Algorithms mine articles to predict bad drug reactions

Is there a better way to predict serious side effects to drugs before FDA-ordered recalls for treatments like Vioxx. Researchers from the Santa Monica, CA-based think tank Rand showed that mining articles from PubMed could help identify adverse events caused by prescription drugs. They developed an algorithm that was used to mine articles published between 1949 and September 2009, searching for ones that identified one of 38 drugs and one of 55 side effects. The group, whose research was recently published in the Journal of the American Medical Informatics Association, found that in some cases the articles could reveal drug safety concerns ahead of actual FDA warnings. Story

Suggested Articles

Through a new online tracker, AllTrials names sponsors who fail to report clinical trial results on time per the FDAAA Final Rule.

Elligo Health Research will test a harmonized real-word data model built by the FDA.

Just about a year after filing its IPO application BGI hits a big wall with a regulatory review.