Big data 'not for the weak'
The University of Mississippi Medical Center is poised to deliver EHR data to in-house patients, starting in the next two weeks, to help determine whether they are at risk for a heart attack in the immediate future.
That’s just one of the big data initiatives UMMC is undertaking, which also include using geomapping and what chief health information officer John Showalter, MD described as “fully automated deep machine learning within our EHR,” that the hospital system will go-live with in early December.
[See also: Analytics means we 'roll up our sleeves']
“We have a strong sense of urgency,” Leigh Williams, revenue cycle director at UMMC said. “We’re focused on tangible meaningful actions for our clinicians.”
Inverted big bang
Alongside UMMC at the Healthcare IT News Big Data and Healthcare Analytics Forum here, a broad swath of healthcare providers, from the large-scale Kaiser Permanente to Carolinas HealthCare System across the country, and Hospital Sisters Health System in Illinois and Wisconsin, are working toward different flavors of big data and analytics.
[See also: Big data doesn't have to be 'Star Wars']
One takeaway: “This is not for the weak,” said Carol Lovin, chief strategy officer at Carolinas HealthCare System.
Whereas the Big Bang started with a single point of origin and spread out from there, healthcare and other industries are moving toward big data in essentially the opposite way: An implosion triggered by combining previously disparate data sources tightly-bound to metadata into a database from which clinicians and researchers can extract what’s interesting to them, said John Mattison, MD, chief medical information officer at Kaiser Permanente.
“That’s a radical departure from the data warehouse model we all know,” Mattison added. The critical differences Mattison pointed out include the fact that big data involves more atomic-level data and atomic-level metadata — and the upside is that healthcare organizations can use a minimalistic data model.
Indeed, that minimalist data model enables healthcare organizations to “transfer data very quickly into new algorithms and metrics,” according to Sarah Mihalik, vice president of provider solutions at Explorys, a Cleveland Clinic spinoff that sells products for clinical integration, population health management, cost of care measurement and pay-for-performance contracts.
“Not having a rigid data model means providers can get more predictive insights from the data,” Mihalik said.
Aligning the right talent
Such a drastic change, of course, will require that healthcare organizations develop or hire new skill sets to compile what Mattison called a “cross-breed of expertise wherein data scientists work in tandem with subject matter experts.”
CIOs historically have driven analytics initiatives, but that is changing and Caradigm vice president of population health Brian Drozdowicz said he is seeing among customers a roundtable of technology experts working with clinical champions, revenue cycle managers, and in some cases chief transformation officers.
Several speakers, including UMMC’s Showalter and Jeanne Huddleston, MD, medical director of healthcare systems engineering program at the Mayo Clinic, recommended that big data teams also include engineers.
At Mayo Clinic, for instance, Huddleston and her team use analytics to streamline, among other things, scheduling and patient check-ins, staffing and workload for the emergency department.
“Putting the engineers in the workplace while we’re with the patients is going to take us where we want to go a lot faster,” Huddleston said. “There is nothing like actually seeing it happen in real life. We need to use real space and real conditions to see what needs to change.”
Priority No. 1: financial sustainability
Showalter confessed that when people ask what his professional title means — he’s a chief health information officer — he does not have a good answer other than making sure UMMC gets a return on technology investments.
“How do we do this in a way that is going to bring financial sustainability?” UMMC’s Williams asked.
That is a reality many hospitals are grappling with as they move toward analytics and big data. As recently as 2011 only about 10 percent of healthcare organizations were using analytics tools, according to Shelley Pryce, HIMSS director of payer and life sciences, adding that estimates suggest by 2016 that statistic will climb to 50 percent.
Getting there has plenty of rewards, better data for treating patients, streamlining operations, cost-savings among those. But it won’t be easy.
“It really is an iterative process to change the organization’s mindset to be data-driven. It does take time,” Caradigms Drozdowicz said. “If there’s a strong analytics background in place you can fairly nimbly make changes to improve operations.”