VistA Alert System due for an Upgrade

The Department of Veterans Affairs is looking for vendors to help it modify VistA, its electronic health record, so that it can better track physicians’ responses to patient care-related alerts that have been generated by the system.

Currently, a component of VistA known as the computerized patient record system (CPRS) generates these alerts, primarily in response to abnormal diagnostic test results. CPRS also captures whether or not physicians click to acknowledge receipt of the alert, but the information stream runs dry after that.

VA officials want CPRS to track physicians' follow-up actions in response to the alert. Such activities might include notifying patients about the abnormal findings, repeating the test, or ordering additional tests. The officials also want to track timeliness of follow-up by physicians who have received an alert.

The officials anticipate the work will require a redesign of the CPRS interface. In their RFP, they also indicate they are looking for vendors to provide source code for three functional and documented prototypes over the next 16 months.

Background
VistA is among the most comprehensive, respected and widely deployed legacy EHR systems in extant. It was developed using the MUMPS language/database.

The current plan to design an alert-driven physician behavior tracking system comes just months after the release of a study by Hardeep Singh and colleagues. They looked at what happened when VA physicians were prompted by an alert that had been triggered by abnormal findings on a medical imaging study.

It turned out that more than 18% of the alerts were never even opened, and nearly 8% of the affected patients did not get timely follow-up. In many cases, the lack of timely follow-up was associated with a deterioration in the patient’s condition.

Needed: A Clinical Decision Support Strategy
VistA has its own problems, being MUMPS-based and all (What? You thought the VA should be able to fix its own EHR? Or at least have a vendor that can help ‘em out?).

But the challenges identified by Singh are just as acute for all EHR system architects. Alerts after all, are just one kind of clinical decision support (CDS) tool. Reminders (for example, to order a mammogram on a 65 year old female that hasn’t had one recently) are another kind, as are structured order forms and clinical guidelines—at least to the extent they can be presented to clinicians during the course of their work.

In her recent review of clinical decision support for the Agency for Healthcare Research and Quality, Eta Berner of the University of Alabama wrote:

“…although EHRs…can improve accessibility and legibility of information, it is unlikely that there will be major improvements in the quality and cost of care from the use of health IT without proper implementation and use of CDS.”

This easier said than done. The set of all possible alerts, reminders and so forth is infinite, and a decision to add any one of them into a physician’s normal cognitive processes and workflow risks causing frustration. And should too many alerts be added, the comfortably numb physician might just ignore them all. So how do you choose which ones implement? Should you let an “administrator” turn alerts on and off?

By the way, when alerts interrupt normal cognitive patterns, they can have the perverse effect of creating new and different errors. For example, imagine trying to type a blog post while being forced to check your email inbox every single time you heard that “ding.” That blog post is going to take forever to write, and it’ll have more errors in it than one that was written with the benefit of some peace and quiet.

So no, this is not easy, but as per Dr. Berner, the quality-enhancing potential of EHRs is on the line here.

Good luck to those VistA consultants!

Glenn Laffel, MD, PhD
Sr. VP Clinical Affairs, Practice Fusion