The Evolution of Root Cause Analysis

Editor’s note: Jessica Behrhorst, MPH, CPPS, CPHRM, CPHQ, is a high reliability subject matter expert at Aptive Resources and a faculty member at the Institute for Healthcare Improvement (IHI) working on educational programs, including Root Cause Analysis and Action – or RCA Squared (RCA2).

Sarah Mossburg: Welcome, Jessica. Please tell us a little about yourself and your roles and describe your interest in root cause analysis (RCA).

Jessica Behrhorst: I am currently working as a high reliability subject matter expert at Aptive Resources, specifically on High Reliability Organization transformation with the Veterans Health Administration. I am also currently a faculty member at the Institute for Healthcare Improvement (IHI), working on educational programs, including Root Cause Analysis and Action – or RCA Squared (RCA²) –an online self-paced course IHI offers, as well as the Patient Safety Executive Development Program. Before joining Aptive Resources, I was a Senior Director for Patient Safety at IHI. I also was formerly the Assistant Vice President of Quality and Patient Safety at Ochsner Health System in New Orleans. 

I became involved in RCA² as part of my role at Ochsner, a health system that was quickly expanding while I was there. We onboarded many facilities, bringing them into our policies and practices. Part of that was how we did root cause analysis and other regulatory functions. Our team began to think about how we were impacting the organization’s culture and how to build a culture of safety to promote the safety practices we were trying to implement. Ochsner’s Chief Quality Officer learned about RCA² during a conference and was so impacted that we soon began implementing it. He believed it could improve how we look at events and support our culture more broadly.

We invited Dr. Jim Bagian, one of the authors ofRCA²: Improving Root Causes Analyses and Actions to Prevent Harm, to train leadership and patient safety teams to understand the nuts and bolts of RCA². After that, I helped lead the effort to implement RCA². It became a passion for me – thinking about how we look at events and improving the way we handle them. It is not just an event that happens and goes on a shelf. When I joined IHI full-time, I had an opportunity to do more work on RCA², creating their course and connecting with national patient safety leaders. Working with organizations trying to make this shift in how we look at events has been great.

Sarah Mossburg: That is helpful context. While I am sure most of our readers will be familiar with it, to level set for this discussion, could you describe root cause analysis?

Jessica Behrhorst: Root cause analysis is an event investigation process that is not unique to health care, it is also used in other industries. When a serious adverse event occurs, RCA is a way for safety teams—in our case, patient safety teams—to investigate why it happened. We are trying to get to the root cause of the issue and create actions to prevent it from happening again.

In health care, RCA is a regulatory process. Accreditation requires that we conduct root cause analysis on sentinel events, which are serious events that cause harm or death to patients or staff. These events require an RCA to ensure we create actions based on the event.

Sarah Mossburg: Are there other ways organizations have used root cause analysis historically?

Jessica Behrhorst: In health care, RCA was often something we did to “check the box” when we had a serious event. Some organizations were thoughtful about it and used the RCA process as an improvement opportunity, but that was not always the norm. It tended to be tied to regulatory functions, which we often think of as quality assurance tasks we must do to function.

The processes could be secretive and confidential because we are dealing with sensitive events, like causing harm or death. They happened behind closed doors, and we did not talk about them much. To frontline staff, it could seem like a black box. They saw that a group of people got together to discuss what happened, but then frontline staff would not know what happened after that. That was very much the perception of an RCA – just something we had to do when we had a serious adverse event.

Sarah Mossburg: How has the RCA process changed over time?

Jessica Behrhorst: Over the last 15 or 20 years in health care, there has been a much bigger focus on how we think about patient safety culture. When we think about tools for patient safety, we should consider how we are improving the safety culture. People have started to think about root cause analysis and other patient safety tools along these lines. We want to use RCA as an opportunity to identify things that are wrong and think deeply about how we create change that improves our systems.

Sarah Mossburg: As an IHI faculty member, you provide training on RCA². Can you describe what RCA² is?

Jessica Behrhorst: RCA² is not just root cause analysis but root cause analysis and action. It is RCA² because the focus is on actions as well. The team that wrote the RCA² white paper identified some issues with the traditional RCA process. They felt that traditional RCAs did not identify the root causes of events or creating sustainable actions around those root causes. They emphasized action planning, making sure that patient safety teams are working on strong actions. 

In my experience with RCAs, our actions often included “educating staff” or “creating new policies.” We learned from other organizations that those are not sustainable actions. Sometimes those actions are needed, but in terms of the root cause analysis and action, we wanted something stronger. In RCA², a hierarchy categorizes actions as strong, intermediate, or weak. Education, checklists, and new policies – those types of things are not strong actions. When thinking about strong actions, we need to ask, “How do we develop strong actions that create a sustainable process and prevent recurrence?”

