often think about. For instance, figuring out how we design a physical environment, meaning space, lighting, noise, distractions, location, and the physical workflow.
Nicole Werner: A few others that might be worth mentioning are teamwork, teams, and team communication. There’s also a lot of work that’s been done on alarms and alerts, and responses to those, and the environmental design. You can start to see we’re really all about the interactions, and I think that’s the unique aspect of HFE.
Sarah Mossburg: So, HFE has broad patient safety applications, in terms of device safety, physical design to prevent errors related to distractions, lighting, teamwork and communication, alarms, and alerts. It sounds like it touches on pretty much every aspect of patient safety we could think of at this moment.
Pascale Carayon: Absolutely, and you know that’s actually a great comment, Sarah, because you know, in patient safety, we very often think about, okay, what are you doing in medication safety? What are you doing about diagnostic safety? What are you doing with infection prevention? What are you doing with improving maternal health and babies? We slice patient safety into domains, and from the perspective of Nicole and I, we scratch our head, and we say, “Well, we’re improving medication safety sometimes, sometimes we’re improving team communication, because that may help with surgical safety.” We’ve also done a lot of work in infection prevention. The specific patient safety domains, the way they are cataloged or categorized, doesn’t really matter to us because very often the way the systems are designed contribute either in a positive or negative manner to those multiple patient safety domains because there are always people in the systems.
Sarah Mossburg: This is a good point to transition into talking about the framework that you developed called the Systems Engineering Initiative for Patient Safety, or SEIPS. You first published that model in 2006, expanded it to 2.0, and then just recently, in 2020, you published SEIPS 3.0. I was hoping you could walk us briefly through the SEIPS 3.0 model and the key differences from the original SEIPS model.
Pascale Carayon: First of all, we think about them as a family, and families have members. We don’t think about the first version, and then the second version being better than the first one, and number three being better than the second one. We also have the SEIPS 101 version that’s more for dissemination. We think about them as a group of different models that have core elements, but each one of them has a slightly different emphasis.
The SEIPS 3.0 model really focuses on the patient and understanding the care process from the patient’s perspective and emphasizing the need to look at care transitions: any time a patient transitions from one broad environment of care setting to another. We know that transitions are vulnerable times that need a lot of attention. That’s really the emphasis of SEIPS 3.0. But all the SEIPS models include an understanding that what people do is embedded in a system, and so people are in the system. They do tasks and activities, they use different tools and technologies, and all of that happens in the physical environment and under certain organizational conditions.
The people in the system are physicians, nurses, patients, and care partners. Sometimes they are single individuals, sometimes they are teams. Those elements of the system, the people, the task, the physical environment, and the organizational context, all interact with each other. It is very difficult to just look at the interface of CPOE [computerized provider order entry], or the medication administration, or whatever, without really thinking of the rest of this system. The dual objective that I was talking about earlier in the definition of HFE is part of our SEIPS models of improving things for patients, of course, patient safety, but also improving things for organizations and the healthcare workers.
The last important element of SEIPS is feedback. Designing good systems is really difficult, and we might not get it right the first time. Therefore, learning and feedback, or continuous improvement is important.
Sarah Mossburg: I wonder if you can help me understand the difference between human factors engineering versus systems engineering?
Pascale Carayon: Our unique perspective is to understand the people in relation to the systems. So that’s why you have someone like Nicole who has a psychology background, and someone like me who’s an engineer, and we spent a lot of time taking classes in psychology and sociology because we need to understand people and improve things for people. We are just a different flavor of systems engineering because we focus our attention, our analysis, our design, and our improvement activity on people.
Sarah Mossburg: I understand that you both have been working in a patient safety learning lab that was funded by AHRQ that uses HFE and the SEIPS 3.0 model to look at the patient journey, particularly transitions in care. Could you tell us a little bit more about the patient safety learning lab itself, starting with what is the patient safety learning lab, who is part of it, and what you do as part of that process?
Nicole Werner: Our patient safety learning lab is really about using human factors engineering, or systems engineering for people, as Pascale just described, to transform how we think about older adults’ care transitions from that point of vulnerability to a point where risk is reduced. Transitions can be a checkpoint to identify and even remediate potential risks moving forward using this intervention process, which we’ve named the patient safety passport. A key part of our patient safety learning lab is based on the SEIPS 3.0 idea of the patient journey. It’s not an episode of care; it’s part of that older adult patient’s journey where they might be experiencing many transitions over time. We’re thinking about this time point as an opportunity to actually improve safety rather than thinking of it as a point where they’re going to get harmed. In our project, we’re using the emergency department as that hub for the multiple care transitions that are happening to and from the emergency department. Our team includes human factors engineers, faculty scientists, students, ED physicians and nurses, hospitalists, pharmacists, health system leadership, the health system IT group, and a family partner.
