Bradley J. Erickson, MD, PhD, FSIIM
Chair

The American Recovery and Reinvestment Act of 2009 (ARRA), also known as the ‘stimulus package’, promises to spend a significant amount of money promoting Electronic Health Records (EHRs) through an incentive system. Those physicians who make “Meaningful Use” of an EHR will receive incremental payments from CMS starting in 2011. That key phrase, ‘Meaningful Use” has been the subject of much discussion. Several groups have already proposed definitions. While those definitions have included many valid points, none has even mentioned imaging, including the initial recommendations of the HIT Policy Committee. That, despite the fact that imaging is a large part of the cost of healthcare in America, and is one of the most rapidly increasing components.

It is also apparent that imaging is a critical component of modern healthcare—blindly decreasing imaging across the board is not a viable option unless we are willing to accept a decline in quality. Imaging encompasses not only radiology images, but also pathology, ophthalmology and the rapidly growing field of telemedicine. Imaging-specific criteria should be included in the definition of meaningful use because imaging is a large part of the cost of healthcare in America, and is one of the most rapidly increasing components. It is also widely recognized that, when measured by data volume, the largest part of any patient’s longitudinal medical record is comprised of images and related data. Images are also significant facilitators and enablers of improved communication among the inter-disciplinary care-team.

What is required is information to help guide the use of imaging to maximize quality and safety while reducing cost. We will have to do more with less. We will have to become more productive. Imaging Informatics is at the crux of this issue—we are focused on leveraging information to achieve these goals. For that reason, we wish to advance the following as important components of meaningful use of EHRs for imaging physicians:

1. Electronic medical records must enable the sharing of images and associated data between health care organizations.  Images should be available anywhere the patient might need them. This requires sharing of reports and annotated images between health care organizations. There are now formal standards and best practices (e.g., Integrating the Healthcare Enterprise) for accomplishing this task.  But despite the proven successful use of these standards by some institutions, adoption remains minimal.   Effectively sharing images and reports could significantly reduce unnecessary imaging examinations, which will reduce costs while increasing quality, improving timeliness of treatment, and enhancing patient safety.

2. Electronic medical records must provide decision support for the selection of the right imaging examination.  For example, currently available decision support systems can help the requesting provider determine whether an intravenous contrast agent should be administered, and can suggest more appropriate and less costly alternative tests.  Evidence now shows that order entry decision support reduces imaging costs and unnecessary examinations.  This scientific evidence is strong enough that some third-party payers have already accepted the use of these systems in lieu of pre-authorization.

3. There must be recognition that physicians do not all do the same thing. One of the listed elements of ‘meaningful use’ is focused on e-prescribing. While this is useful to some physicians, a substantial number of physicians do not prescribe medications outside of the immediate delivery of medications during a procedure. As such, e-prescribing should NOT be a required component of meaningful use for some physicians. At the same time, imaging physicians should make use of decision support tools where appropriate, including 3D and 4D visualization tools, computer-aided detection and diagnosis tools, and other image enhancement methods.

Certainly, there are more elements that some will wish to propose. Indeed, my original list was longer, but some of those were more ‘stretch goals’ than items that could realistically be in place for most practices in the next 2-4 years. It is worthwhile discussing some of those pieces (e.g. structured reporting is certainly something that would increase efficiency, value, and quality) but I would like to have this blog post focused on near-term practical targets.

Please share your comments below.

Dr. Erickson is a Professor in the Department of Radiology at the Mayo Clinic, Rochester.

My co-leader (Tony Seibert) and I took pictures during our interactive session on setting policy for storing images from the new 64 slice plus CT units.  Some of the units can generate 15,000 images for a study and these clearly will need careful handling.  So the question is – Do you save them all?  Do you save screen captures of the processed data which is usually more useful than the 15,000 individual images? Do you save both?  If you save the “raw” 15,000 images, how can you show in the future how they were processed and displayed for interpretation and if you only save the processed data, how can you look back and reprocess if desired?  The round tables each given one of three scenarios and asked to develop a policy to storage.  The groups seemed to enjoy the exercise and no two groups came up with the same policy.  I believe there is no right or wrong answer, but your answer has to be in a written policy.

The picture below shows a group of our attendees hashing out their policy on storage.

SIIM 2009 was one of the most interesting meetings I’ve ever attended.  The changes in the format of the meeting meant that there were sessions designed to challenge the minds of all attendees, no matter how sophisticated their knowledge of Imaging Informatics.  I will try to bring a bit of the experience to everyone over the next few days.

One of the most entertaining and interesting sessions, was our keynote speaker, Henry Petroski, PhD, and his presentation “The Design and Failure of Useful Things”.  His premise was that limitations of design or objects are what trigger innovation and future development.  Read one of his more recent books “Success through Failure: The Paradox of Design” – not yet available for your Kindle or one of his earlier books “Design Paradigms:  Case Histories of Error and Judgment in Engineering” on your Kindle.   I guarantee that you will enjoy his books and his meditations on the everyday things we all take for granted. 

As I listened to him speaking, I reminisced about the early days of PACS.  The earliest project I was involved with was moving images from radiology to the ICU – we used the old coaxial cable running from a digitizer to display stations on each ICU.  This was designed to solve the limitation of having only one copy of a radiograph when it was needed for the radiologist’s interpretation and the clinician’s treatment. This was not a network as we know it today; it was point-to-point with thick cables running through the walls and ceilings of the hospital.  Most of the younger readers of this blog won’t remember the cables thicker than my thumb that could not readily bend around a corner.  In our initial system, we could not add more display stations without running more dedicated cable.  Then we had networks with multiple devices on each cable and the diameter of the cables decreased.  They weren’t fast enough, so they got faster.  Sometimes our equipment was mobile, so we had wireless – and so on.   For displays, we had a dedicated workstation for each modality because there was no DICOM and every manufacturer had a proprietary format.  Everyone knows how this limitation lead to the current state of display workstations.  Now we are facing the limitations in the display and navigation of the large image sets created by our new modalities.

The picture below is Henry Petroski on the left, Kathy Andriole, chair of the program committee responsible for the successful re-engineering of the 2009 annual meeting in the middle, and Brad Erickson, SIIM Chairman on the right.

Drs. Petroski, Andriole, and Erickson