The reading room is the center of diagnostic function in a radiology department. It’s also the site of numerous interruptions as radiologists perform complex tasks.
At Cincinnati Children’s Hospital Medical Center, diagnostic radiologists were being interrupted on average every 186 seconds. The distractions were numerous: phone calls, clinical questions, pages, and more. Ethan Smith, MD, an associate professor of radiology and director of thoracoabdominal imaging, and Janet Strife Chair for Quality and Safety at the University of Cincinnati, teamed with colleagues to address the issue.
Dr. Smith recently answered a few questions about his project, which has resulted in an 80% increase in median time between interruptions for the 35 faculty radiologists, eight pediatric radiology fellows, more than 200 technologists, and 12 reading room assistants.
How did the idea come about to do something regarding interruptions?
“Stress and burnout in the workplace are major concerns that affect all of us. We found during faculty meetings and other informal discussions that one of the major stressors for radiologists in our department (both faculty and fellows) was a perceived excessive number of interruptions that occurred while the radiologists were trying to accomplish their daily work. Interpreting radiology studies requires a substantial amount of focused concentration and cognitive effort, and frequent interruptions have been shown to lead to increased time to complete complex tasks, decreased performance, and increased stress. We set out to do a formal quality improvement project in order to understand how frequently interruptions were occurring in the reading room, and if there were specific interventions we could pursue in order to limit the number of interruptions.”
What did your project entail?
“First, we formed a team of radiologists, reading room coordinators, CT/MR technologists, and radiography technologists. The group came up with a simplified process map of how interruptions occurred and how interruptions were handled currently. We then listed all the possible reasons interruptions could occur, and what could go wrong in the current process. From these lists, we developed ‘buckets’ of different interruption types and chose the most frequently occurring types as our targets for intervention. Most of our interventions ended up focusing on processes related to communication within the radiology department (most frequently communication between technologists and radiologists). The most successful interventions focused primarily on standardizing when we communicated (for example, reducing the number of required image ‘checks’), and how we communicated (for example, by developing an automated phone tree to reduce unnecessary interruptions).
“The data collection was done manually, with a person observing in the reading room for one hour per day and recording the length of time between interruptions, the number of interruptions, and the type of interruption (in person, pager, or phone). Our primary metric was the time between interruptions, because by increasing the time between interruptions, that would increase the amount of time the radiologists had to do their primary task, which is to interpret imaging.”
What have been the results?
“We have definitely seen improvement. Our initial goal was to increase the time between interruptions by 50%, which we did accomplish. However, we found that even increasing the time between interruptions by 50%, there was still a sense among radiologists that reading room interruptions remained a major stressor. Since completion of the formal project, we have collected more data and are developing additional interventions.”
What has been the reaction from your reading room personnel?
“So far, the work has been very well received. The simple act of collecting the baseline data and understanding how frequently interruptions were occurring was very helpful. The baseline data helped everyone in the department understand and acknowledge how significant the problem of interruptions was in our reading room. Once the scope of the problem was clear, it made it easier for everyone to get on board with the interventions we wanted to test. There have been minor issues along the way that have required that some interventions be abandoned or altered, but in general, most have been successful and have been really well received by the radiologists, technologists, and reading room coordinators.”