As part of the National Science Foundation's "Big Data" initiative (.pdf) UC Berkeley was recently awarded a $10 million grant. There, researchers will create an open-source platform to collect, organize and make sense of vast amounts of data, including information recently made public by the Centers for Disease Control and Prevention and several other federal agencies.
In a new iHealthBeat report from Deirdre Kennedy, health-care experts, including Stanford's Atul Butte, MD, PhD, discuss (.pdf) the UC Berkeley project and the opportunities and challenges of mining big data for health care and scientific research. About the potential of using electronic health-care records for research purposes:
Butte says it's possible to mine EHR without violating patient privacy. Stanford did just that. The university released a widely publicized study this year that found women reported higher levels of pain than men. Researchers gleaned that information from thousands of patient records by just combing through one piece of data — how they rated their pain when nurses asked them.
"It was the largest study ever for pain. That data was just sitting there in the repository waiting for someone to do something with it," [says Butte.] "It didn't even have to be publicly available. Data is on its way to getting more and more public. I actually think the new challenge is what do we want to ask of that data?"