Where traditional RCAs tend to be a very closed process, RCA² democratizes the process so that more people are involved. We want it to be a simple process that people understand. There are three basic questions we use all the time: “What happened?” “Why did it happen?” “What are we doing to keep it from happening again?” RCAs do not have to be complex. In the past, regulatory agencies created lengthy forms that asked standardized questions. Although they appeared very thorough, sometimes the questions did not address the complexities of the event we were investigating. RCA² uses the “Five Whys” model to simplify and improve that.¹ It is not about answering 30 questions about an event – it is about digging down into the causal chain of why this event happened.

Another difference between traditional RCA and RCA² is how analysis teams are structured. In traditional RCA, there were large teams. In RCA², we have small working teams without the staff involved in the adverse event. Those staff members are interviewed to get information, but we do not want them as a part of that small team that will look at why this happened. This is done to support culture and psychological safety. Some organizations need help with this piece because they are accustomed to having the staff involved in the event on their teams. While I think it is important to include those staff members, we should consider the effect it will have on our overall psychological safety and culture. Some of these events can be very traumatic, and we need to support the staff who have been involved. Removing them from the RCA² team helps us think about causality in a very objective way. 

Sarah Mossburg: Earlier you talked about an action hierarchy and mentioned some weaker examples. Could you share some examples of stronger or sustainable actions?

Jessica Behrhorst: Strong actions tend to include human factors considerations, adding things like forcing functions or those types of changes that force us to make the right decision and take some of the human decision-making out of it. A non-healthcare example is a gas pump. If you pull up to a gas station with diesel and gasoline pumps, you cannot put the diesel nozzle into your gasoline car because of how it is designed. It is a forcing function. In health care, we can think about the design of the processes that our staff use daily and how we can change them to ensure that they are not making wrong decisions. We have a lot of distractions in our environment and should consider how we can help staff make the right decisions.

Sarah Mossburg: How much of a role does safety culture have in successful implementations of RCA²?

Jessica Behrhorst: It is a critical role, and it is symbiotic. To have a strong safety culture, we must have patient safety tools and processes that support that open culture, that culture of being able to speak up, that non-punitive culture where we are focused on improvement, not on blame. So, as we think about each of our safety processes—a communication and resolutions program, root cause analysis in action, or peer support for the caregiver programs—all of those programs have to work to support this transformational change. At the same time, if we have leaders who are punitive in their response to events, we are not going to see that change in the culture. They must support each other to change the culture and make each program work successfully. 

Sarah Mossburg: Are there specific aspects of safety culture that need to be in place before RCA² implementation?

Jessica Behrhorst: They do not necessarily have to be in place before, but they must be happening at least at the same time. If we do not have them at all, it is hard for it to be successful. Do we have an organization where staff feel comfortable reporting? How are we looking at human error versus reckless behavior (i.e., just culture)? Do we have a learning organization where we are thinking about improvement? Those things are helpful to have in place or to implement at the same time as RCA^2. If we are doing an RCA² in an organization without psychological safety, I guarantee that many leaders will want to be involved because they want to know who is to blame. At the same time, when we go out and interview staff in those organizations, they are not going to tell us what happened. It affects the whole process if we do not have psychological safety. It's the same thing if we do not have just culture. Staff may think that an RCA will get them fired when all they did was commit a human error. And related to being a learning organization, we want to make sure that we are putting that quality hat on at the end of this process and building sustainable improvement. If that is not part of this organization, we could put that RCA back on the shelf and not look at it again.

Sarah Mossburg: How have organizations implemented and used the RCA² methodology since its introduction?

Jessica Behrhorst: A lot of organizations have moved to RCA² because they have seen the negative effects of the traditional RCA process on their culture, and they want to do it in a different way. As an example, a traditional RCA would often feature large meetings with leadership, the staff involved in the adverse event, and patient safety and risk managers. They would discuss in front of the whole group, “Why did you make this bad decision?” It created a very non-psychologically safe environment and felt punitive to staff who had been involved in the event. As a result, staff members often did not want to be a part of this. In our organization, people would call in sick because they did not feel comfortable and did not want to be a part of the RCA. These organizations were already thinking about how to change their culture and how RCA could be changed to support this cultural change, which drew them to learn more about RCA², not just because it is a new tool but also because it supports the cultural change that these organizations are trying to make.

