Melinda Jean Yeh MD, Presenter: Nothing to Disclose

Stephanie Waygwen Hou MD, Abstract Co-Author: Nothing to Disclose

David E. Avrin MD, PhD, Abstract Co-Author: Stockholder, Amirsys, Inc Consultant, Amirsys, Inc

Thomas H Urbania MD, Abstract Co-Author: Nothing to Disclose

Subcritical imaging findings—those which require follow up but don't pose an immediate threat—create a challenge in the radiologist workflow. In 2011, a semi-automated system for communicating subcritical results was implemented. Subcritical reports are flagged using an XML keyword, and dedicated support staff communicate subcritical findings to the responsible clinician. Our purpose is to assess the performance of this system by analyzing the follow up and significance of flagged radiological findings.

IRB approval was obtained. Retrospective review of consecutive subcritical alerts from 3/21/11 through 2/19/12 was performed (n=2805). The most common study types were CT abdomen/pelvis (17.3%, n=484), PET/CT (8.6%, n=242), CT chest (7.9%, n=222), and US pelvis (7.9%, n=221). In all cases, the responsible physician was contacted, most often on the next business day. The first 253 consecutive subcritical results were analyzed to assess follow up rates and significance of the result (defined as resulting in a change in patient management or outcome). Overall, 62% of findings were followed up (n=156), and 33% of those findings were significant (n=52). When specific follow up recommendations were made (n=119), adherence was 81% (n=96). Additionally, 63% of findings were incidental (n=159) (defined as being unrelated to the study indication). Incidental findings had a lower rate of follow up (48%) compared to non-incidental findings (84%), as well as a lower rate of significance (20% versus 45%).

A substantial proportion of patients being imaged at our facilities did not receive all their medical care within our network, so the percentages above likely underestimate the true rates of follow up and significance. The lower rate of follow up for incidental findings (48%) is of concern given that 20% of these findings were clinically significant in our sample.

A semi-automated system for communicating subcritical findings allows for prompt notification of referring clinicians, a relatively high rate of follow up, and frequently identifies issues that affect patient care. Further refinements should focus on increasing the rate of follow up for incidental findings.

Yeh, M, Hou, S, Avrin, D, Urbania, T, A Semi-automated System for Communicating Subcritical Results: Follow Up.  Radiological Society of North America 2013 Scientific Assembly and Annual Meeting, December 1 - December 6, 2013 ,Chicago IL. http://archive.rsna.org/2013/13025124.html