CMS’ announcement that it was ending the Medicare Health Support (MHS) demonstration project caused a spasm of criticism from the DM industry earlier this year. Despite the efforts of high-powered legislators who requested an extension to the MHS program, CMS still plans to end the project this year.
The project began in 2005 as a way to test DM programs in the Medicare population in the areas of HF and diabetes. MHS has traveled a rocky road in the past three years, with three of the eight awardees (LifeMasters Supported SelfCare, McKesson Health Solutions, and CIGNA) dropping out of the project.
A CMS official says MHS is just the latest demonstration project with DM interventions that did not prove worthwhile for the Medicare population. (CMS policy does not permit public attribution of comments by its staff members.)
Other DM-inspired demonstrations have focused on COPD, CHF, and diabetes, and have included call center nurses, health educators, social workers, and other population-based care coordination. Most of the projects have been “largely disappointing,” says the official.
“Overall, we have not seen significant impacts on behaviors for the Medicare beneficiaries who are participating in the project. We’re not seeing significant impact on utilization of services and therefore on costs,” says the official.
Those involved in the project and industry leaders following the demonstration have complained that CMS has not been providing up-to-date information. Thomas Wilson, PhD, DrPH, an epidemiologist at Trajectory Healthcare, LLC, and founder and board chair of the Population Health Impact Institute in Loveland, OH, says CMS has not offered enough program information to allow individuals to independently judge the program’s successes and failures.
Wilson says the methods by which claims are made should be transparent, adding that if results only are disclosed, they are unintelligible unless the methods used to produce them—including risk adjustment techniques—are also provided.
“What we have learned is that we have not learned as much as we would have liked,” Wilson says. “My general feeling about this whole thing is that it’s taxpayer money and there should be more transparency regarding the methods that are being used to assess performance—both clinical and financial.”
The interim report on the project’s first six months, released in summer 2007, provided the healthcare community with some useful lessons. However, since that report, CMS has not been offering that kind of information, Wilson says. “The answer to questions ‘Did it work?’ and ‘Did it not work?’ must be based on replicable evidence—methods and results—available to the public. We still don’t know the basis of CMS’ opinion,” he adds.
Part D pharmacy claims data is one example of the information breakdown, says Michael F. Montijo, MD, MPH, FACP, senior vice president of solutions and government lead at Healthways in Nashville. CMS did not give the companies the claims data until more than two years into the program, he says.
Not having that kind of real-time data affected Healthways’ ability to intervene with at-risk patients, Montijo says. “The ability to use that effectively was obviously diminished,” he adds.
Paul Serini, executive vice president of XLHealth in Baltimore, which is involved in the MHS project, as well as CMS’ chronic condition special needs plans (SNP), likened the MHS project to giving the companies only half of a toolkit. The DM vendors in the MHS program are not provided with pharmacy claims data to more effectively track participants. That is not the case in the SNP.
“Our ability to do what we do best and help the member is greatly enhanced in a SNP model because, as the health plan provider, we have real-time visibility for both medical and pharmacy claims,” Serini says.
One thing the MHS project revealed was a sicker patient population than was expected. Serini says MHS beneficiaries were “twice as sick as SNP [beneficiaries]. That means you better have a lot more on-the-ground nurses and a lot more programs to do a lot more things than you thought you may have needed at the beginning.”
Rachel Haltiwanger, vice president of operations at XLHealth’s Tennessee MHS program in Brentwood, says on-the-ground nurses communicate regularly with the beneficiaries and their physicians.
For the physicians, nurses provide another connection to the patients, and for the beneficiary, nurses serve as advocates and healthcare experts who understand the system’s maze.
“Having someone be the champion to help them navigate through the medical management is important in making sure that they can get the appropriate care they need, raising awareness of warning signs that put them at risk, and educating them on the importance of follow-up,” Haltiwanger says.
A CMS official says the federal agency heard complaints from both sides of the sickness debate. Some companies said the CMS patients were too sick, whereas others said they were not sick enough to allow for adequate program savings.
The CMS official says the DM companies may have had trouble saving money and improving outcomes in MHS because the industry isn’t used to caring for an older population.
“It’s sicker and older than the population that disease management vendors are used to dealing with. They do some work in Medicare Advantage, but Medicare fee-for-service beneficiaries are generally older and sicker than Medicare Advantage beneficiaries and they are older and sicker than the commercially insured population where most of disease management takes place,” says the official.
One problem was that several beneficiaries dropped out of the project because of death, losing benefits, or being moved into hospices, Montijo says.
“This high level of attrition, of which there was around 15% a year for most of the programs, produced a design where beneficiaries were dying faster than they could be reached to start the program. All beneficiaries must consent to participate, and that is the first step. I don’t think anyone realized how heavily weighted the success of the program was to the early days of the program. The maximum opportunity for savings in the design was day one, and the least opportunity would be the last day,” Montijo says.
He says the beneficiaries could be divided into the following three groups:
The first group could have benefited from less intensive and less costly wellness programs rather than DM. Intervening with the second group increased costs while not reversing their sickness trends. And the third group is where DM companies had an effect. “The cohort design gave you a population that was not as homogenous as what we thought it might be,” Montijo says.
The MHS project taught Healthways that to achieve the greatest effect, it was more important to target the neediest individuals in ways that would help them respond. The company has taken that lesson and is starting to implement it in other product designs.
Healthways offered face-to-face services to those who were institutionalized, while providing intensive care management that included outbound phone calls to the remaining beneficiaries. That targeting led to a higher degree of sophistication using predictive models to find those who were at risk and whom Healthways’ services could help. “We honed and honed predictive models to better identify those who we should and those who we shouldn’t be intervening with,” Montijo says.
That outreach generated a high level of engagement. More than 87% of Medicare beneficiaries who were approached to take part in Healthways’ Washington, DC, and Maryland project participated. Moreover, Healthways received thousands of cards and letters from beneficiaries, physicians, and caregivers about the importance and value of this program.
Healthways has taken the lesson of targeted outreach from the MHS project and implemented it across its commercial population. With the program ending, the DM industry must wait until late 2009 or early 2010 for the independent analysis of the MHS program. Until then, DM will wonder why the MHS program was not extended and whether programs within MHS can be emulated.
“The methods that a vendor, especially a government vendor, uses to assess the value of their service should not be a secret,” Wilson says. “While a DM service can certainly have proprietary tools as part of its intervention package, the methods that are used to assess the value of those services—especially when delivered to a government purchaser—should not be a secret.”
MHS comings, goings
The following Medicare Health Support (MHS) awardees are still taking part in the project and serving these regions:
Green Ribbon Health—Central Florida
Aetna Health Management, LLC—Chicago
Health Dialog Services Corporation—Western Pennsylvania
Healthways, Inc.—Washington, DC, and Maryland