Analytics At The Heart Of Healthcare
Under one of the industry's most ambitious IT initiatives, UPMC plans to spend $100 million over the next five years to create a data warehouse that combines clinical, genomic, insurance, financial and other information from more than 200 sources. It's partnering with Oracle, IBM, Informatica and dbMotion.
The initiative ties into the aspect of U.S. healthcare reform that aims to change how care is delivered and paid for. Under this notion of population health and accountable care, providers will get paid to keep a group of people -- say, all the employees at a company -- healthy, creating financial incentives for providers to step up their preventive care and minimize mishaps that lead to hospital readmissions.
"You can't do that without analytics," says Dr. Steve Shapiro, UPMC's chief medical and scientific officer. For example, in a recent UPMC study of the use of a particular catheter, it found that the catheter wasn't used consistently, but that on average it led to better outcomes. But the device is also more expensive, leading to the next layer of clinical data research to understand which patients benefit under which circumstances. All that data will eventually be presented to doctors to help them make decisions. (UPMC created a company, Evolent Health, with the Advisory Board Co. to market population health services and technology.)
That type of benefit-cost analysis is just the start. Researchers such as Adrian Lee will push UPMC's IT department to do much more with data, to allow personalized medicine based on a patient's individual genome.
Lee, director of the Women's Cancer Research Center at the University of Pittsburgh Cancer Institute and Magee-Womens Research Institute, is studying the role of hormones such as estrogen and progesterone in the development of breast cancer. His work focuses on genomic data unique to each patient. That research will start to home in on particular genome combinations for certain types of cancer, applying to only four or five cases even for an operation as huge as UPMC. In order to create large enough data sets to study such small subsets, UPMC has no other option but to find better ways to share data with other healthcare providers. "It will force sharing," Lee says.
Change And Challenges
UPMC is no health IT nirvana. Chief medical information officer Dr. Dan Martich acknowledges that in the rush to implement EHRs to comply with government mandates and financial incentives, "we broke a lot of things." By letting physicians cut and paste in e-records, for example, an in-hospital progress note that used to neatly sum up the patient's daily status in half a page now can run 19 pages. "What we've created, unfortunately, is an ongoing patient blog," Martich says.
How frustrated are doctors? Physicians are notorious for ignoring internal communications, so Martich had low expectations when he sent staff physicians, residents and other clinicians a survey about EHRs. One-fourth responded, and about 1,000 of them included comments.
So Martich is leading an "e-record simplification" initiative aimed at capturing the data needed for quality, compliance, billing and patient use in the most efficient way.
A lot of change is being forced on the U.S. healthcare industry, and technology lies smack in the middle. Amid those changes, UPMC is determined to find new ways to put health IT to good -- and lucrative -- use.