40% of images, radiology reports requiring follow-up are never reviewed

Though the combination of PACS and electronic medical records (EMRs) have given ordering providers a chance to look at images and radiology report themselves, many are still not reviewed, according to a new case study published in the Journal of the American College of Radiology.

“Clinicians’ independent access to these two components allows them to measure the ‘value’ of the images (accessed by the PACS) versus the ‘value’ of the radiologists who provide the reports (accessed via the EMR),” wrote lead author David M. Yousem, MD, of the Johns Hopkins Medical Institution in Baltimore, and colleagues.

Still, in spite of clinicians having access to reports on their own, a past study found 7.5 percent of all radiology studies go unread.

Yousem and colleagues sought to determine the characteristics of studies that were “most at-risk” of being missed, in an effort to decrease the number of unread studies. They specifically reviewed neuroradiology studies from all patient settings that were collected at their during September 2016.

Date and time stamps were obtained for when the orders were requested; when the patient completed the exam; when the imaging was accessed; and when the radiology reports were viewed. The information was tracked over an 8-month period using the viewing logs of the EMR to determine which studies had neither the imaging nor the radiology report accessed in the eight months after the imaging order was placed.

Images were classified as “normal,” “minor unimportant findings,” “findings requiring intervention or follow-up” or “critical.” They were also classified as being inpatient, emergency department, outpatient or community-based (patients of providers who are part of remote outpatient practices and connected with the facility, but who may not have direct access to PACS or EMRs). Finally, they were also classified by department.

Of the more than 8,200 neuroradiology-specific imaging studies, Yousem et al. found 90 percent were completed and approximately 8.4 percent had neither images or imaging reports viewed in PACS or EMRs. They also found:

  • 12 different departments had a minimum of one study ordered in which neither the images nor report were viewed.
  • Otolaryngology (12 percent) and neurology (11.7 percent) had the first and second highest percentages of never-viewed studies compared to the number of imaging studies ordered.
  • Outpatient imaging (44 percent) and outpatient community setting (33 percent) were the first and second largest number of studies with images and reports that were never reviewed. Both also had a higher rate of studies (14 percent) that were never viewed with respect to total orders.
  • Of the 124 never-viewed studies, 20 percent were classified as normal, 40 percent were minor unimportant findings, 40 percent were classified as findings that require intervention or follow-up and less than 1 percent were critical.

“Minimizing the number of studies that are not viewed is important to patient care,” Yousem et al. wrote. “It ensures that important and perhaps time-sensitive findings have appropriate clinical or imaging follow-up.”

The researchers recommended an “alert system” that can help ensure studies do not go unread. Or, the researchers noted, patients can also choose to participate in a system in which they would receive an alert if the radiologist or the ordering physician has not reviewed the images or report. An institutional list of critical findings may also help ensure there are no delays in diagnosis or treatment.

While radiologists are taking responsibility in ensuring imaging follow-up, the researchers noted this increases their workload. To avoid mishaps and negative consequences, action must be taken.

“As this study demonstrates, the rate of studies with images and reports that are not viewed can be high, particularly in the community outpatient setting by remote referring physicians,” the researchers concluded. “It is important to address this high rate and possibly implement ways to minimize the number of these non-viewed studies.”