Sarah Mossburg: Can the RCA² process be customized for different organizations? What are notable ways organizations have innovated the process?

Jessica Behrhorst: When we teach RCA², we stress that it is a toolkit, not just one tool. It has several different pieces. Risk-based prioritization is an important part of RCA². It determines which events warrant dedicating resources for a root cause analysis. Another piece is the event mapping and causality process. In traditional RCA, many organizations use the fishbone (or Ishikawa) diagram to show causality. In RCA², we use the Five Whys to ask why this event happened and create causal chains. And we also have the action hierarchy to help us think about sustainability. Some organizations use the whole suite of RCA² tools, while others are not ready to make the complete change and implement these new tools. For example, they may start with the action hierarchy or the causation process with the Five Whys. They may leave the rest of the process the same. The recommendation is to use all of these, but we know that that amount of change can be difficult for teams. I encourage organizations to use the tools that their organization is ready for. As they go through the process, they may be able to add additional tools. 

There is also a lot of innovation with organizations using RCA². People are evolving the process and thinking about what works best for their organization, such as setting up teams and who is involved in a patient safety team or RCA team. Those can be customized based on organizational needs or structure. At smaller hospitals getting six people together for 3-hour meetings can be difficult. RCA² is flexible enough to work in those settings. We also see that traditional event investigation is mostly designed as event investigation in an acute care setting, but we know that a lot of our health care happens across the continuum—in ambulatory areas, at home, and in other care settings. We have staff working in these settings, thinking about how they can use and modify these tools to work for them.

A couple of specific organizations come to mind related to innovations – MedStar Health has a strong culture around human factors, so they have been able to integrate human factors into the actions they come up with. Beth Israel Deaconess Medical Center has done a lot of work around emotional harm, and they have modified some of the RCA² tools to look at patient emotional harm and address it in more sustainable ways.² One last example is New York Health and Hospitals, which has done fantastic work around equity. They have taken RCA² and molded it into a way of thinking about event investigation that considers equity.³ It is critical to think about the equity lens as we think about patient safety and our patient safety tools. 

Sarah Mossburg: These are great examples of innovations. You mentioned earlier that RCA² is a toolkit and that there are pieces that people could pull and use. Is there any one piece that you feel must be used by an organization or that you would specifically recommend? 

Jessica Behrhorst: Two parts of it are the most important, and they are the bookends of the process. One is risk-based prioritization, which looks at all events and considers the likelihood of these events happening in the future and their potential severity. We tend to think retrospectively about events that have caused harm—What did we do wrong? What can we fix? But we really want this to become a more prospective tool. How can we think about changing processes that have not yet harmed patients? When we look at events through the lens of this risk-based prioritization, we look at not just the actual harm but the potential harm of the event, which helps us decide whether we want to do an RCA² on this particular event. I advise people I train to think about where they want to put their resources and where their time is most valuably spent. That is a really important tool for deciding to use RCA for more than just sentinel events. That is the tool at the beginning.

At the end of the process, it is critical to have sustainable actions and think about them in terms of improvement. I often think back to my early career. When we did RCAs, we put everything in a binder, which would go to the director’s office. We would pull it down again if a similar event happened, but putting those events and actions away does not create sustainability. We should make sure to build strong actions and think about how to sustain them.

Sarah Mossburg: Earlier, you talked about conducting a root cause analysis as a regulatory requirement for accreditation. How does changing to an RCA² impact that?

Jessica Behrhorst: It actually does not change anything. We are still fulfilling our regulatory requirement to investigate serious events. We expanded it because we are looking at events that we are not required to look at from a regulatory standpoint and are still doing a thorough investigation. Several years back, RCA² got approval from The Joint Commission. RCA² is a process that fulfills and surpasses a regulatory need to help us create sustainable change in our organizations. It also supports that cultural transformation that we are looking for.

Sarah Mossburg: You mentioned that ambulatory and home health are starting to use RCAs. How are they doing that?

Jessica Behrhorst: In the ambulatory setting, fewer adverse events are reported just because of the nature of the setting. We tend to have fewer events to look at for a root cause analysis. To adapt RCA² to this, ambulatory care organizations can use the risk-based prioritization tool to look at events that could have caused harm. Often, near-miss events are happening but are not being thoroughly investigated. Another thing I have seen in the ambulatory setting is events that caused only minor harm. For example, a refrigerator storing vaccines was accidentally left off all weekend and, therefore, all those vaccines have gone bad. If we find this out later, after we vaccinated a few patients, we do not need to do a root cause analysis on each event individually. We have a singular causality and can look at events as a group. That is an easier approach in the ambulatory setting, where events often impact multiple patients. Both of these are ways to overcome the lower volume of adverse events and still get value out of RCA². 

