Healthcare organizations throughout the country are lining up for a race to receive the massive Medicare database of claims data covering the more than 40 million consumers who participate in fee-for-service Medicare and prescription drug plans (Medicare Advantage plans are not included.) It's not clear which organizations will be judged worthy by Medicare to use the data. All must already demonstrate expertise in using healthcare data, promise to build provider ranking tools, and give the public free access to ratings.
[In Pictures: The 10 Best Places to Retire in 2012]
The decision to release the information was announced early in December. While mandated by the healthcare reform law, the ground rules for accessing and using the data were hammered out in negotiations with the healthcare industry and detailed in 100-plus pages of regulatory language. Although private insurers and other healthcare providers already share a lot of information, including Medicare data, the claims data release is expected to enable more detailed evaluations of the services provided by doctors, hospitals, nursing homes, clinics, and other healthcare providers.
Among data of particular importance to older patients, hospital readmission information will be a high priority, as will extensive cost information. While based only on claims payments to providers, the information includes details on what the payments were for, and thus can be used to build pictures of which tests and procedures patients received and how their care compares with national and regional healthcare standards.
"There has been a real dearth of health information generally," commented Niall Brennan, the director of the policy and data analysis group at the federal Centers for Medicare & Medicaid Services (CMS). "Even I, as a Medicare expert, have no clue who I should choose as a doctor. So I wind up calling one of my friends to see who he recommends."
While the information will identify healthcare providers, public disclosure of information that would identify specific patients is not allowed. Still, there is sensitivity over the possibility that such data could be mistakenly released or stolen, and Brennan says any organization judged eligible to receive the claims data must agree to stringent data use and oversight rules, including outside auditing of their activities.
The new rule becomes effective January 9, Brennan said, when organizations will be able to download and submit their applications. To qualify to receive the database, an organization must already be familiar with healthcare claims data and have access to claims data from sources other than CMS. Organizations that can cobble together the most complete records of healthcare services from all providers will be in the best position to receive the new Medicare data.
Brennan said he hoped the agency can "provide data to the first wave of applications by the end of March." After that, the data must be incorporated into existing ranking tools or new software must be developed to provide the rankings to the public. Before that can happen, he explained, healthcare providers must first receive their rankings and have an opportunity to review and appeal them. "They are going to have 60 days to do that," Brennan said, "and that was a very important provision that was included in the rule" to convince providers to support the new regulations.
The fastest the public would see ratings that include CMS data would be late summer or early fall, he said, but only if an organization had already developed the software and presentation tools for public reporting. Building a rating system "is pretty expensive," he noted. While the information is to be provided to the public at no charge, there may be ways for organizations to sell more intensive reports to healthcare providers. Also, for example, they might be able to post the ratings on freely accessible websites but make money from advertisements that run on the ratings display pages.
"I think it will be an evolutionary process between CMS and these organizations," Brennan said. "We want to spur innovation. And we want these guys to have acceptable business models, but not at the risk" of using the information improperly.
There is expected to be sensitivity among providers to low ratings. But Brennan said he hoped that would be replaced in the future with a different attitude. "I do believe over time that providers will come to look upon this not only as a positive development but as a very useful and effective way to give them feedback on their practice patterns," Brennan said.
In many cases, existing state and local healthcare initiatives that already issue provider rankings may have an inside track to receive the CMS information. "Our assumption is that these will be regional or state-based organizations," he said. The agency initially said it expected perhaps two dozen organizations throughout the country would get the information but the total could well be higher.
"Because it's a publicly available data set, I think there will be many more users than the CMS has predicted," said David Lansky, CEO of the Pacific Business Group on Health, one of the regional nonprofits that will apply to receive the information.
Adding Medicare claims data to the provider information already available "improves our ability to understand how doctors and hospitals are performing," he said. "But the data itself has limits," he cautioned. "It's claims data. It doesn't tell you an awful lot about the illness or the treatment or the outcome."
CMS also purposefully did not instruct organizations how they must present the data. "While you can measure lots of things, part of the role we play is to provide some focus on the things people should be looking at," said Lansky.
It's likely, for example, that some organizations will emphasize physician performance while others will focus more heavily on hospitals and other institutional sites where healthcare is provided. That's the case in Minnesota, where group healthcare practices are prevalent, says Jim Chase, president of Minnesota Community Measurement.
"We haven't so far" provided data on individual doctors, Chase said. "We've found it better to report on the site of care." Chase thinks there will be more interest in the information by healthcare providers than consumers. "It really isn't an individual patient who says, "Oh, here, I'm going to do something differently.'"
"The public historically has not had a lot of interest," said Lansky. "It has not been convenient for people to get access to this information when they needed it, and the data has not been as targeted as people would have liked."
However, once the data is easily accessible in online reports on doctors and hospitals, new ways of using it will likely be tested, and broader public interest could grow. Publication of provider ratings may also trigger renewed sensitivity to the information being released. "We've had pretty good success in [providing ratings for] the primary care area," Chase said. "Measurement is a little bit newer in the specialty care areas, so there may be some [provider] concerns there. And whenever you get into the cost of care, people get nervous."
Chase said he expects that it will take his group until 2013 to reflect the new information in its public ratings. Lansky said he was shooting to begin making public use of the Medicare data this fall.