Skip to content Skip to navigation

Report from eHI Forum: Moving EHRs from Transactions to Analytics

August 14, 2012
by David Raths
| Reprints

There is a “talent arms race” going on in healthcare analytics, claims Matthew Bates, a senior executive and partner with consulting firm Accenture. Speaking at the eHealth Initiative’s National Forum on Data and Analytics in Healthcare in Washington, D.C.,Bates said analytic maturity has an organizational component and is about skills and talent management as much as about technology infrastructure. This is a hard time to be an employer in the field, he said. “As soon as you train people, if they don’t like you, they have three job offers and can leave.”

Bates’ talk reiterated one of the main themes of the Aug. 9 conference: that organizations now have more data than ever, but the question is: what is the right data to solve a particular problem? “It’s not about the amount of data you have or the number of sources,” Bates said. “It is about what you are doing with it.”

In his keynote address, John Glaser, PhD, CEO of Health Services at Siemens Healthcare, described several of the ways that clinical information systems are shifting from transaction-oriented to “intelligence”-oriented. Healthcare entities will be under increasing pressure to prove they are providing quality, safety and efficiency. “We will have to get better at managing people over time,” he said, “and we are still not very good at following the evidence.”

Glaser,previosly CIO of Partners Healthcare in Massachusetts, argued that we are entering a new era in healthcare delivery in which large volumes of data will allow for the orchestration of complex processes, such as tailored cancer therapies, computer reconciliation of data inconsistencies, and the use of EHR data to compare the relative risk of medications.

During the daylong forum held at the Gaylord National Convention Center, several pioneering organizations presented examples of how they are starting to use electronic measures to drive quality improvement processes. For instance, Jeffrey Hill, PhD, measurement and evaluation lead for the Rhode Island Beacon program, described some lessons learned from the project’s effort to harmonize measures, definitions, and reporting. The Beacon is working to transform patient-centered medical home care processes across the state, starting with 28 practices that account for about a quarter of the patients in the state.

 “It is difficult but possible to harmonize measures and definitions before even thinking about aggregating data,” Hill said. “These types of reports can be used to engage providers in quality initiatives and practice transformation, as long as it is convenient,” he added. Third, he said, comparative data is key. It is preferable to offer physicians not just performance levels, but also rates of improvement and explanations for why it is happening. “It is possible to link practices performing less well with ones performing very well, and they both can learn something. Best-practice sharing goes both ways.”

Niall Brennan, director of the Office of Information Products and Data Analytics, Office of Enterprise Management, in the Center for Medicare & Medicaid Services (CMS), reminded attendees that more advanced analytics are essential to the whole health reform agenda. “We can’t do any of it without getting more control of our data destiny and turning it into actionable information,” he said. “CMS has always been about data, not necessarily analytics, but we’re changing.”

For instance, pioneer accountable care organizations are now getting monthly claims data feeds. “We are working on ways to make the data we provide more helpful, so that analytics firms can turn that raw claims data into actionable clinical information to help ACOs achieve their goals.” By 2017, he noted, every physician will be rated on cost and quality and will either receive bonuses or cuts. “We are moving to value-based care, so we have to give them evidence. ACOs are the tip of the iceberg. We are moving to routine data sharing on a vast scale with all our stakeholders.” He mentioned one piece of the Accountable Care Act calls for the creation of “qualified entities” that can get the federal claims data and combine it with data from other sources to produce combined performance reports for providers.

Like Accenture’s Bates, Brennan said the real challenge is agreeing on the type of things to analyze.  “You can have all the data in the world, but without the right analytics, it is still just a bunch of useless numbers.” In the past, attitudes about analyzing data at CMS have perhaps been overly conservative, he said. “We are trying to change that. A lot of that is about culture change.

After listening to a few hours of presentations about registries, dashboards and clinical decision support, Paul Harkaway, M.D., a critical care physician from Ann Arbor, Mich., tried to bring the conversation back to the patient. “Chronic disease management is all about relationships,” he said. “That is still analog. We have to learn about our patients. What we need to come up with are digital tools that facilitate us having meaningful relationships. Please make sure you aren’t making tools that get in the way.”