The answers may be lurking there in the data, if you just look. But there are risks to looking, too.
Jenny Frankovich from Stanford University's Lucile Packard Children's Hospital tells NPR's Amy Standen about the time she had a young patient with lupus who was in serious condition as her kidneys were beginning to shut down. In similar cases, Frankovich knew that some children also developed clots, but giving an anticoagulant is also risky.
"There wasn't enough published literature to guide this decision. And really, the best route was to not do anything," Frankovich tells Standen.
But then Frankovich had the idea to search through her hospital's electronic medical records for the number of lupus patients they'd had and the number of them that had developed clots. She did this, Standen notes, with a few keystrokes. And based on the data Frankovich dug up, the clinical team decided to give the patient the drug. The patient didn't develop a clot and, to Frankovich's knowledge, is now doing fine.
This, Stanford's Atul Butte tells NPR, is an example of how medicine and medical research may soon look. "We already have the measurements and the data," he says. "The struggle is to figure out what do we want to ask of all that data?"
But Standen notes that Frankovich's hospital, rather than adopting such an approach, has actually moved away from it. The hospital says the system isn't ready to be used like that — what if, Standen says, there were errors in record transcription or if Frankovich had used the wrong search terms?
"I mean, for sure the data is there, right?" Frankovich says. "Now we have to develop the system to use it in a thoughtful, safe way."