Our patient safety learning lab started off with this idea of understanding the system, doing a work system analysis, where we used different methods, like observation, and contextual inquiry, to really get an in-depth understanding of the system and the processes, mapping those processes, and determining the barriers and facilitators in that system that are influencing those processes. But we’re also using quantitative data from the electronic health record to model the processes and their associated outcomes. We’re using both of those approaches to get that deep understanding of the system.
We used that analysis as a foundation to guide the design of our patient safety passport, which is a participatory design approach and based on those in-depth work system analyses, we decided to focus on two specific transitions. One is the emergency department to the skilled nursing facility. When an older adult is already in a skilled nursing facility, they go to the emergency department, and they go back to that skilled nursing facility. The other is discharge to home from the emergency department.
Pascale Carayon: The decision was made based on including a lot of stakeholders, and we had a lot of meetings with our multidisciplinary teams, and then looking at data. The transition that I looked at is the one where an older adult goes from the ED to home and the discharge process. We redesigned that care transition with multiple stakeholders. Our participatory design process happened right at the beginning of COVID, so we moved a lot of things to the virtual environment. The core team that participated in the design process included an ED physician, an ED nurse, sometimes we called on other ED physicians and nurses, people from IT [information technology], our family care partner, and then a number of human factors engineering students and researchers.
We looked at that care transition as one that actually begins early during the ED visit, and so we implemented an intervention that improved the way physicians enter instructions that then show up in the discharge summary. We did a lot of work on a pretty big redesign of the discharge instructions document, working with our IT colleagues to figure out what we could do, given the constraints of the EHR. We also worked a lot with our ED nursing colleagues to improve communication at the time of discharge. Everybody wants to go really quickly; the patient and the care partners want to leave the ED, and physicians and nurses are extremely busy. And then, we supported the nurse when they do their follow-up call about 24 hours later. We really figured out how we could get all of these different perspectives, different opinions, different viewpoints together to improve a process that had many different components. The multidisciplinary participatory process is great; it’s figuring out how you manage to get to an end product that is satisfactory and has a good impact.
Our intervention was implemented more than a year ago. We used all kinds of different methods to evaluate it, including interviewing patients and nurses, collecting data from physicians, and doing a survey of patients.
Sarah Mossburg: And do you have preliminary results around how the intervention was received?
Pascale Carayon: The feedback is pretty positive. As you can imagine, there are pluses and minuses, and that’s also what we look for in our evaluation. Again, thinking about the feedback loop. We did typical human factors analysis to redesign the discharge instructions, and used what we call usability heuristics, or good design principles, for improving the documents. For instance, moving things around, changing the font size, bolding things, and trying to eliminate a few sections. The discharge instructions summary is a lot shorter than it used to be, and it’s better organized. We’re getting good feedback from our patients on that, but we are also hearing that there is still a lot of information. Well, we wanted to exclude some information, but we could not do it. The design process is a continuous improvement process.
Sarah Mossburg: Dr. Werner, I think you mentioned skilled nursing facility transitions as the other transition that you looked at a little bit more in depth. I wonder if you could speak to that.
Nicole Werner: Sure. We brought in key stakeholders, including skilled nursing facility practitioners, health IT folks from the health system, nursing, and ED clinicians. We wanted to include people that are tasked with improving care transitions from the emergency department to the skilled nursing facility. We ended up focusing on a specific type of transition where older adults come to the emergency department from a skilled nursing facility with a suspected urinary tract infection. There were a couple of different reasons for focusing on that particular situation. The first is that it’s a really important transition in terms of diagnostic accuracy and antibiotic stewardship, and lots of different patient safety issues can happen as a result. Second, there had already been some decisions in the healthcare system to use an intervention to improve the emergency department to skilled nursing facility transition processes. We tried to focus on something that was really important from a safety perspective and would complement what they had already been working on.
We were specifically engaging with the advanced practice providers who were overseeing the older adult in the skilled nursing facility, who would be able to make prescribing decisions for those older adults. We wanted to create an automated communication process because we found that there were gaps in communication between the emergency department and the skilled nursing facility. At each stage of that transition, emergency department and skilled nursing facility staff had some disconnected and conflicting ideas of what needed to happen in terms of communication and expectations. To address some of those gaps, we created a trigger in the electronic health record. Anytime an older adult came to the emergency department from certain skilled nursing facilities that were overseen by these advanced practice providers, an automated trigger set forth a series of interventions to improve communication and coordination.
