Radiology reports represent the clinical knowledge of the profession, the tangible, value-based action item upon which clinicians depend. They convey details of diagnosis, information of clinical relevance, and evidenced-based recommendations, all expressed with clarity and accuracy.
Or at least they should. Not all do, and not all radiologists use reports to effectively communicate the relevancy of their profession.
Curtis L. Langlotz, MD, PhD, professor of radiology and biomedical informatics at Stanford University School of Medicine and medical informatics director at Stanford Health Care in California, has spent much of his career focused on the study of and improvement of radiology reporting. Dr. Langlotz recently wrote a book on the subject: The Radiology Report: A Guide to Thoughtful Communication for Radiologists and Other Medical Professionals.
The first six chapters of the book are a style guide for radiology reporting, which a review by Brent Adler, MD, published in Pediatric Radiology, calls “the best radiology focused style guide we have” and “an excellent primer for critical evaluation” of how radiologists work.” Dr. Adler, a pediatric radiologist at Nationwide Children’s Hospital in Columbus, OH, states that he “feels obligated to recommend the book to every resident and attending radiologist.”
When Applied Radiology asked Dr. Langlotz in a recent interview what motivated him to write the book, he said it was because radiology residents have not received sufficient training in radiology reporting. He said that experienced radiologists tend to write better reports, because “with experience they tend to become more comfortable making definitive decisions and also develop a keen sense of what is important. These factors lead to brevity, clarity, and efficiency.”
Dr. Langlotz said he decided to collect and consolidate the published wisdom on what makes a good radiology report. “I also wanted to share some conclusions drawn from my own research and from personal experience that might be beneficial. In addition to preparing a style guide, I added material that I thought a reader would find interesting, such as the history of radiology reporting, the applications of information theory to reporting, and the history of speech recognition systems.”
Precise communication and clarity of language is important for a well-written report. The use of the term “normal” is acceptable with additional explanatory comment, Dr. Langlotz said.
“The word ‘normal’ does not account for the panoply of variations among normal individuals. If a report states that the liver is ‘grossly normal’ or ‘essentially normal,’ a referring clinician may think: ‘In what non-essential ways does this patient’s liver deviate from normal? Are these deviations significant for the clinical decisions I must now make?’ The use of ‘relatively normal’ prompts the question: Relative to what?’ A younger patient? An older patient? This same patient’s most recent exam? Other patients who undergo the exam?” he said.
A report should not be ambiguous or use language that hedges, such as use of the term “no radiographically visible signs of disease.” It is acceptable to be uncertain. “Uncertainty is inherent to radiology and should be expressed clearly in the radiology report,” Dr. Langlotz said. “On the other hand, obfuscating uncertainty with ambiguity of language is counterproductive to clear communication.”
10 Tips for More Effective Radiology Reports
When appropriate and feasible, radiologists should include evidence to help clinicians and patients interpret a report. An example would be a spine imaging report that also contains information about the frequency of common imaging findings in asymptomatic patients.
Many surveys indicate that patients want access to their radiology reports. Dr. Langlotz predicts that direct access to radiology reports will have a positive effect on their quality. “Patients will equate a radiologist’s attention to detail in creating the report with the attention to detail used in detecting disease,” he said. “Consider the failure to edit out ‘The gallbladder is normal’ from a report template when the patient actually has undergone a cholecystectomy. Clinicians might see simply a clerical error. But the patient may believe the radiologist looked at someone else’s scan,” he said.
The Radiology Report — An interview with Curtis Langlotz, MD. Appl Radiol.