2014 Patient-centric Imaging Awards | Health Imaging

Radiology Order Entry Clinical Decision Support

Mount Sinai Medical Center, New York City


Radiology orders on inpatients are entered through our EHR. Previously, providers could enter the clinical indication from a structured list or via free text, but there was a desire for clinical decision support at the time of order entry to diminish inappropriate utilization of radiology services.


We deployed a radiology order entry clinical decision support system, targeted at CT and MRI, to implement ACR appropriateness guidelines within our EHR environment. A structured list of “clinical Indications” provided by the decision support system replaced our homegrown list. Providers were presented with this list.

Implementing such a program has several steps.  We are midway through the overall process. The first part of the process involved change management principles.  Along with our EHR team, we educated our provider staff to changes in the Radiology Ordering interface that were needed to implement this system.  We made some modifications to our user interface to address initial concerns.

For nine months, we ran the system in the background—that is the system provided an appropriateness score of 1-9, 9 being the strongest indication of an exam being appropriate. This was a data collection phase to understand the ordering practices and patterns of our providers.  We did not show the providers the decision support scores during that phase.


We learned that on a weekly basis, approximately 10 percent the exams ordered received a low score between 1-3. Just over 2 months ago, we turned on display of the decision support score, showing providers the score when it is between 1-4 and suggesting alternative exams through the automated interface. We are just beginning to analyze the data to look for change in ordering patterns.

We have anecdotal comments that this has been helpful. After 4 to 6 months, we will analyze the data to determine if there is a change in overall performance


Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s