There are not many reports about what happens in the home care setting. We have individuals going out into individual homes, where some serious events occur, like patient falls and medication errors. First, we need to make sure that we hear about those events before we can use the risk-based prioritization tool from RCA². It was designed for acute care, but the tool is flexible. People implementing it in the ambulatory or home health setting need to sit down with others in their organization and think about the potential risk areas, the risk levels they want to see, and potentially modify the risk-based prioritization tool to decide what events they will look at, where do they want to use their resources. 

Sarah Mossburg: What are the main lessons you have learned from years of completing RCAs and RCA²? Where do healthcare organizations tend to struggle or succeed?

Jessica Behrhorst: One thing that I share with people learning or implementing RCA² is that this is a very flexible process. Also, we know this can be a very emotionally charged type of investigation and often a very protected type of investigation. People are sometimes uncomfortable with changing the process of who to include and how we share the findings of the RCA. The process is very flexible, but it does involve having conversations with staff at your organization about how we are going to implement these tools. As a patient safety manager, it’s not the right approach to say “I’m going to change the way that we do root cause analysis, but I’m not going to talk to the risk management staff because they probably have some objections about it.” Part of using this toolkit is related to understanding how we collaborate in our own organization to ensure that the tool supports the cultural transformation we want. We can only achieve that if we have some organizational agreements and alignment around these tools. People are sometimes intimidated by the thought of changing their whole process, but you do not have to; you can ease into it. The rules are not hard and fast. 

Sarah Mossburg: We spent a good amount of time talking about the history and where we are now with RCA². What do you see as the next steps for this work? Are there other areas that you would like to see further researched?

Jessica Behrhorst: Some organizations are finding different ways of using RCA²,which we fully support.  One is thinking about the emotional harm we are causing to patients as well as staff. If we have staff that feel harmed emotionally and physically, they are not going to be able to take care of our patients. I think formalizing the way we integrate equity into this tool is also critical. Also thinking about how RCA² is an integral part of a patient safety improvement program is important. We talk about RCA² as a standalone tool, but to succeed, we need to have all these pieces supporting our patient safety culture. I do not think we need a ton of research, but we do need to formalize the way we look at RCA². We also need to begin thinking about how it is being used in other care settings. We have said this is a flexible tool to adapt to different settings, but we can be more specific and intentional about how people are using it in those areas. That will be helpful as our health care continues to transition into different settings.

Sarah Mossburg: Are there any regulatory or organizational policy changes that should be considered or explored related to RCAs or RCA²?

Jessica Behrhorst: The newly released patient safety structural measures⁴ include initiatives like CANDOR.⁵ And if we think about some of the cultural elements of RCA², how can we include those in some of our structural measures as well? Some of that may be related to how we share the results of our RCA. We did not talk a lot about that, but it is a critical part of RCA². The last action piece also includes follow-up. As a learning organization, how are we sharing those results with the staff involved in the event, the unit where it happened, and other units where a similar event may occur as well? Having policies will encourage organizations to share those findings in a way that protects our staff in terms of psychological safety and just culture.  I would love to see some of these elements included in future iterations of some of our patient safety structural measures.

When I think about RCA², I think about it as part of a greater cultural transformation. I have heard some people say we do not need to do RCA anymore, but I think there is a reason why it has remained a regulatory requirement. When we have serious events or events that could potentially cause serious harm, we need to find out what truly happened. If we think about it as a way of improving our culture of safety and increasing our ability to become learning organizations, it is a critical tool. 

Sarah Mossburg: Well, thank you so much for chatting with us today. This was an interesting discussion. 

References

1. 5 whys: Finding the root cause. Institute for Healthcare Improvement. Accessed November 22, 2024. [Free full text]
2. Diaz-MacInnis L. Healing after harm: addressing the emotional toll of harmful medical events. Beth Israel Deaconess Medical Center. August 22, 2018. Accessed November 7, 2024. [Free full text]
3. Chandra K, Garcia M, Bajaj K, et al. A systemwide strategy to embed equity into Patient Safety Event Analysis. Jt Comm J Qual Patient Saf. 2024;50(8):606-611. [Available at]
4. Mate, K. IHI helps hospitals prepare for new CMS measures to make care safer. Institute for Healthcare Improvement Insights.  August 21, 2024. Accessed November 8, 2024. [Free full text] 
5. Communication and Optimal Resolution (CANDOR). www.ahrq.gov. August 2022. Accessed November 22, 2024. [Free full text]