No kidding. The label on the vial of pills I was taking for vertigo warned that they might make me dizzy. Well, at least the vertigo was under control.

Big Data And Side EffectsDespite all of the research the pharmaceutical companies do on their products, not all of their side effects are immediately identified. That’s why the Food and Drug Administration asks physicians to report side effects through a system known as the Adverse Event Reporting System. Assuming patients report the side effects and the docs fill out the related forms, the system works. Maybe not quickly, but one day – perhaps.

The good news is that technology may be coming to the rescue. According to an article in The New York Times, scientists at Microsoft, Stanford, and Columbia University have been able to detect evidence of unreported prescription drug side effect by analyzing search engine queries.

Siting a study reported by the Journal of American Medical Informatics Association, the Times said that researchers used automated software tools to examine queries of 6 million Internet users in 2010 related to paroxetine, an antidepressant, and pravastatin, a drug that lowers cholesterol. What did they find? Evidence that a combination of the two drugs caused high blood sugar.

Here’s how they did it. The researchers looked at 82 million searches for drug, symptom, and condition information. They identified searches for terms paroxetine and pravastatin, as well as queries that included both drugs. The Times said they computed the likelihood that users in each group would search for hypoglycemia and about 80 related symptoms.

That data showed that people who searched for information on both drugs were more likely to search for terms related to hypoglycemia. And people who searched for both drugs did so in a short period of time.

According to the Times, the researchers said they were surprised by what they called the signal strength that they detected in their analysis, and concluded that there “is a potential health benefit in listening to such signals.”

Thinking about the upside of this kind of research makes me dizzy. But then maybe that’s just my pills.

The New York Times article was published on March 6. At the time, the American Medical Informatics Association Website didn’t offer any information on the study.