Who's at the table for your root-cause analysis?

A crowded house makes for better understanding

Failure and near-misses offer a significant opportunity to change the way you do something in a way that will benefit patients. But is there a best method of doing a root-cause analysis (RCA)? According to St. Joseph's Hospital in St. Paul, yes.

Those in charge of RCAs presented a best practice for RCAs at the November Minnesota Alliance for patient safety conference.

People often assume if they have a program in place, it must be effective. But According to Rosie Emmons, RN, performance improvement specialist at the hospital, it's not a comfortable process. Emmons has been doing training in RCA across her state for five years. The sessions are always full. That tells her that people are looking for ways to improve how they investigate sentinel and other reportable events.

Here are a few suggestions.

• Put them all in the room. She thinks one mistake people make is limiting who is at the table. It's not enough to have the people directly involved in the event involved in the analysis of it. Rather, you have to look wide and then go deep. Everyone who had anything to do with it needs to be there. That means people from the lab, transportation, or even the day nurse if the event happened at night.

"Often those people may not have a lot to add, but they can help you get a bigger picture of what happened," says Robert Moravec, MD, St. Joseph's medical director. "A couple days ago, we were trying to understand a surgical event, but until we saw the device in question and opened it and looked at it and saw how the event happened, it wasn't clear to us. Even though everyone who was in the room at the time was there, we had to go further."

Emmons adds that if you are going to change a process as a result of the adverse event, you have to have everyone at the table who will be affected by that change.

• Stick to the path. Another mistake is getting sidetracked by tangents — even if they are important. Emmons says you can be talking about an issue related to staffing and how not having enough nurses on a particular shift was an issue in a particular event. If you aren't careful, you can end up talking about how hard it is to plan for a night when there are so many admissions "and before you know it you are talking about staffing in other scenarios."

They may be important issues, but you can't lose focus of your main task: talking about the event at hand and developing a strong action plan to deal with it.

• Look for strength. Make sure you go with a strong action plan. If you don't make clear changes, you're going to reinforce the way you currently do things, which is what led to the problem to begin with, Emmons says. "Nothing will change." You have to make a clear break from the way you did things in the past and start doing it a new way. Simply putting up a poster or sending out a reminder memo about an existing procedure won't bring you the change you seek. "If you are sitting in a room doing a root-cause analysis, you already know this is a repeated problem that relates to a sentinel or reportable event."

This isn't to say you have to go completely back to the drawing board, says Moravec. "You may just need to change one small step in a bigger process." For instance, in surgery, if you have an object fall into a patient and it isn't realized until later, you may only need to change a process by repositioning a tray or keeping surgical sets behind the patient until required.

• Look from every angle. One of the reasons it's important to have a room full of people rather than one or two is that everyone has a different perspective. Think of the movie Vantage Point and how what happened changed depending on who was telling the story. Emmons says by the time an hour has passed, she will have people with a better understanding of what happened and why because of the variety of viewpoints. "Everyone has to be on the same page and see it all the same way. Then we look at the main issues and determine how we will improve." That means letting everyone have a chance to tell his or her story.

Can there be conflict and finger-pointing? Yes, says Emmons. Sometimes they have to do follow-up meetings or take time outs — "I explain that this isn't about blame but about fact-finding," she says. But ideally, everyone in one room at one time as close to the time of the event is best.

• Leading from the head. That's easier to do if you make root-cause analyses something that the leadership in the hospital takes seriously. Moravec is convinced that St. Joseph's does it better than other hospitals, and the main reason is that he or — sometimes and — someone else at the senior level is present at every single one of these meetings. "Our CEO has mandated that a senior leader is at every root-cause analysis. It might be our CEO or our CNO, or executive director of patient care. If it involves a doc, I'm there, or the director of surgery. And the senior staff holds the directors and reports accountable for implementing whatever change is decided. I don't think most places are like that."

If you are a nurse on the day shift and the event involved a night-time fall, you might think it's not important that you show up. But if your chief nursing officer is coming in — on her day off — for a meeting, are you going to miss it? Doubtful, says Moravec. The culture is that this is an imperative, not an option. "Lending a level of importance to the event helps everyone take it more seriously."

• Preaching to the choir. When you are part of a multihospital system like St. Joseph's, having buy-in at the top helps when it comes to spreading the gospel of change across facilities. "We may have an ER event and put in an action plan. But we have physicians that work in three ERs and so we have to put it in place at all of them. We have to keep track of where it is in the organization. Meanwhile, the other hospitals have their own changes that they try to spread to us," says Moravec.

Because senior staff are so involved in every root-cause analysis, and they also are involved in inter-hospital committees involving patient safety and quality improvement, mandating change is probably easier in the HealthEast system than in others. "We keep tweaking things," Moravec notes. "We have documents about RCAs and actions, and a database where we note where things are on the spread to other areas. We bring these things to site quality meetings and review them there, and to the system patient safety committee and review them there." They continue to evaluate new models of reports both for inter- and intra-hospital spread. In something of a meta-QI statement, Moravec notes that when they figure out what works for documenting the spread, they'll spread that out, too.

Discussions about what's happening elsewhere and who is responsible where for what all happen regularly. But there are barriers to it even in a system that has worked on making QI, well, spreadable, says Moravec. "You are working against the habits of highly effective people that have been ingrained for 20 years of practice; they have developed workarounds that have worked so far. They haven't had a failure at their site and they are short on time." That's why having a strong action plan is imperative, he adds. "Putting it on a piece of paper and emailing it out, that's not going to do it."

Other issues they grapple with include changes that involved facilities or equipment expenditure or that have to be approved by medical staff. How to handle those is an ongoing challenge with no simple answers.

• Measure it. People may not think you need to worry about a topic like measurement when doing a root-cause analysis, but Moravec says it's vital. "A good RCA will also put in place a measurement program. If you are doing a checklist for the OR, after you make the change, you want to look at the next N cases to see if it is done and verify the change."

• Find a secret weapon. Moravec says Emmons is a secret weapon. "She has taken this on with a passion and skill set that not many have," he says. "She is objective, non-judgmental, and able to ask why at three or four levels." For instance, a surgeon may try to take ownership of an error. She'll ask why something happened. If he says it was because of something else, she'll ask why that happened. She will keep doing this up the list of reasons until she is satisfied they found the ultimate "why." Many people stop at one or two such questions.

She is also determined to create a strong action plan. That she is a nurse makes clinical staff respond more readily to her. And Moravec says that while she is very patient, she is happy to interrupt and willing to get someone who is sitting in a corner to open up. That circulating nurse who is nervous because the chief of surgery is at the table? Moravec says Emmons has a unique ability to get that person talking and telling his or her side of the story. "She can pull people in."

For more information on this topic, contact:

Robert C. Moravec, MD, Medical Director, St Joseph's Hospital, St. Paul, MN. Telephone: (651) 232-5936.

Rosie Emmons, RN, Performance Improvement Specialist, St. Joseph's Hospital, St. Paul, MN. Telephone: (651) 232-3000.