How to Combat Radiologist Cherry-Picking

Elizabeth Bergey, MD.

As humans, we often gravitate toward the path that yields fast, easy results rather than the best course overall, especially when we’re under pressure. When a radiologist chooses the case they want, rather than the next one on the worklist, it is referred to as “cherry-picking.” It can have a negative impact on both patient care and staff morale.

In the past, proximity and team collegiality helped keep this behavior in check. With a team that worked in a common location–and often had coffee together–cherry-picking was easier to spot, and a sense of team comradery provided a natural deterrent. Today, a geographically dispersed team, combined with increased workloads and complex worklists, has made cherry-picking more prevalent and more problematic. For a radiology practice to function optimally, it’s an issue that needs to be addressed.

Discordancy and complexity lead to cherry-picking

Cherry-picking is often the result of a discordancy between the wRVU (work relative value units) credit awarded to the interpreting radiologist and the actual amount of time, energy, and skill it requires to interpret a study. This happens because of flaws within the RVU system. Compounding the problem in radiology is the absence of complexity modifiers.

Another factor contributing the cherry-picking issue is the size and complexity of the worklist. With a long, complex list of cases needing attention and no clear system to designate case distribution, it’s easier and more tempting to cherry pick.

Active, passive, and perceived cherry-picking

Whether it’s blatant or more subtle, it’s likely that cherry-picking is having a bigger impact than we realize. It might be the more active, obvious cherry-picking, where someone bypasses the first four cases on the list and takes number five instead. There may even be a few in the group who treat it like a sport, gaming the system and cherry-picking at every opportunity.

Less obvious is the cherry picker who sees a tough case at the top of the list and decides it’s time to use the restroom and check their email. This passive cherry picker is often not alone in their wait and watch approach – that temporal bone study might sit at the top of the list for a bit. Then, when someone finally takes it, boom, boom, boom -- numbers two through six are quickly grabbed as the passive cherry pickers jump back into action.

There's also the concept of perceived cherry-picking, where people in the group are skipping cases because they're assigned to do something special. Perhaps they’re working on a research project related to renal ultrasounds. They’re going to be skipping all of the thyroids to get to the next renal.

Potential impact on care, productivity and morale

On the surface, it can seem that cherry-picking is frustrating but relatively harmless. However, it can have a profoundly negative impact on clinical care, productivity measures, and team morale.

Left unchecked, cherry-picking:

  • Increases risk to patient safety - Cherry-picking can lead to delays in patient care when high priority studies are not given the prompt attention needed.[1]
  • Creates productivity challenges – It can be nearly impossible to effectively compare work effort using wRVU as the metric and challenges radiologist management analytics when some radiologists are cherry-picking. Those who are gaming the system have an artificial increase in their apparent effort while the radiologist who is ‘playing by the rules’ suffers from an apparent decrease in effort. In addition, the process of choosing cases can cost each individual radiologist more than 15 minutes of time each day, creating an unnecessary decrease in productivity that multiplies to alarming levels across a large practice.  
  • Decreases staff morale – Cherry-picking can harm team dynamics and cause a tremendous amount of job dissatisfaction, leading to increased levels of burnout. 

Using technology to adapt to higher workloads and increased complexity

The contemporary enterprise often includes several PACS, with each having its own worklist. Even if a consolidated worklist is present, it can be challenging for humans to optimally manage the exam workload in real time across a dynamic workforce of diverse readers. To address this issue, many PACS vendors provide some form of “auto serve” algorithm that distribute worklists to the appropriate groups of readers. 

The more sophisticated tools choose the next case for a radiologist to read based on the best interest of the enterprise. Called Intelligent Worklists, these solutions can yield enormous productivity gains, eliminate “list anxiety,” and help boost turnaround times.

However, even the most sophisticated of these still have shortcomings in today’s dynamic radiology environments. The best of these may provide an early alert when bottlenecks are anticipated and extra resources need to be called in, but many rely on assumptions that all readers are the same, lacking the ability to take into account the wide variety among readers. There needs to be a better way.

That better way is here. GE Healthcare’s newest PACS systems are designed to optimize your most valuable resource – your radiologists’ time - through integration with Helix Pace and Balance from Quantum Imaging. This innovative intelligent workload management solution can eliminate cherry-picking and allow you to properly credit radiologists for their work, while at the same time prioritizing optimal patient care.

Discover how GE Healthcare’s Intelligent Workload Manager solution can help your enterprise address cherry-picking and boost your radiologists’ productivity and morale. Contact us for more information.

 

ABOUT THE AUTHOR

Elizabeth A. Bergey, MD, is President and CEO, Quantum Imaging & Therapeutic Associates. Dr. Bergey joined Quantum in 2001 as a diagnostic radiologist. In 2005 she was elected to the Board of Directors and served as treasurer. In 2007 she was elected as Quantum’s Board of Directors Chairman, President of the physician group and CEO of the professional corporation.

REFERENCES

[1] Chan, T., Howard, N., Lagzi, S., & Romero, G. (2018). Cherry-picking and Spillover on Service Level: Evidence from a Radiology Workflow Platform. http://dx.doi.org/10.2139/ssrn.3273494