Researchers at several prominent New York City-based hospitals are using patients' emergency contact information to build family trees in hopes of using the data to improve care—but one bioethicist says the practice is unethical, Heather Murphy reports for the New York Times.
How the data mining works
Nicholas Tatonetti, a data scientist at Columbia University Medical Center, said his team has sought new ways to study genetics' role in adverse reactions to medications in an effort to "design safer drugs for patients." But they lacked a large data set that contained multiple generations of detailed medical histories, as most heritability research has relied on the studies of twins or of families that carried a specific disease.
According to the researchers, EHRs present a new way to research heritability of diseases and medication responses among a diverse set of patients. The researchers detailed their approach in a study in the journal Cell.
First, the team mined anonymized data from about 2 million patients at Columbia University, Icahn School of Medicine at Mount Sinai, NewYork-Presbyterian Hospital, and Weill Cornell Medicine. The researchers searched for patients whose emergency contacts had also received care at one of the institutions and extracted diagnostic codes and other health information from labs and bills, such as height, body mass index, and blood serum levels.
Murphy reports that the data can be used to examine why some treatments, such as the cancer medication mercaptopurine, are highly effective in some patients but toxic in others.
"It's a way of looking at genetics but without having any genetic data," said Tatonetti. He added that if more data can be gathered, "we're going to be able to advance medicine much more over the next 50, 60 years by observing patients instead of just guessing."
Why some ethicists are concerned
But some bioethicists argue that the approach is unethical because patients—and emergency contacts—have not explicitly consented to having their information used for research.
Craig Klugman, a professor at DePaul University, called the practice "really problematic" from an ethical standpoint. Klugman compared the researchers' use of emergency contacts to Cambridge Analytica's use of personal data on Facebook through a personality quiz, but he argued that this practice is actually worse. "On Facebook, there is no expectation of privacy. It is clear that we pay for this free service by giving away our information," he said. "In medicine, there is a strong expectation of privacy going all the way to the Hippocratic oath."
Tatonetti acknowledged that more explicit consent forms could be valuable in the future, although researchers argued that the broad research consent forms that most patients complete at the doctor's office were broad enough to cover this use.
Klugman also raised a separate concern: The patients whose records are used in the research won't actually be notified of any findings due to a privacy agreement. Klugman said that the idea that "we're going to find information to help your health but we're not going to give you that information" undermines the purpose of the research.
Tatonetti noted that his team went through a thorough vetting process with Columbia's review boards, and he added that he hopes future participating institutions will ask patients to sign agreements allowing researchers to share their findings with them (Murphy, New York Times, 5/17).
Access our health care cheat sheets on top-of-mind legal landmarks
With the new tax law, antitrust laws, HIPAA, and countless others, the health care landscape has become an alphabet soup of legislation. To help you keep up, we've created a series of cheat sheets for some of the most important—and complicated—legal landmarks.
Check them out now for everything you need to know about the Affordable Care Act, antitrust laws, fraud and abuse prevention measures, HIPAA, MACRA, and the two-midnight rule.