Analytics: making sense of it all
April 30, 2015 in Medical Technology
The U.S. Department of Health and Human Services kicked off the year with a January announcement it called “historic”: a doubling down on the shift to value-based care that paves a more aggressive path away from paying for volume.
[See also: CBI resources improve care, lower costs]
HHS wants to tie 30 percent of fee-for-service Medicare payments to models such as accountable care organizations and bundled payments by the end of 2016 – and hopes to reach 50 percent by 2018.
The agency feels comfortable setting such explicit and ambitious goals for Medicare’s embrace of alternative payment models thanks in large part to the fact that, “through the widespread use of health information technology, the healthcare data needed to track these efforts is now available,” according to HHS.
[See also: Analytics means we 'roll up our sleeves']
That means many providers have just a couple years to get their act together and start putting their data to work in service of more efficient care for healthier patient populations. Essential to that? Clinical and business intelligence technology.
Travis Turner, president of clinical integration at Fredericksburg, Va.-based Mary Washington Healthcare is one of those working hard to bring analytics to bear on both sides of the payer-provider divide.
Mary Washington recently applied population health technology from MedeAnalytics to data from some 3,000 employees in its self-insured health plan. The goal is to use its insights to help improve quality reporting, risk stratification and more. And as a newly-minted member of the Medicare Shared Savings Program, Mary Washington plans to broaden its use of the tools as it embraces accountable care.
“At the end of the day, it’s got to be swift, it’s got to be quick, but it’s got to be done right,” says Turner of HHS’ aggressive new pivot from fee-for-service to value-based contracting. “For us, this is that first step in being prepared for what’s coming.”
Key to MWHC’s improvement strategy is to align its quality metrics with those of its affiliated physicians, he says. Intelligence tools, helping aggregate and normalize clinical and financial data, can help the health system self-evaluate, manage patient populations and spot instances worthy of early clinical interventions.
There are challenges aplenty. With more than 400 physicians in 100 practices, “you can imaging the number of electronic health records,” says Turner. (Hint: more than 20.)
“We’re challenged with working with these vendors to extract a lot of the quality data that’s required to participate with the MSSP program,” he says. “Every vendor generates PQRS, meaningful use, HEDIS type metrics in a different format. We’ve really got to boil that down. And it becomes a big barrier to moving forward when we’re trying to delineate across 20 different vendors.”
Still, MWHC has seen some gratifying early wins, says Turner, including some decrease in hospitalizations of patients spotlighted predictive modeling. “We’ve seen avoidability in the ED go down, we’ve seen ED per 1000 go down,” says Turner.
And it’s still early yet. It may seem like a modest start, focusing on a few thousand members of a self-insured health plan but, he says, “this is how you start with dipping your toe into population health management.”
At the 2015 HIMSS Annual Conference Exhibition earlier this month, there were dozens of keynotes and education sessions focused on analytics and population health.
From grappling with the data-intensive challenges of CBI, tips on building organizational support for analytics programs, putting clinical information to work for risk-adjusting bundled payments and taking advantage of emerging strategies such as geographic information systems, applications of IT for population health were front and center in Chicago.
But while more and more providers are showing the willingness and the wherewithal to embark on more complex analytics initiatives – their bottom lines depend on it, after all – there are still some roadblocks preventing more enthusiastic adoption.
Luke Shulman is principal consultant at Burlington, Mass.-based Arcadia Healthcare Solutions, which develops data aggregation and analytics tools for EHRs and beyond. One big challenge he sees in the field is related to user experience and design.
“Early on I was working with a client who was entirely on green screens – this was an insurance group – and even though there’s absolutely no user experience through a green screen they could finish any task in 30 seconds,” says Shulman, by way of example.
When they eventually replaced that with “a system of windows and clicks, that looked better, everything got slower,” he says.
“I think there’s a lot of history that’s built up from the forms and flowsheets and templates that is ubiquitous within the EHR that is holding us back right now,” says Shulman. “We fully realize the importance of getting all that stuff structured and formatted. In our other work, which is to get these new analytics and population health tools used, we see it becoming a huge barrier.”
Another hurdle he sees is the same one so much of healthcare is trying to fix: interoperability.