Integrating Images with EHR at Cleveland Clinic: Lessons Learned
Three years ago, Cleveland Clinic integrated radiology images into its Epic enterprise electronic health records system.
During the past 18 months the organization has integrated many other types of images into the EHR as well, using the Imaging Clinical Information System platform and vendor-neutral archive of Agfa Healthcare.
But the task is not done and no one can say when it will be, says Cheryl Petersilge, M.D., chair of radiology at Hillcrest Hospital in Ohio and medical director of the imaging integration initiative at Cleveland Clinic. More images from more places across the 10-hospital delivery system keep popping up. Every time we turn around, theres a new place asking us to integrate their images.
Images and reports currently in the archive that physicians query via the EHR come from radiology, ophthalmology, endoscopy, wound care and many different types of ultrasound systems. Who knew how many images ophthalmologists create? Petersilge asks. Theyre among our biggest customers and I would never have guessed that.
What is not in the integrated imaging/EHR system yet are cardiology images. Cleveland Clinic has a large cardiology system across the enterprise with its own separate image management system and migrating that data is a large project to come, with no real timetable yet established.
The integration initiative started with radiology and an early lesson was the need for having an indexing strategy so the archive wouldnt simply house a bunch of unsearchable studies, Petersilge recalls. An organization needs a naming strategy with standardized ways of naming specific types of studies across the enterprise. Building that strategy and the workflow processes to support the moving of information between the EHR and the imaging system platform put the project on hold for several months.
Today, the mature service is in high demand among physicians, Petersilge says. They understand the benefit of having all the clinical and imaging information available via one conduit at the point of care, with no bouncing between computers. To me, thats really what EHRs are supposed to be, she adds.
Having images immediately available has brought a benefit that wasnt anticipated, Petersilge notes. Image data is absorbed faster by the brain than text, and more easily retrieved from the brain, which improves the quality of clinical decisions and overall care, she believes.
Other lessons from the project include:
* When embarking on the journey, CMIOs, CIOs and physicians have to think about images beyond radiology and cardiology, and understand that the project will balloon considerably.
* A governance mechanism is necessary to standardize indexing/naming strategies, and a robust physician advisory group must have influence. Only your consumer can tell you whats really important to them, Petersilge says.
* The project isnt just about integrating the image management system, archive and EHR; its also about the workflow for all constituencies.