Start small for big data sharing wins

When it comes to data sharing and data standards, “we’re not going to solve this problem overnight,” said John Supra, CIO of South Carolina’s Department of Health & Human Services, speaking during the closing panel session at the State Health IT Connect Summit on April 2. “It’s going to take a lot of time.” A good starting point, he said, is building common elements, such as common rules, policies or definitions.

States have different needs and objectives, said Joe Bodmer, director of the interoperability initiative at the Department of Health & Human Services’ Administration for Children and Families. He believes in incremental improvement, he said. “We do not have to do it all at one time.” But it can be hard to break through different agencies’ agendas. By picking low-hanging fruit—“something that’s innocuous to all the various programs”—communities can establish governance and trust and begin to work toward larger projects.

Smaller projects teach people how to play together, said Bodmer. “These projects should have a charter that lays out roles and responsibilities and then you have a comfort zone because people know how to work together.”

Bodmer said communities have been successful by looking at shared users rather than shared data. “We want states to get creative and innovative.” Come up with good, valid reasons why these programs should be part of waivers and funding, he said. “We’re putting tools out there for states. Take advantage of them. Use them.”

Sharing a small amount of data “can accomplish something very big,” said Supra. He also is interested in finding ways to share the good ideas that worked in one state. “There are very different states but they have some of the same challenges. We want to figure out how to apply bits and pieces in different states. That level of thinking is what lets us drive change a lot more incrementally than we have in the past.”

Some states have said that the data requirements for different federal programs conflict, said Jessica Kahn of the Centers for Medicare & Medicaid Services’ Division of State Systems, Data and Systems Group. “We don’t want to create disincentives to comply. We struggle sometimes to get the full picture but shame on us if we continue to fund things that could be integrated.”

Standards develop because a group of people decided it was important to share this data, said Supra. Start with something small and make it meaningful, he said. “If a bunch of states do something and the feds are encouraging it and giving us incentives, there is a point down the road where those who don’t want to be on board have to be handled with a stick rather than a carrot. Once there is enough capacity to say this makes sense, there is a role the feds need to play on big important pieces.”

When an audience member asked about quality measures, Kahn said the issue isn’t how many quality measures are needed. “The sticking point I still see is that a provider could report the same quality measures four different places in four different ways. The data just moved was still so redundant that we barely even make an improvement in the problem.” Rather than worrying about picking the right measures, if you can consolidate through health information exchange how quality data are reported from your providers and enabled and they can give it to you, you’ve narrowed the channels, she said.

Kahn said there is a market for vendors that can take data and slice them the way providers need it “so it’s not the provider’s headache and payers get data in a way that’s reliable.”

There is never going to be less demand for data and certain types of providers just don’t have the IT people on staff to manage it, said Kahn.

Another audience member asked about keeping standardization current, which is not happening for EHRs at this point. “That’s a huge conundrum,” she said. If organizations don’t modify their EHR to achieve better usability, people won’t use it. On the other hand, if they do modify the EHR it doesn’t interoperate.

Getting data standards “is what will make those other problems go away,” Kahn said. “If you are collecting data on a granular enough basis, you will be able to report it out again.”