One of the challenges was that there was not an easy way to identify which patients this intervention applied to, but that was actually an innovative part of the project, and that I’m personally very satisfied with, because I think it solves a huge challenge for the healthcare system, too. We were able to create this new process within the admission process and the electronic health record where these patients were able to be identified to trigger that series of interventions.
Sarah Mossburg: That leads me into the next thing that I would love to touch on, which is about why patient experiences and feedback are important in transitions of care, and how they’re being incorporated into some of the work that you’re doing.
Nicole Werner: Well, currently in our healthcare system, I would say, the only constant in a transition of care is the patient and their family. We rely on them to be the linchpin in these transitions, to have the knowledge of where they were, what was happening, their medications, etc. And I think there’s been a historical tendency to think of a transition as a short time period like a discharge. A care transition, from the ED or from any other healthcare setting, back to home or to another setting, is not one moment in time. And what we found in our research on older adults transitioning from lots of different settings is this is really a longitudinal process for them, and from their perspective it’s much longer than it’s typically been conceptualized by healthcare professionals. They’re often given a lot of instructions and told to do things, but the implementation of that by the patient and their family isn’t always possible, given their circumstances, or they don’t understand it, or they don’t know the next steps to take. I think that it’s not just important, but it’s essential to have the patient and family perspective in this type of work because of that longitudinal nature and because our health system is currently designed to rely on them for information.
Pascale Carayon: From a human factors perspective, what Nicole is saying is that we need to understand the patient experience, and to us that translated to understanding the whole system for the patient. A lot of that has been done in other care settings, but the ED is really tough because everybody is busy, and no one wants to be in the ED, and the goal is to get patients out of the ED. So, from a research perspective that adds a lot of challenges. Very early on in our project, we wanted to understand what it’s like to be a patient in the ED. We had a couple of our students who were research assistants spent about 200 hours in the ED pre-COVID following patients and care partners during their entire stay in the ED. We documented a lot of the activities that patients and care partners engaged in, and that’s really something that no one had done before. It helped us to understand the role of patients and care partners as knowledge workers, and emphasized that we need to figure out how to better engage patients and care partners during that ED encounter and on the broader patient journey.
Sarah Mossburg: That’s interesting to note that patients and care partners are the one constant in that journey, and especially at those transition points, they have critical information that the healthcare team can use to provide better care for patients. You both mentioned challenges related to implementation of the projects. Dr. Carayon, you spoke of bringing together lots of different people, and having a lot of discussions and challenges coming to agreements on potentially what and how to intervene. Similarly, Dr. Werner, you spoke about engaging with the dynamic systems that are already looking at some of these issues and wanting to complement the work being done rather than disrupting existing quality improvement initiatives. What other key challenges did you experience as part of this process?
Pascale Carayon: So of course, things changed during COVID. For a couple of months, we could not collect data, and then patients had no care partners with them. There were challenges related to us having to move a design process from a physical room to the virtual environment. In any project, there are going to be surprises. We don’t know what they are, but we know that stuff is going to happen.
Nicole Werner: We’re connected to the system, connected to all the different possible people. That’s our work as human factors researchers because the systems that we have to adapt are dynamic and open. In that instance of the ED to skilled nursing facility transition that we focused on we really connected to them. So, we know that they have this other intervention, and we can engage the person who’s leading that intervention and bring them onto our design team to make sure that we’re not overstepping. We’re not impeding on what they’re trying to do.
Sarah Mossburg: Are the interventions from the patient safety lab being more broadly implemented or potentially replicated?
Pascale Carayon: I can speak about the one we did on the extended discharge process. That whole design process was focused on the particular patient population of older adults who come to the ED with a fall. The entire process has now been implemented for every patient in all of the EDs of that healthcare system. Whether it’s going to get implemented in other systems I don’t know. I mean, this is an interesting question. Hopefully through what we do and what we provide to them, we teach them a lot of things, and you know some of the people we’ve worked with are in other organizations, and then they take those skills and methods with them. A lot of it is probably going to be also diffused through other mechanisms, such as publications, presentations, our students getting hired, and so on.
Nicole Werner: I think two of the emergency department physician faculty that we work with now have funding from AHRQ. At least one uses similar methods to our work system analysis and a participatory design approach to carry forward these methods in their own. I really love seeing that people that we work with are carrying forward those methods.
Sarah Mossburg: That’s wonderful. Thank you both so much. This has been really interesting conversation; it has been really interesting to learn about your work.