AJR InBrief Banner 

AJR February 2018

Lane F. Donnelly
Corresponding Author
Stanford University
Palo Alto, CA

“Practical Suggestions on How to Move From Peer Review to Peer Learning”

Many radiology departments and imaging facilities are moving from peer review to a peer learning model in order to educate radiologists as part of a collaborative, continuous learning paradigm. Peer learning is seen as a more constructive, supportive way of looking at errors than peer review. The authors of an article published online ahead of print in the March 2018 issue of AJR show that peer learning is a more constructive, supportive way of looking at errors than peer review.

Led by Lane F. Donnelly of Stanford University’s Lucile Salter Packard Children’s Hospital, the authors noted that essential resources needed for movement toward a process of peer learning include peer review software, faculty time, and leadership support.

The first step in transforming from peer review to peer learning is for practice leaders to explicitly identify and discuss the transition with relevant stakeholders, including members of the radiology practice, hospital leaders, and relevant administrators.

The authors of the study found that stakeholder skepticism often turns to support when the rationale, the plan, and the practice leaders’ commitment to legitimate improvement, are illustrated in depth.

The authors’ research on how cases are selected for peer review systems showed that random selection of peer review cases is often emphasized so that error rates for individual radiologists can be calculated.

The second step recommended by the authors is to place greater emphasis on active identification of cases with learning opportunities. Some institutions have already abandoned or are considering abandoning random auditing of imaging cases. Others believe that specific types of deficiencies, such as inappropriate follow-up recommendations and incorrect reporting structure, can be identified by means of random audit, and this provides a mechanism to better identify these types of learning opportunities.

The third step is to replace numeric scoring of errors with qualitative descriptions of learning opportunities. For example, it may be helpful to classify contributed cases as “great call” or “learning opportunity.” A great call is defined as a case in which a radiologist has made the correct finding and interpretation but there is a reasonable chance that another radiologist would likely have not done so. Learning opportunities are cases in which there has been a perceived error, deviation from best practice care, or a system-related problem.

Errors are an inherent part of the practice of medicine, including radiology, the authors stated. The challenge, and the great opportunity, is to simultaneously accept this reality and leverage it for improvement.


Do you have a story idea for InPractice?
Contact Mike Mason, Communications Manager, at mmason@arrs.org for more information.