This is bizarre. It turns out that many doctors don’t know what the real, FDA-approved indications are for the medications they’re prescribing.
This guide analyzes how artificial intelligence – including machine learning – can be used by pharmaceutical and medical device companies to improve the clinical data review and cleansing process.
In a Wall Street Journal article that discusses the risks of off-label uses for prescription drugs, the author cites a 2009 study of 1,199 physicians (mostly primary care doctors and psychiatrists) that tested their knowledge of 22 drugs. Doctors were able to correctly identify the medications’ true indications only 55% of the time. Naturally, this is a concern, since adverse events are significantly more prevalent when patients use medication for off-label uses.
On top of that, diagnoses and indications aren’t listed with prescriptions, making it hard to track off-label drug use. However, Gordon Schiff, associate director of the Center for Patient Safety Research and Practice at Brigham and Women’s Hospital (BWH) in Boston, is trying to change this.
According to a BWH press release, “the project aims to demonstrate that adding the indication to the medication order decreases medication errors, speeds up drug ordering for the prescribing clinician, improves the appropriate[ness] of drug choices and aids patients and pharmacists in education and counseling on the reasons for the drug’s prescription.” The project also entails the redesign of a model computerized prescriber order entry (CPOE) system that incorporates the medication indication.