Radiologists and patients tend to interpret certain phrases differently, according to a new study published in the American Journal of Roentgenology. Now that patients are reading their own radiology reports more and more, today’s specialists must choose their language wisely.

“A variable interpretation of the radiologist's diagnostic confidence may lead to confusion and dissatisfaction with the report,” wrote Andrew J. Gunn, MD, with the department of radiology at the University of Alabama at Birmingham School of Health Professions, and colleagues. They added that radiologists “should endeavor to use strategies to convey imaging results clearly and effectively.”

Gunn et al. surveyed more than 50 radiologists and more than 100 patients, asking them to assess the “the perceived likelihood of true metastatic disease” of 10 phrases commonly found in radiology reports. The phrases included “likely represents,” “suspicious for,” “cannot exclude” and so on.

Overall, the authors found a “significant difference” between patients and radiologists when assessing the statistical likelihood for most phrases. Patients, for instance, associated “probably meta-static disease” with the highest likelihood of malignancy. Radiologists associated it with the sixth highest likelihood.

Radiologists, meanwhile, ranked “diagnostic for metastatic disease” as conveying the highest likelihood of malignancy. Patients ranked it the third highest likelihood.

The study also revealed that, with some phrases, the two sides were actually on the same page. “There were some similarities in how patients and radiologists perceived the phrases,” the authors wrote. “For example, ‘represents metastatic disease’ and ‘diagnostic for metastatic disease’ were interpreted by both groups as representing a high likelihood of true metastatic disease, even though radiologists still assigned a higher statistical likelihood to each phrase. The qualifying phrases ‘may represent metastatic disease’ and ‘metastatic disease cannot be excluded’ were assigned low likelihoods of true metastatic disease by both groups.

The authors noted that their study did have limitations. For instance, they had a higher response rate from patients than radiologists and “it is unclear how participation rates would affect the results of the study.”