The practice of using a medication for something other than its primary purpose is a common one, and off-label prescriptions can be powerful weapons in the fight against certain cancers and other difficult-to-treat conditions. But while it’s easy to track which pills are being dispensed, information on the disease the drug is intended to treat is more difficult to glean from unstructured EHR notes. Researchers at Stanford University are trying to remedy that gap in clinical understanding by detailing a new method of scraping EHR data for valuable information on off-label usages.
A study from 2006 indicates that while 21% of all prescriptions are for drugs intended for off-label use, fewer than 30% of these decisions are supported by peer-reviewed data. While there is great potential for many pharmaceuticals to be used in beneficial new ways the original developers couldn’t even imagine, there is also a risk of harm to patients who may not react as a provider could hope.
To prevent adverse events and to track the novel ways certain medications are helping patients, the research team combed through the Stanford Translational Research Integrated Data Environment (STRIDE), a collection of data on more than 2 million patients treated in the Stanford health system over the past decade. By extracting deidentified medication data and pulling out associated medical conditions, the team was able to produce a list of 403 distinct uses for off-label drugs.
The researchers then correlated the uses with the cost of the drug and the potential for high-risk side effects, which separated the medications into two large categories. Low-cost, beneficial drugs were flagged for further clinical trials that may lead to additional FDA approvals. High-cost, high-risk drugs are prime candidates for reevaluation by physicians and possible action from regulatory agencies.
Pharmaceutical companies already spend billions of dollars each year developing, marketing, and gaining FDA approval for medications targeted for specific diseases, but stand to make even more money by expanding the range of uses for one medication that may take a decade or more just to bring to market. Drug companies may invest in similar off-label tracking systems to see exactly how their products are used and how to spot patterns that may encourage new discoveries.
“Just as detection of abnormal spending is now a routine feature of credit card services, someday off-label use detection could be a routine part of health-care systems,” says Nigam Shah, MBBS, PhD, lead author of the study. “The combination of electronic health records and molecular evidence can make a stronger argument for agencies to fund clinical trials.” Source