Implementation of an institutional communication pathway using electronic flagging and organisational audit for non-emergent actionable radiologic findings

In the modern hospital, a radiology report is more than just a list of observations; it is a critical bridge between a scan and a patient's care. When a doctor orders an image of the body, they are looking for answers, and the radiologist is the expert who translates the picture into a story that guides treatment. For urgent, life-threatening discoveries, hospitals have long had strict rules to ensure the message is delivered immediately, often through a direct phone call. However, medicine is full of findings that are not emergencies but still demand attention. These are discoveries that, if ignored, could lead to a missed diagnosis or a delayed treatment plan, yet they do not require a frantic call in the middle of the night. The challenge for any large medical center is ensuring these important, but non-urgent, notes are actually read and acted upon by the doctors responsible for the patient's care.

A team of researchers at Gunma University in Japan set out to solve this specific problem of communication. They wanted to know if they could build a reliable system to catch these important findings and make sure they were followed up on. To do this, they created a new pathway that combined a digital alert system with a human check. In their system, when a radiologist finished a report and spotted a finding that needed attention but was not an emergency, they would assign a special electronic flag to the report. This flag acted as a digital marker, signaling that the report contained something important. But the system did not stop there. About two weeks later, a dedicated team from the hospital's Medical Quality and Safety Division would review these flagged reports. They would check the hospital's electronic records to see if the referring doctor had actually opened the report and if they had documented any next steps. If the report sat unread or if the follow-up was missing, the safety team would step in to contact the doctor directly. This study was not designed to prove that the system was perfect, but rather to describe how it worked in the real world and what kinds of findings it caught.

The researchers looked back at nearly 29,000 diagnostic imaging reports generated over a six-month period at their university hospital. These reports covered a wide range of scans, including computed tomography, which uses X-rays to create detailed cross-sections of the body, as well as nuclear medicine exams, magnetic resonance imaging, and standard X-rays. During this time, the radiologists used their new electronic flagging system to identify non-emergent actionable findings. They found that in about 1.83 percent of all the examinations, there was at least one finding that required timely clinical attention. This means that roughly one out of every fifty scans contained a discovery that, while not an immediate crisis, needed to be addressed to ensure the patient received the right care.

When the team broke down these findings by the type of scan, they saw that computed tomography scans produced the most flagged results. This makes sense, as these scans provide a very detailed view of the body and are often used to look for complex issues. The most common type of finding they flagged was related to cancer. Specifically, the most frequent category was the unexpected progression or spread of a cancer that was already known to the doctors. The second most common category was a new suspicion of cancer that had not been diagnosed before. For these new suspicions, the researchers followed up to see what happened. In about one-third of the cases where a new cancer was suspected, further testing confirmed that the patient indeed had cancer. This confirmed that the system was successfully catching serious conditions that might otherwise have been missed or delayed.

The true value of this study, however, lies in what the organizational audit revealed about human behavior and system reliability. The researchers discovered that even with a digital flag, things could still go wrong. When the safety team checked the records two weeks later, they found that in ten cases, the referring doctor had not even opened the report. In another fifty cases, the doctor had opened the report, but there was no clear record in the medical file showing that they had taken appropriate action or planned a follow-up. In total, the safety team had to intervene in sixty cases to ensure the patient was not left without care. This finding suggests that simply sending a report with a warning sign is not enough to guarantee that a doctor will act on it. The safety team's role in checking the records and making direct contact served as a crucial backup, catching the cases where the initial communication had failed.

The study highlights that clinically important findings are a regular part of routine radiology, appearing in nearly two out of every hundred scans. By combining a structured electronic alert with a human audit process, the hospital created a practical model for managing these findings. This approach moved the responsibility for communication beyond just the radiologist and the referring doctor, involving the hospital's quality and safety division as a partner to ensure nothing fell through the cracks. The researchers noted that their system was built to fit their specific hospital workflow and was not necessarily a universal solution for every medical center. Nevertheless, the results show that such a pathway can effectively identify important findings and provide an extra layer of protection for patients, ensuring that the story told by the scan is actually heard and acted upon by the medical team.