Top Ten Tech Trends: Analytics for Readmissions Reduction Work: A Long, Complex Journey Ahead | Healthcare Informatics Magazine | Health IT | Information Technology Skip to content Skip to navigation

Top Ten Tech Trends: Analytics for Readmissions Reduction Work: A Long, Complex Journey Ahead

February 18, 2014
by Mark Hagland
| Reprints
The scramble is on to leverage the IT tools that can help providers effectively reduce avoidable readmissions

The mandate embedded into the Affordable Care Act (ACA) for hospitals to reduce their avoidable inpatient readmissions has proven to be a major—if predictable—shock to the U.S. healthcare system. Major, because already last August, the Centers for Medicare & Medicaid Services (CMS) notified the executives of 2,225 hospitals across 49 states that they would lose somewhere up to 2 percent of their Medicare reimbursement in 2014, based on CMS finding too many avoidable readmissions for heart attack, heart failure and pneumonia patients at those facilities (with hospitals losing forfeiting up to 1 percent of Medicare reimbursement in fiscal year 2013), with 18 U.S. hospitals to see the maximum 2 percent reduction in Medicare reimbursement in 2014. Predictable, of course, because such measures had been cleared by federal authorities as punishments for poor performance when the ACA was passed back in March 2010.

So if hospital executives have known for nearly four years now that these penalties were potentially coming to their organizations (and indeed, the first penalties were already applied in Oct. 2012, based on 2008-2011 data), why does avoidable readmissions work remain one of the biggest challenges facing healthcare leaders now? A complex knot of reasons is involved, say experts, but fundamentally, it all boils down to this: under the fee-for-service reimbursement system, there had never been (prior to 2010) much motivation for hospital executives to try to figure out, let alone fix, the “problem” of avoidable readmissions for common diagnoses, given that all the incentives under FFS payment have always been towards filling beds. As a result, development of both analytics tools and analytical processes, around this issue has until recently been severely delayed.

Now, of course, there is a scramble underway to figure out how to harness the analytics and report-writing tools and the data warehouse infrastructure, to reduce and eliminate avoidable readmissions. On the Medicare side, hospitals are receiving publicly reported data from CMS to help them sort things through, while on the private side, health insurers are beginning to share claims data with hospitals and physicians, while ramping up their plans to emulate the ACA readmissions reduction program in their contracts.

Meanwhile, the stakes continue to be raised on the federal side, with CMS preparing to raise the maximum penalties to a 3 percent reduction in overall Medicare payment for admissions beginning Oct. 1, 2014, and preparing to expand the number of conditions covered to include chronic obstructive pulmonary disease (COPD), and elective hip and knee replacements. At 3 percent of Medicare revenues, some hospitals operating on very slender margins to begin with, and also facing penalties under the value-based purchasing program and healthcare-acquired conditions program also mandated by the ACA, could ultimately face shuttering. As Chas Roades, the chief research Officer at The Advisory Board, told Kaiser Health News last August, “The financial penalties aren’t huge right now, but hospital leaders recognize that the penalties will get bigger, and that scrutiny over readmissions rates will continue to grow.” http://www.advisory.com/daily-briefing/2013/08/05/cms-2225-hospitals-wil...

An uphill climb, even for pioneers

So why is this all so hard? Just ask the leaders of pioneering organizations in this area, like the 25-hospital, Arlington-based Texas Health Resources, which has been using data analytics for readmissions work for at least a few years already, and which has been working with neighbor Parkland Hospital on a collaborative initiative. Ferdinand Velasco, M.D., THR’s chief health information officer, puts it this way: “I do think it is going to be a journey. And this will be a Top Tech Trend for many years. I think this will be a similar transition around data analytics, as the EHR transition was. We now have a tremendous amount of data, but it’s going to take a long time” to consistently be able to apply analytics, using that data, to readmissions work.

