Referring to Andis Robeznieks' article on Using EHRs to extract data on adverse drug events:During my career as an emergency physician, I probably filled out a half-dozen adverse drug reaction reports. They involved Clinoril (Sulindac) and Captopril causing life-threatening emergencies.
The interesting fact is that I saw many more adverse drug reactions, but without my level of concern rising to life-threateningthe task of calling the pharmacy for a form, filling it out and submitting it was too great. There were many more patients in which I suspected a drug reaction, but failed to do the reporting.
For a well-designed EMR to perform such reporting electronically, at a click, is certainly within easy reach. For example, an EMR could have a control that is labeled "report as possible drug reaction." With that simple click, the patient's age, gender, co-morbidities, current medications and length of use, symptoms and exam findings could be generated into an XML report and transmitted over the Internet to a central site (such as the Food and Drug Administration) in seconds.
What is needed to make this report meaningful to a computer? A defined vocabulary that we all agree upon. Instead of "angioedema" or "hypotension," those terms plus an accompanying defined code could be sent in the report. Instead of simply supplying a drug name, its name plus NDC or RxNorm code should be sent.
The utility of this is many more adverse drug reactions could be expected to be submitted and a computer could sort through the submissions to find patterns. Remembering back to the enormous tongue swelling that occurred in several of my patients taking Captopril, it seems a computer could quickly make this association. It was a reaction that I only saw with ACE inhibitors. This technique is a no-brainer.
Currently, the continuity of care record, or CCR, promotes the use of controlled vocabularities without dictating which to use. XML schemas are simple to create and even simpler to read and write. The last step where EMRs promote granularity"angioedema" is a granule of data and not one undefined term in a whole paragraph of textis now being seen with the newest EMRs.
While putting a form online or even in an EMR is helpful, it only slightly lowers the barrier to submitting a suspected drug reaction. A far better alternative is to have one-click, computer identifiable data submission.
Matthew Chase, M.D.CEOMedtuityWesterville, OhioTo submit a letter to YOUR VIEWS, click here . Please include your name, title and hometown.