mead cohen berger shevtsova garfinkle michta grygiel blankenhorn
Delivering Health
The Downside to Big Data in Health Care

Big Data’s march through the health care industry really seems to be making news this week. Take, for example, this Wired profile of Modernizing Medicine, a company that aggregates data from “knowledge gathered from roughly 3,700 providers and more than 14 million patient visits.” Health care providers can then draw on this data to make better treatment decisions. Powerful search algorithms deliver precisely targeted results.

Sounds good on the surface, but some of the details about the system raise questions. This passage jumped out:

For now, the recommendations from Modernizing Medicine are largely based on what is most popular among fellow professionals–say, how often doctors on the platform prescribe a given drug or order a particular lab test. 

It’s in fact very popular for doctors to over-treat, or to prescribe more expensive versions of generic drugs. Systems like these could end up propagating these bad practices on a much larger scale. This may be fixable: improvements to the system are coming later this month that should allow doctors to tap IBM’s Watson supercomputer to cross-check treatments against the results of clinical trials, research, and even reported outcomes from other doctors using the service.

But there’s a bigger problem. As we noted about a year ago, the temptation with big data is to centralize care decisions in a federal health care bureaucracy. The centralizing vision here is clear: a national health care system overseen by powerful teams of experts equipped with best practice data from supercomputers appeals very strongly to a certain kind of wonkish techo-solutionist. If bureaucrats over at CMS can get access to what the data says are “best practices” they can increase their ability to micromanage doctors.

If we ever solve our health care policy problem, it will be in large part due to technological innovation. Mobile tech could help create cheaper ways to deliver care to patients; certain kind of big data systems could empower nurse practitioners to provide more primary care on their own. Tools that monitors our vitals could be a big boost to preventive medicine. But these are all good because they decentralize care, empowering care providers and patients to get better, more personalized treatment. Companies like Modernizing Medicine, insofar as they facilitate decision-making that balances health outcomes and cost, are all to the good. But by themselves, they’re probably not silver bullets, and if they inadvertently concentrate more health care decision-making power in DC, they may do a lot more harm than good.

Features Icon
show comments
  • rheddles

    If we ever solve our health care policy problem, it will be in large part due to technological innovation.

    No. No. No. If we solve our health care policy problem it will be due solely to making individuals the purchasers of health care services.

  • Corlyss

    As the Snowden fiasco has shown, even the most secure site on the planet couldn’t protect itself from illegal intrusion by a mole, while almost daily reports of hackers hacking into sophisticated banking systems revive cautions about banks and retail credit card purchases. Given the fact that narcoterrorists are now switching to Medicare and health care fraud because the risks are so low compared to the rewards flowing from a slovenly administrated system, am I the only one concerned about putting all this whiz-bang big data into a single repository?

© The American Interest LLC 2005-2016 About Us Masthead Submissions Advertise Customer Service