Process Designed by ASEM Faculty Effects Change in a Wide Range of Industries
To err is human. And when human error is judged to be the cause of serious workplace accidents, identifying the person to blame may be an equally human reaction.
As longtime safety professionals, Ivan Pupulidy, Ph.D., and Crista Vesel, MSc., understand all too well the high stakes involved in human endeavors. But when an accident results in catastrophic consequences, Pupulidy says well-intentioned investigations could have unintended results. “If investigators are searching for a person to blame, they will usually find a scapegoat,” Pupulidy says, “but that does very little to prevent similar accidents from happening again.”
Over the past 18 years, Pupulidy has been working to change the narrative around accident investigations. As the Director of the Office of Innovation and Organizational Learning for the U.S. Forest Service, Pupulidy developed and implemented a process called the Learning Review in 2013. This new process sought to shift the focus from laying blame for human error to instead understanding why it made sense for people to do what they did.
Since Pupulidy’s retirement from the Forest Service, he and Vesel have helped implement the Learning Review in various industries around the world, including aviation, light rail in Europe and Great Britan, and most recently in the New Zealand Healthcare system. Pupulidy explained, “We were very proud to hear that the entire medical system in New Zealand has adopted the Learning Review.”
Today, the Learning Review and many of its main characteristics are key components of courses taught by Pupulidy and Vesel in the UAB Advanced Safety Engineering and Management Program (ASEM).
Recently, the two sat down with us to discuss what makes the Learning Review different from other types of accident investigations.
UAB Engineering: To start with, can you explain in general what the Learning Review is?
Pupulidy: The Learning Review was originally designed for the Forest Service to replace the “Serious Accident Investigation Guide. It was designed to understand the context that surrounds human decisions and actions.
UAB Engineering: Why was that necessary? Surely earlier accident investigations considered human actions.
Pupulidy: There is a tendency when investigating an accident to look at humans as the source of the problem. When you look at a normal work environment, you see that humans are usually the source of safety. As humans, we intentionally create safer conditions at the point of work, when we sense high risk or face something new to us.
So why do we look at humans being the source of the problem when something goes wrong? Well, often it is that there has been a bad result. Traditional investigations tend to judge the actions as being “bad” if there is a negative result. But when the same kind of work or the same kind of innovation (action) is successful, we reward it and say, “That’s thinking outside of the box.”
Vesel: There is a lot of drama around accidents. When a serious accident occurs, you get the money, the time, and the people to investigate the accident, but there is an urgency to the investigation that can often prevent learning from the accident. Investigators already have an idea in mind of where they want to go with the investigation to get to “the cause” of the accident.
UAB Engineering: It seems like a natural tendency for an accident investigation to focus on finding the cause. But you say that approach often stops us from learning from the accident. How so?
Pupulidy: Using the old guide, we determined that the Forest Service was, in many cases, blaming firefighters for their own deaths. When catastrophic accidents resulted in loss of life, the investigations commonly stated that the firefighters failed to follow the rules, regulations and procedures. We agreed that was part of the story, but the question we had to start asking was, “Why did it make sense for them not to follow rules, regulations, policies and procedures?” These were highly trained professionals who, in moments of crisis, made decisions that resulted in their death. It is ludicrous to think that these decisions intentionally led to the accident. In asking the question why it made sense, we considered how the situation contributed to the actions taken by the participants.
This questions moved us to start to think differently. When we approached an investigation from this perspective, we started asking much better questions. The result was a very context-rich narrative and a very content-rich suite of findings that we brought to leadership, and leadership said, “We need more of this.”
UAB Engineering: The Learning Review was implemented in 2013. How much difference has it made?
Pupulidy: Some of the gains we’ve had from it have been profound. Before, we had a lot of accidents. We had 400 firefighter line-of-duty deaths between 1992 and 2007—an average of around 20 fatalities per year, which remained steady for years. That number has dropped off to less than 5 per year. In fact, it’s dropped off enough to where we don’t even look at the statistics the same way anymore. We used to look at the number of fatal accidents each year, now we focus on learning and the old trend analysis is no longer important. It shows a flat line over the past several years. The Learning Review contributed markedly to that.
UAB Engineering: Firefighting is a dangerous occupation with a lot of unpredictable factors. How does a process designed to investigate firefighting accidents work in other industries, such as healthcare?
Pupulidy: Regardless of what industry you’re working in, there is always the risk of human error. To err or not to err is not a choice we get to make. So, one of the first steps of the Learning Review is to build a network of influences map, which displays the influences or performance-shaping factors that existed during the event.
We can’t change the fact that humans are going to make mistakes. But we can change the conditions under which humans work. The network of influences map gives us a way to recognize, make sense of, and start managing those conditions. Rather than try to change the worker, leaders begin to manage the situation that surrounds the worker.
The next thing we do is to redefine situational awareness. That’s a concept that has had a lot of goofy definitions assigned to it—not the least of which is that firefighters or any individual in the workplace needs to be somehow omnipotent.
If we redefined that to say, “Let’s make them aware of the pressures that surround them,” that’s a very different and useful conversation for frontline personnel to have.
Statement from New Zealand's Te Tahu Hauora Health Quality & Safety Commission
“As part of our newly released ‘Healing learning and improving from harm: National adverse events policy 2023, Te Tāhū Hauora Health Quality and Safety Commission, New Zealand, recommend the learning review method to review health care harm. The learning review is designed for complex systems like health and is heavily influenced by human factors and safety science. It reflects the need to move from blame to understanding while addressing the biases of more traditional methods. Learning reviews discuss how harm events are often a product of the system not individuals making mistakes, thus creating learning opportunities at a system level that can be translated into sustainable actions for improvement. The systems safety team at Te Tāhū Hauora look forward to collaborating further with Ivan as we roll out the learning review education program.”
UAB Engineering: One of the courses you teach in the ASEM program deals with the language of investigation. How does the language or vocabulary you use affect outcomes in the Learning Review?
Vesel: One of the best indicators of a shift in culture is a shift in language, and you can see that shift in organizations who have implemented the Learning Review. Unlike probably any other accident model out there, we don’t use the word “cause” in Learning Reviews. Once you stop looking for cause, you can open yourself up to the influences found in the system or culture. And you create a whole new narrative around the investigation.
This is what helps us find learning opportunities in normal, everyday work—not just the accidents. The accident, frankly, becomes less interesting when we begin to see cultural changes in normal work environments.
Pupulidy: We understand that the best thing we can do in learning mode is to engage in dialogue. So frequently these Learning Reviews present a complex narrative from multiple perspectives. We take that, and we pose open-ended questions that we don’t necessarily know the answer to. By engaging in that dialogue with humility, we ask, “What thoughts emerged for you as you worked your way through this narrative?” That piece is designed so a frontline leader can take a crew and bring them together and say, “Hey, let’s look at this scenario and let’s see how it applies to our work.”
UAB Engineering: Both of you teach in UAB’s ASEM program. Is Learning Review incorporated into the curriculum?
Vesel: Definitely. In the ASEM program, we teach a course on the Learning Review called “Learning-based Response to Accidents and Incidents”, as well as a course in the language of safety, crisis leadership, and a course in critical thinking that is preparation for the capstone where we bring in elements of social psychology.
Pupulidy: In Spring 2024, we will start two new courses: Applied Semiotics in Safety, and Human and Organization Potential (HOP). Many of the concepts that are at play in the Learning Review are relevant to all of these courses and play an important role in preparing our students to improve safety programs around the world.