Up in Philadelphia, Stephen J. Klasko, M.D., president and CEO of Thomas Jefferson University and the Thomas Jefferson Hospital System, says of his health system, “We’ve recognized the burning platform to get this done. We have to promote not over-utilization or under-utilization, but optimal utilization, for the first time. And we’ve recognized at Jefferson, and what David [David Nash, M.D., founding dean of the Thomas Jefferson University Jefferson School of Population Health] in his school has recognized, is that 90 percent of the reason that people get readmitted is that they have symptoms and don’t know what to do.” Connecting patients with their physicians electronically through mobile connectivity, Klasko believes, will be one of several keys to reducing readmissions, as will increasing patient engagement through such motivators as enrollment in high-deductible health plans.

What’s more, as a November 2013 Health Policy Brief from Health Affairs noted, “A study by Yale University researchers identified six strategies that were modestly successful in lowering readmission rates for patients with heart failure. These included partnering with community physicians, partnering with local hospitals, having nurses reconcile medications, arranging follow-up appointments prior to discharge, sending discharge papers to patients' primary care physicians, and assigning staff to follow up on test results after discharge. Hospitals that implemented more of these strategies,” the brief went on to say, “had substantially lower readmission rates. However, the study also found that several strategies intended to reduce readmissions actually increased readmission rates, potentially because they reduced informational and logistical barriers to hospitalization.”

So how does all this translate into “next steps” for healthcare IT leaders? “To me, the most important thing is having a system and an infrastructure that allow you to get reliable data,” says Scott Tongen, M.D., a director for the Pittsburgh-based Aspen Advisors consulting firm. “And then the next important thing,” says the St. Paul, Minn.-based Tongen, whose clinical background is as a hospitalist physician, “is being able to demonstrate that the data is accurate and means what it means. Too often, physicians have a tendency to question the data if they don’t look good, to question the data.”

And when it comes to physician engagement and buy-in, Tongen says, the key is “being able to use the data to identify physician best practices, and then use the data in places where you’re not getting best practices. We worked on readmission for congestive heart failure” at Allina Health System in the Twin Cities when he was there, Tongen notes. The core of the challenge around physician buy-in, he notes, was helping to guide physicians past data findings that were not helpful, such as data around “readmissions for any reason,” and leading them forward around data they could accept and take action on.

Progress in Philadelphia

At the four-hospital Penn Medicine integrated health system in Philadelphia, considerable progress is being made in leveraging data and analytics to reduce avoidable readmissions, reports Christine VanZandbergen, associate CIO of clinical applications for the system, where, she says, “We are driven from a clinical outcomes and quality perspective by our CMO, P. J. Brennan, M.D., and his partnership with a variety of operational and clinical leadership, per our Blueprint for Quality and Safety. Our overarching goal,” VanZandbergen says, “is eliminating preventable mortality, and eliminating preventable readmissions.”

Indeed, the folks at Penn Medicine have an explicit goal of eliminating all avoidable readmissions for several conditions, by July 1 of this year. And while Penn isn’t yet willing to publicly share results to date, VanZandbergen says that considerable progress has been made, using sophisticated analytics-driven processes, and focusing on patients determined through analysis to be at the highest risk for readmissions. Not surprisingly, she says, a bottom-line result of all the analyzing taking place in her organization is that the single strongest predictor by far of a readmission is a recent admission.

What the folks at Penn, Jefferson, and Texas Health Resources are all learning is that working across systems, and indeed, ultimately, community-wide, will be essential to real progress in readmissions reduction, as with the Texas Health Resources-Parkland Hospital ongoing collaboration.

Penn’s VanZandbergen says, “I hope that we’ll be able to expand our data to include outside organizations, since all we have access to right now is internal data; and so that means claims data. And we’re in the process of negotiating relationships with more than one of our payers, to look at these high-risk populations, and work with claims.” Will this be a long, complicated journey going forward? All those interviewed for this article agree that it will. Yet all are also optimistic that the U.S. healthcare system will eventually get to where it needs to get on this journey of 1,000 miles.

 

 


The Health IT Summits gather 250+ healthcare leaders in cities across the U.S. to present important new insights, collaborate on ideas, and to have a little fun - Find a Summit Near You!


/article/top-tech-trends-analytics-readmissions-reduction-work-long-complex-journey-ahead

See more on

betebettipobetngsbahis