EHR data spurs real-time evidence-based medicine

This time, House, M.D. fans, it was lupus. The article “Evidence-Based Medicine in the EMR Era” published in the Nov. 10 issue of the New England Journal of Medicine might have read like a House television script, but it was a real-life glimpse of what the most optimistic health IT advocates are hoping will become commonplace in U.S. health care: Mining EHR data to arrive at treatment decisions.

Stanford pediatricians were presented with a 13-year-old lupus patient who they thought should be given anticoagulants, but were resistant because, typically, they aren’t given to patients that young — even when critically ill. They didn’t have “the book” to rely upon in this case, and a consult with in-house experts proved “fruitless and failed to produce a consensus,” the authors wrote.

So in four hours, they did a retrospective study of similar patients in the hospital’s data warehouse, called the Stanford Translational Research Integrated Database Environment (STRIDE), and decided to move ahead with the treatment based on the previous results of 98 pediatric lupus patients treated between 2004 and 2009. While the patient responded favorably, the authors said they will never know if they made the “correct” decision, but they did know that — in absence of randomized trial research to support their decision — they acted on the evidence of the best data available, coupled with their experience.

“Our case is but one example of a situation in which the existing literature is insufficient to guide the clinical care of a patient,” the authors wrote. “But it illustrates a novel process that is likely to become much more standard with the widespread adoption of EMRs and more sophisticated informatics tools.”

Health IT, for the win! While the clinical decision support involved in the anecdote wasn’t quite real-time, it’s still a great story. It’s definitely potential fodder for an upcoming plot starring Hugh Laurie’s limping misanthrope Dr. Gregory House.

No one, yet, is claiming that retrospective EHR data will replace the long-relied-upon model of randomized clinical trials. But there are so many illnesses and out-of-the-ordinary cases that don’t fit the mold of standard treatments, such as the teen lupus patient this article’s authors were faced with treating, quickly. It’s just one anecdote, but this case proves that — used judiciously — the data stored in EHR systems can be more than just a paper-saving technology; they can save lives. It gives CIOs tasked with implementing EHRs and building data warehouses a human face to attach to all the seemingly soulless bits and bytes.