Microsoft's web-search mining uncovers side effect of drug pairing
Web search records could provide early signals of harmful drug interactions. And thanks in part to consumers' inability to resist going online to learn about their symptoms, researchers found a wealth of data to triangulate the interactions of two widely used drugs and an undesirable side effect.
The study united efforts from researchers at Microsoft ($MSFT), Columbia University and Stanford University. The collaborators tapped records from 2010 activity on Google, Microsoft and Yahoo search engines with an automated tool. The approach confirmed that those likely to be taking the antidepressant Paxil or paroxetine and the heart pill Pravachol or pravastatin were twice as likely to search for symptoms of high blood pressure than those not taking the two drugs together, the New York Times reported.
The results, which appeared March 6 in the Journal of the American Medical Informatics Association, basically confirmed the findings from a previous study from Stanford's Dr. Russ Altman. In that previous study reported in 2011, the group discovered the increased risk of hyperglycemia on the antidepressant and cholesterol med from FDA records provided by physicians. Yet the new web-search approach, which the Times likened to Google Flu Trends, could find harmful drug-to-drug effects earlier than the FDA system.
"We don't know how much potential it has," Hojjat Salmasian, a Columbia bioinformatics researcher not involved in the study, told ScienceNow. "But based on the results that have been published so far in this study and other similar studies, it seems like this is a very important field to explore."
Altman describes the findings as the tip of the iceberg.
"I think there are tons of drug-drug interactions--that's the bad news," Altman told the Times. "The good news is we also have ways to evaluate the public health impact."
On a separate but related front, the FDA has been working on speeding up the agency's ability to spot adverse events from drugs with the Sentinel effort. Meanwhile, non-agency scientists and companies are racing ahead with their own side-effect trackers.