Radiologists can take steps toward producing actionable reports that are clear, accurate, and easy for referring clinicians to understand, according to a talk delivered on March 4 at ECR 2026.
Presenter Paul McCoubrie, MBBS, a consultant radiologist at Southmead Hospital in Bristol, U.K., and a columnist on AuntMinnieEurope.com, outlined how radiologists can improve their reporting and avoid pitfalls.
Paul McCoubrie, MBBS, talks about how radiologists can employ actionable reporting with referring clinicians at ECR 2026 in Vienna, Austria.
“My take-home message is to think about actionable reporting as a process, and it’s not just the linear model that we used to do,” McCoubrie said. “This is getting the request and the vetting right before the protocol.”
Defining ‘actionable reporting’
Actionable reporting has conflicting definitions. The American College of Radiology (ACR)'s 2014 guidelines state that abnormal scans should prompt rapid alerts and categorization by urgency. But the Royal College of Radiology (RCR)'s 2018 guidelines prompt alerts for any reports, and later models focus on action.
“Overall, this is more complicated, and this is more than just steering people to action,” said McCoubrie, who is the author of the books "The Rules of Radiology" and "More Rules of Radiology."
He offered the following definition for actionable reporting: An actionable report starts with a request for an appropriate investigation for the right patient, which is then performed at the right time and finishes when a clear, succinct, readable report containing appropriate guidance is safely delivered with appropriate urgency to the requesting clinician.
The resulting process includes conveying alerts from constructed reports to referrers, which then leads to clinical action.
McCoubrie also outlined five principles of actionable reporting: answering the clinical question, the need for brevity, clear communication, avoiding unreadable reports, and providing effective guidance.
Regarding this last, he said radiologists should strive to create a “Goldilocks” report, where just the right amount of guidance is given, not too much or too little. Citing previous reports and studies, McCoubrie said reports should be about three to four sentences.
Current communication between radiologists and referrers
In another talk, McCoubrie discussed controversies in actionable reporting. Part of this presentation included an interactive session where attendees (n = 96) were polled about their practices’ communication.
While 49% of attendees reported that they have one system for all communications about urgent results, 23% said they expect referrers to read the report.
Attendees also listed results communicated over and above the written report, including unexpected abnormalities that may need action (89%), a report addendum that may change management (65%), expected but serious abnormalities (60%), unexpected cancers (53%), and recurrence or progression of known cancers (18%).
And 45% of the polled audience said they have no formal policy about alerting referrers to priority radiology reports, just informal rules, while 53% said they do not categorize abnormalities by emergency.
Although some hospitals may institute alerts for findings that may trigger urgent action, McCoubrie said excessive alerts may hinder radiologist productivity.
“They will hate the system,” he said. “They will not use it, unless it’s a single-click system.”
However, he added that not having alerts for benign or normal findings may encourage referrers to not read all reports, which carries risk and may lead to radiologists taking the blame should the findings be serious.
Common pitfalls and how to avoid them
Prior research suggests that between 23% and 46% of clinicians may not follow recommendations from reports. Reasons for this include disagreement, uncertainty, or lack of awareness of the reports between radiologists and referrers.
McCoubrie listed the five most common pitfalls in reporting:
Wrong test, time, and protocol. This can be prevented at the source via “clear and honest” clinical questions. McCoubrie said systems should encourage face-to-face discussions when patients are very sick or have complex cases.
Dysfunctional categorization. Having a formal system that works and having a critical/urgent category are needed. And while emphasis should be placed on reading all reports, reporting systems should not create “alert fatigue” for radiologists.
Report text not actionable. Text should answer the question and be clear, brief, and readable.
Poor alert system. While no one system is best, McCoubrie said verbal communication is best for emergency situations, clerical interventions are best for routine cases, and electronic alerts are efficient yet complex. McCoubrie added that electronic alert systems are often neglected but should be a requirement.
Appropriate action not taken. Conclusion and guidance may not be clearly phrased, and the referrer may have either disagreed with or never seen the report.
Our full coverage of ECR 2026 can be found here.
