How Physicians Are Driving Data Analytics Advances in Health Care

Don Tennant
Slide Show

Five Ways Automation Speeds Up Big Data Deployments

You’d think physicians would have enough to worry about, with all that medical stuff they have to deal with. But it seems that in addition to the stethoscope draped around their necks, a lot of them are wearing a data analytics hat.

That’s only one of quite a few takeaways I gleaned from a fascinating conversation with Mike Berger, vice president of enterprise analytics at Geisinger Health System, a hospital system in Danville, Pa. I met with Berger at last month’s Teradata 2014 Partners conference in Nashville, and in addition to sharing his experience as a Teradata customer, he shed an informative light on the role medical practitioners are playing in the advance of data analytics in the health-care sector. “I know within my organization, there’s a lot of pressure to jump to this Hadoop thing—jump to Big Data—because it’s new, and it’s sexy,” Berger said. “The Mayo Clinic is doing it, so we better be doing it, too.” And where is that pressure coming from? Berger had no problem identifying the source:

It’s coming from new clinical leadership that is a little younger, and a little more fearless, which is fun—it’s not a bad thing, it’s a good thing. They want to jump into the sexy stuff that’s getting all the profile  … We have some very, very smart doctors who do more [than just clinical work]. You’ve got a handful of physicians who are given very IT-esque leadership functions. CMIO—chief medical information officer—is a very common title now. It’s a clinician who’s a geek, and who loves data. For the past 10 years, they’ve been very much focusing on implementing electronic medical records [EMRs]—that was their core function. Now a lot of those systems are in, so the evolution of that is, what are you going to do with it? What’s a secondary use for all of that data? So now, CMIOs are getting into analytics and data. Before, it was just transactional applications—some of them are still buried in EMR worlds. But I think at the more progressive health systems, the EMR is commoditized. You paid a lot of money for it, and you still will. But IT can handle it. Now, it’s all about how you’re going to work with the data in a useful way.

Berger said the challenge is determining the right way to transition from the traditional data warehouse to the world of Big Data:

Moving out of one, into the other, doesn’t really happen—you need to maintain both at the right balance. Invest in what’s new, but maintain this classic stuff, because it probably will be there forever. Maybe not—maybe there will be something that really does allow you to get rid of it. But you have all these skills and competencies and processes in data warehouse—actionable analytics that have been embedded into workflows. To pull all that out, and redo it, just to do it in a Big Data platform, doesn’t make sense. I think people feel like it’s all or nothing, and I think the Teradata-articulated strategy is that you need some big data, and you need some classic relational data warehouse. The competitive winners are going to be the ones who figure out how much you need of both.

I mentioned to Berger that at a media briefing the previous day, Teradata officials noted that everything Teradata offers can be offered in the cloud. I asked Berger where, if anywhere, the cloud fits into his strategy with respect to Teradata. He said it’s too little, too late:

We already invested in all the physical hardware. If I was going into this from scratch, that could potentially be part of my strategy. … When we went from a [Teradata Data Warehouse Appliance] 2650 with three nodes to three extra nodes, that could have been as easy as somebody turning a dial, if it was all in the cloud. The challenge is that our hardware is still great, and there’s no secondary market for this stuff. So we’re stuck with what we have. When those are ready for retirement, I think it’s a viable consideration. But we’re still a little bit of a ways away from that.

Health Care

Finally, I noted that the Teradata officials had spoken fairly extensively about having adopted a range of open-source technologies, and I asked Berger how meaningful that is to him and what he’s doing at Geisinger. His said open source is a tricky proposition in health care:

It’s not really applicable to what we’re doing yet. Health care is extremely concerned about [ensuring that] all of that patient data is sitting on a network that’s extremely locked down. Every application that’s running on the network needs to have gone through an information security office audit process that’s very, very exhaustive, and very painful, and very detailed. That doesn’t work with open source. Open source is a bunch of techies that go to the Web, and they download stuff, and they install it. You can’t do that anyplace on our network where there’s patient data. They have quarantined areas where the high-performance computing cluster lives, where they’re not allowed to see patient data. So everything we send to it has to be on a very specialized, dedicated pipe. … They can’t connect the way you would want to connect to Teradata. If we were Nationwide Insurance or eBay, where IT probably has a lot more free rein to do downloads, we could install things and try them. I can’t do that. I think that hurts us in our ability to just try things. The vendor is not going to put in the time that it takes to do their part of that audit, unless they feel like they’re going to get cash for it. And in open source, it doesn’t work, because oftentimes inherently, it’s not secure. They haven’t figured out the bugs, and they want you to try it so maybe you’ll help figure out the bugs for everybody else. Health care doesn’t want to be the place to do that. So it’s a Catch-22 for us. We haven’t figured that out yet.

A contributing writer on IT management and career topics with IT Business Edge since 2009, Don Tennant began his technology journalism career in 1990 in Hong Kong, where he served as editor of the Hong Kong edition of Computerworld. After returning to the U.S. in 2000, he became Editor in Chief of the U.S. edition of Computerworld, and later assumed the editorial directorship of Computerworld and InfoWorld. Don was presented with the 2007 Timothy White Award for Editorial Integrity by American Business Media, and he is a recipient of the Jesse H. Neal National Business Journalism Award for editorial excellence in news coverage. Follow him on Twitter @dontennant.

Add Comment      Leave a comment on this blog post
Nov 20, 2014 4:42 AM James James  says:
Good article, data analytics should be a key weapon in every company’s fraud protection arsenal. A well designed data analytics strategy can strengthen your internal controls and proactively manage risk in a globalized economy. I work for McGladrey and there's a whitepaper on our website that offers good information on the above topic that readers will find very informative. Reply

Post a comment





(Maximum characters: 1200). You have 1200 characters left.



Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.