Improved communication helps radiologists comply with CF protocol ... for a while

Improving communication with radiologists related to critical findings (CFs) has a positive impact on compliance, according to a recent study published in the Journal of the American College of Radiology. After time, however, the influence of that communication appears to fade.

David M. Yousem, MD, MBA, and colleagues from the Johns Hopkins Medical Institutions in Baltimore, randomly selected 50 radiology reports per month from November 2012 to April 2016.

“These reports were reviewed by the division chief of neuroradiology,” the authors wrote. “If a report contained a CF but had not been flagged as critical or had not been called to the referring physician by the radiologist, or if a report that did not contain a CF had been flagged as critical or had been called as such, it was classified as [a standard operating procedure] error in our study. Afterward, the division chief e-mailed the radiologist who had made the error about the CF mismanagement with the report and an explanation as to why that notification or flagging was inappropriate.”

Yousem and colleagues then tracked the rate of errors over time, tracking how the department responded to having feedback provided by their division chief.

Overall, the study included 2,100 neuroradiology reports, and more than 97 percent of those reports featured CFs that were handled appropriately. A total of 49 notifications were delivered to 15 radiologists over the course of those 42 months, and 11 of the 15 radiologists received two or more notifications.

The authors found that the monthly error rate decreased beginning in month 11 of their study, but then it jumped back in month 33. The monthly error rate from the 11th month through the 32 month was just 1.28 percent, down from 3.98 percent in the first ten months.

What caused the rate to go back up? Yousem et al. noted that in previous surveys, CF notification systems had lost their impact after two to three years, so this may just be a continuation of that pattern.

The authors also mentioned that new hires could be partly to blame.

“It is also true that three new faculty members were added to the team during the late upturn, which may have affected compliance, as they had not learned of the program previously,” they wrote. “At this juncture, a new initiative to reachieve the so-called Hawthorne or notification effect may be required.”

The team concluded that their program did make a difference, but interventions may be required in two- or three-year intervals to sustain improvement moving forward.