Skip to content Skip to navigation

At the Mayo Clinic, Reaching Out to Collaborate on Big Data

February 21, 2014
by Mark Hagland
| Reprints
Ryan Uitti, M.D., of Mayo-Jacksonville, is helping to lead a very broad collaborative effort using big data

Ryan Uitti, M.D. wears numerous hats. He continues to practice neurology at the Mayo Clinic-Jacksonville, and was chairman of the neurology department there until just a couple of years ago. Then, in 2012, in addition to ongoing work doing clinical research, and his continuing patient care practice, Dr. Uitti began working with the Center for the Science of Health Care Delivery, one of several special centers at the Mayo organization designed to promote specialized, important activities (the others are the Center for Regenerative Medicine, which focuses on organ transplant issues; the Center for Individualized Medicine, which focuses on the implications of genomic research for patient care delivery; and the Centers for Health Care Delivery Research, Systems Engineering, Value, Population Health, and Surgical Outcomes).

Dr. Uitti is now the director of the Center for the Science of Health Care Delivery, and is spending 40 percent of his time in that capacity. And it is in that capacity that he is leading an initiative between Mayo Clinic and Optum, an analytics company in healthcare. Together, the two organizations created Optum Labs, a collaborative research and innovation center; and just this month, several new partners joined the collaborative: the American Medical Group Association, the Boston University School of Public Health, the Lehigh Valley Health network, Pfizer Inc., Rensselaer Polytechnic Institute, and Tufts Medical Center.

That collaborative will focus on the sharing of information resources, proprietary analytical tools and scientific expertise, to help its participant organizations test new care pathways and other opportunities to drive innovation in care delivery.

And it is in that context that Dr. Uitti will deliver a keynote address entitled “Big Data, Value Analysis and Population health Science at Mayo Clinic,” at the Health IT Summit in Atlanta, sponsored by the Institute for Health Technology Transformation (iHT2), being held April 15-16. The Institute became a part of Vendome Group, LLC, Healthcare Informatics’ parent company, in December 2013. Below are excerpts from HCI Editor-in-Chief Mark Hagland’s recent interview with Dr. Uitti.

Tell me a bit about the Center for the Science of Health Care Delivery, and its goals?

Essentially, Mayo has been doing research and education as major parts of our mission. We’re celebrating 150 years this year. Thus, we are continuing our investment in R&D, which for us is research that’s never more than an arm’s length from our patients, and educating everyone involved in healthcare delivery. What’s different here is that we’ve made a strategic decision to focus our research—this is much more directed and focused. The three centers formed simultaneously—Center for Regenerative Medicine, Center for Individualized Medicine, and Center for the Science of Health Care Delivery—along with the other Centers, are all very focused.

Ryan Uitti, M.D.

In the case of the Center for the Science of Health Care Delivery, a part of our work is focused on our new initiative with partners from United Healthcare and Optum, the Optum Labs collaborative, formed two years ago in order to work with clinical and claims data. It’s a not-for-profit collaborative, and we have access to data on over 100 million lives from the past 20 years, in terms of claims data, which encompasses administrative claims data, pharmacy claims data, administrative and physician data—both procedure and diagnostic code-based data.

So this goes beyond data on Mayo patients, obviously?

Yes, that’s right; we’re contributing data from 5 million Mayo patients, obviously de-identified using scrubbers. And the data that Optum has encompasses 100 million lives. And that gives us all the ability to use huge numbers, and with the analytics programs in place, you can really ask a lot of questions about the demographics and health of people.

And in terms of our participation in a broad consortium, the thing is that, from Mayo’s perspective, we always had a pretty good feel for what happened to patients from when they entered our doors until they left Mayo, but we had a major black hole in terms of what happened to patients before and after. Now, we can identify surrogate persons within the Optum database to match up with similarly matched demographic data from the Mayo patient population, and as a result, we can get a sense of what people with a given disorder might experience. So there’s a lot of science that can be done here because of the scale. So you can say, OK, disease X, 20 years prior, is actually associated with condition Y; and because the patients involved in a particular statistical cohort might be on a particular drug, we can determine that that sequence lowers or increases their risk, or whatever.