Health care data analytics, as a technology, is expanding on two fronts: the clinical data side and the supply chain side. Health systems are investing in the technology in both areas for one reason -- cost reduction.
For several years, the health care industry has been moving to standardize inventory on GS1 barcodes,
As long as the ACO is a community-based concept then the exchange is fundamental.
At minimum these systems can aggregate data after the money's been spent and show where improvements can be made. Data analytics systems also can offer decision support, advising employees when ordering products that are off-contract how much more they'll cost. In some cases, organizations can restrict off-contract ordering.
But the emerging accountable care organization (ACO) payment model could lead to explosive growth in health care data analytics technologies in 2013. Hospitals and community health systems are hard pressed to perform two data-aggregation tasks required for ACO participation: computing the total cost of a medical procedure from diagnosis to final follow ups -- for example, physician visits, radiology imaging, surgery and hospital stay, and physical therapy for a hip replacement -- and population health management. In that example, reimbursements are pegged to improving the A1c level of all the diabetics in an ACO, regardless of primary care physician or which facilities within the ACO a patient visited.
Collecting, standardizing and running analytics on clinical data sets isn't for the faint-hearted. In fact, some might see it a job for a health information exchange (HIE), which aspires to traffic in data from all providers in a particular region. In Texas, Sandlot Solutions, a private HIE owned by the 600-physican North Texas Specialty Physicians, has launched its own ACO called Plus.
Another, Maine HealthInfoNet, is collaborating with Arcadia Solutions to create a data warehouse that ties the statewide HIE's clinical data to payer claims data. After a Bangor Beacon Community pilot project to show that the new system can pair clinical data with claims cost data and display it in a physician-readable form to inform care decisions, HealthInfoNet plans to roll this service out system-wide next year. In this way, the HIE hopes to give physicians a general idea of what a procedure's soup-to-nuts total cost (as well as quality and outcomes data) will be in ways that are impossible to automate now.
The outcomes data has been particularly tricky to compute, said Devore Culver, HealthInfoNet's CEO. "A lot of the analysis based on cost data [alone] infers outcomes," Culver said. "The fact that you would use more or less units per event of care may or may not truly represent a better or worse outcome if you don't see the clinical data. That's what we're trying to do here, begin to look at the question of whether having the two coordinated together on an event-of-care basis, and a longitudinal basis, leads to some enhanced view of what can be done better."
Some of the next analytics projects the HIE is working on include "near real-time" notification of physicians when patients are treated elsewhere, and analyzing data to find repeat admissions to multiple emergency rooms in short timeframes. Some ACO proponents said all analytics need to be run in-house on their own data warehouses and question the trustworthiness of third parties such as payers, for example, to assist in such a monumental data project.
But Culver sees the mix of provider and HIE aggregating and running analytics on clinical data as a workable partnership. "As long as the ACO is a community-based concept -- meaning the patient has the [choice] to go anywhere he or she wants for their care -- then the exchange is fundamental," he said. "The minute we slip into what I would call channeling, which sort of feels like the bad old days of managed care, then the exchange is, perhaps lessened in its value. But not diminished to the point where it isn't valuable, because you're still going to want to see comparative [costs] against the community at large."
Michael Gleeson, vice president of quality improvement and analytics for Arcadia Solutions, also sees the HIE-ACO data aggregation and analytics partnership as a way to fast-track the implementation of these complex systems. "A lot of the other Pioneer ACO clients are spending months and years trying to build out the capability to support those initiatives. With analytics on top of the HIE, it's something you can get a really good start on, almost immediately."