The New England Journal of Medicine just published a great piece comparing imaging service utilization in the U.K.’s National Health Service vs. the United States.
The article describes how in the National Health Service in the UK, Radiologists had to act as gatekeepers to preserve the relatively scarce resource of Imaging Services. The infamous “Dr. No” kept the the requestors at the top of their game and quality improved.
The article contemplates the choice between using Clinical Decision Support rules such as ACRSelect or the Radiologist to perform gatekeeping. The flaw in this thinking is that it is not an either/or choice. Decision Support (e.g. ACR Select) is the vehicle to improve and focus the dialog between ordering physician and radiologist. Its not a gate, its a collaborative tool. The notion of anyone in medicine performing gatekeeping in medicine is inconsistent with goal to improve care and outcomes.
The author throws the proverbial baby out with the bath water. The proper way to improve quality and manage imaging resources is to do both CDS AND consults. The idea that EVERY request for an exam should involve a call with a radiologist is extremely inefficient and contrary to the goal of resource management. If healthcare providers invest in CDS to gather feedback in the CPOE cycle, and remove the clearly inappropriate (the 1’s) and the appropriate (the 10’s) from a consult is a benefit to everyone. CDS acts as first pass filter to further define and focus episodes whereby a consultation is required.
The author discusses educations process, but he ignores the fact that CDS is also an educational tool. In fact, implemented correctly, its probably the best tool for education because it provides knowledge at the ‘point of practice’, learning on the fly and in ‘context’, which is arguably the best way to learn anything.
The most interesting points he makes is in the description of his interaction with “Dr. No” in the UK. Of course with no CPOE, Clinical Decision Support wasn’t an option. When Dr. No asked him “what the course of action was if the exam was negative?”, was interesting indeed! It points to a fact that we already know… Asking MORE questions would get a better answer. Unfortunately, the entire principal behind today’s CDS system design is to provide value while causing as little ‘inconvenience’ to the ordering clinician as possible. Even then, that the biggest hurdle to acceptance is to get a a clinician to simply enter a cogent reason for exam! The problem isn’t what a radiologist is or isn’t willing to do as the author suggests. I know of no radiologist who is unwilling to take a call from a referring for guidance. But rather, the problem may well be a culture whereby referring clinicians believe that their practices shouldn’t be questioned. The irony is, it happens all the time with Radiology Benefits Managers and many others areas of medicine and the reason why it works is because payment is withheld and they have no choice, not because there is any perceived clinical value.