Most drugs have certain side effects and some of the effects present themselves only when used in combination with other drugs. While some of the rare effects are detected only after it is available in the market and used by large number of people. Hence, some adverse effects are often missed during clinical trials.
In an attempt to speed up the detection of adverse effects of the drugs, a team of researchers from Microsoft and Stanford University analyzed the queries made to popular search engines including Google, Bing and Yahoo by 6 million internet users.
Analysts looked for queries on an antidepressant paroxetine and the cholesterol suppressant pravastatin. It was found that 10% of the people who had searched for details on both drugs had also looked for information about increased blood sugar, hyperglycemia. This shows a combination of these two drugs caused hyperglycemia.
Scientists had also looked at 62 other combinations of drugs half of which were known to cause hyperglycemia. They realized that the data mining process showed 81% accuracy in predicting whether a drug combo had adverse reactions or not.
Researchers thus conclude that the new field of Bioinformatics holds tremendous potential and remains to be explored.