Just because the Food and Drug Administration (FDA) approves a drug doesn’t mean it is safe. Sometimes adverse events, side effects, allergic reactions, or unsafe drug interactions occur after approval and it can take years before anyone ever realizes there’s a problem. But a new algorithm designed by United States scientists is helping researchers pore through databases with mountains of drug interactions and has yielded thousands of previously unknown side effects. The work, published this month in the journal Science Translational Medicine, is designed to dig through the countless adverse events reported to the FDA’s MedWatch database.
When the FDA reviews drugs for approval, it bases its decision on the results of clinical tests, which helps to determine both the safety and the efficacy of the drug. They are often done in controlled settings using carefully defined criteria to determine which patients are eligible for enrollment. But once a drug is approved, previously unknown side effects can surface. These are typically reported to the FDA and become a part of the agency’s MedWatch database.
Side effects can be culled from this data, but drug interactions are much more difficult to spot. And that, researchers say, is the beauty of the new algorithm.
The algorithm was crafted to remove bias. Then the team used this method to compile a database of 1,332 drugs and possible side effects that were not listed on the labels for those drugs. The program found an average of 329 previously unknown side effects for each drug – far more than the average of 69 side effects listed on most drug labels. A similar database was established that focuses on interactions between two different drugs.
Researchers who developed the database say they plan to present their results to the FDA and suggest the algorithm could be used with existing drug-surveillance programs to remove bias.