Population Health

Harnessing analytics has led to big gains for Maine-based health system to dramatically reduce readmissions, identify high-risk patients and utilize real-time data.

There are two major roadblocks for analytics in healthcare, says Sriram Vishwanath, professor of engineering and data science at University of Texas, Austin. They have to do with two different mindsets: "the analytics is a commodity" and "it's my data and I won't share." In the "commodity" mindset, providers feel pressured to join the other organizations with analytics in place and rush out to hire low-cost engineers to create an analytics system. But according to Vishwanath, "this mindset is dangerous, as it leads to a lot of sub-standard dashboards all being paraded around as predictive analytics solutions. "Analytics isn't just putting together a bunch of engineers," he added. "Quality, team-skill, years of experience and depth of understanding the matter. It's important to recognize that analytics should be done right the first time, by working with an exceptional, high-quality team of PhDs." With the "I won't share" mindset, meanwhile, data is seen as a precious commodity, and providers find it painful to let go of their data. This mentality is a common complaint, he said, but it's "gradually changing – albeit at a glacial pace." It's important for all entities involved in healthcare to work together. Not just institutions as a whole, but also within an organization, from the MDs to the C-suite. "Working together is critical here, to accept that neither side knows it all and must learn from the other," said CHP's Suresh. Other hurdles to overcome include bridging the gap between the providers who feel analytics are an overwhelming waste of time and the vendors who press that analytics are a fix-all solution for every problem. Neither of these extremes is true, said Suresh. "Predictions can never be 100 percent accurate. If they were 100 percent accurate, one would call it a fact, not a prediction," he said. "Is it right 9 out of 10 times? Well, then you have a great predictive engine in your hands. Healthcare predictions are meant to supplement, support and guide and can never be 100 percent accurate." Learn more at the Healthcare IT News Big Data and Healthcare Analytics Forum in San Francisco, June 14-15, 2016. Register here. "Healthcare has a long journey in finding true value from analytics," he added. "There's tremendous value for healthcare from analytics when analytics is done right. Healthcare analytics will, someday, change the way we manage care. This is not an if, it's a when." To accomplish this task all stakeholders must be on the same page. For Vishwanath, those organizations attempting to leverage analytics must lean on professional vendors with established platforms in place and come prepared with goals. Analytics tools are only effective when they are designed to meet a specific need within an organization. "If the vendor can prove its accuracy and does so with relative ease, put them on your target list," Vishwanath said. "If they obfuscate with lots of buzzwords, walk away." The Healthcare IT News Special Report: Bridging the big data-analytics gap Data-Analytics Gap: How real-time strategies can improve care quality and efficiency Case Study: St. Joseph Healthcare sees dramatic improvement serving high-risk population with HealthInfoNet Case Study:  UnityPoint Health connects providers and hospitals with predictive analytics to improve patient care Case Study: Children's Hospital of Pittsburgh develops customized prediction indicator for children with PeriHealth

The confluence of new care models and technology are enabling data scientists to pinpoint gaps in access to care, address social determinants of health, and map data that informs tactics to improve outcomes at the patient and population level.

The pediatric endocrinologist at Stanford University's Lucile Packard Children’s Hospital is known for his HealthKit pilot study on Type 1 diabetes patients.

Vice President Joe Biden unveiled the precision medicine database on Monday, speaking at its operations center at the University of Chicago. Genomic Data Commons is a National Cancer Institute initiative and is central to the National Cancer Moonshot and Precision Medicine Initiative.

Research firm’s latest report also reveals that electronic health records software vendors have made substantial clinical workflow enhancements since 2014 but lacking interoperability continues having an impact on users.

Nicholas Marko, MD breaks down the differences between big data analytics and business intelligence, and explains how Geisinger chooses between centralizing and federating data.

A surge of recent developments – the rise in apps for care management and wellness, advances in research data, precision medicine gains – is placing a great responsibility on healthcare leaders to break down data silos to simplify data sharing among providers. And consumers, too.

Healthcare IT News and HIMSS are accepting topic and speaker proposals for Pop Health Forum 2016, scheduled for September 12-13.

Geisinger Chief Data Officer Nicholas Marko, MD, says deriving value from data depends on a sound roadmap. Here’s a look at how to determine what you need to get there.