The Department of Health & Human Services (HHS) plans to award a contract to build a new multi-payer/allclaims database by September 2010 that will enable health researchers to better study healthcare trends and treatments.
But government healthcare players and observers think the database could initially provide other, secondary applications that in the short-term would prove even more beneficial to improving patient care and outcomes.
These include facilitating better continuity of care and case management for dual-enrolled Medicaid and Medicare patients, identifying population-based trends and decreasing fraud, waste and abuse, they said.
"We're excited that they're moving forward with this," said Ann Kohler, executive director of the National Association of State Medicaid Directors, noting that Medicaid agencies have been asking for access to current Medicare claims data for years.
The plan also addresses the Holy Grail of health IT: higher quality and lower costs, according to Kohler. "The most expensive clients in the Medicaid programs are the dual-eligibles and they're also usually the most frail," she said. "We think that pulling all this data together in an easy-to-access national database would not only help states coordinate these patients' care in a better and more effective manner but it would also make sure that there's no contraindicated care that's happening.
"What's more, it would give states a much clearer picture of the cost of treating these clients."
Wes Rishel, vice president and an analyst for Gartner's healthcare provider research practice, said there were "plenty of good reasons" for such a database.
A universal database could go a long way toward helping Medicare and Medicaid identify double-dipping providers and other profiteers who succeed by splitting up their billing among different states and government payers, he said.
It could also help CMS know which dual-enrolled patients, for example, are not being tested or treated for diabetes and other chronic diseases. "This would allow a better view of the overall population statistically, which is a very good thing to do," Rishel said.
The database, which will be paid for with funds from the American Recovery and Reinvestment Act, is also being touted for its role in bringing wide-ranging, longitudinal claims data to HHS's "comparative effectiveness" research portfolio.
"For certain types of comparative effectiveness questions, such as surgeries, hospitalizations and readmissions, claims data can be used to either answer these questions or test hypotheses that guide further research," said an HHS spokesman.
"The goal is to likely start with claimsbased data and to build a sustainable model that can be scaled over time" including to different data sources."
Initially, according to HHS, the database would not include all payers, just longitudinal claims data from Medicaid, Medicare, other government payers and private insurers that want to participate.
The agency in late January awarded a $1 million contract to Avalere Health LLC to conduct a design study on the proposed database. Among the questions HHS officials hope to have answered are: What risks and benefits exist for various technology models? What types of questions that could be answered with claims data? How can it be best utilized? How could the data resources be scaled over time?
Dissenting view
But not everyone thinks a claims data strategy is the path for HHS to take. Kerry Weems, senior vice president of health solutions at Vangent Inc. and the former administrator of the Centers for Medicare and Medicaid Services, said HHS should put its funding into building databases stocked with more timely clinical data compiled from providerbased electronic health record systems. That represents "skating to where the puck is going to be," he said.
Weems said claims records provide delayed, frequently inaccurate information with too little clinical specificity to be of real value to health research or quality conclusions. Moreover, he worries that because many states rely on a managed care model"which doesn't always result in claims being filed"the conclusions will be skewed.
"The proposition is: If you have a limited amount of money to invest, should you invest it in claims or in a system of being able to harvest real, rich meaningful data from EHRs?" he said. "I know what the answer would be for me."
Gartner's Rishel agrees that a universal database of clinical data would be ideal, but starting with claims data might be more practical way to begin the endeavor.
"Claims data is where you have to start"it's what we have the most of right now," he said. "And I'd rather we get started with some rough data now and be working through those technical issues and then improve it as we go along than to wait for perfect data.
"If we do that, then years from now, it will be a lot easier to crank that better data into your equations and come up with the research and analysis that will truly improve patient care."
