Since the Obama administration launched its accountable care
initiative as part of healthcare reform
in 2012, the CMS
has announced which hospital and physician networks in one of the programs have met cost targets and received shared savings. But so far, it has published little data on quality of care delivered by these networks, which were designed to deliver better care as well as lower costs for Medicare patients.
The $380 million in Medicare
savings produced by accountable care organizations in the Shared Savings and Pioneer programs in 2012 have been well-publicized. But the CMS has not yet delivered on its stated goal of transparency for quality-of-care measures, which will be used to evaluate Medicare ACOs to determine whether they will receive financial bonuses or penalties. Quality measures ensure that ACOs—which are eligible to keep half to 70% of what they save—do not inappropriately skimp on care to win the savings bonuses.
“If you're going to collect 33 (quality) measures, I don't know why it is that the public doesn't have access to that information,” said Dr. Robert Berenson, an expert on Medicare policy and senior fellow at the Urban Institute. Public access to the data adds “credibility and accountability” to the ACO program and could help Medicare beneficiaries decide whether to join an ACO if proposals to establish enrollment in the program are enacted, Berenson said. Lack of confidence in the validity of the quality data may be one reason for delays, he said.
Click to enlarge
More transparency will come, said Dr. Kavita Patel, managing director for clinical transformation and delivery at the Brookings Institution's Engelberg Center for Health Care Reform. Hospitals and doctors may be skittish to publicly report their quality data, while CMS officials may see a limited release in the first year as one way to “just get them comfortable,” Patel said.
“Year one was sort of the dress rehearsal,” said Dr. Anita Ung, medical director for quality improvement and performance measurement at Atrius Health, a Pioneer ACO in Massachusetts.
The CMS declined to answer questions about why it has opted for a limited release of quality data or about the deadline for ACOs to post performance online.
The CMS is largely responsible for public disclosure of quality performance for more than 300 ACOs that care for more than 5 million Medicare patients. First-year results are available for the 145 ACOs that have operated since 2012 in the CMS Shared Savings Program and in the CMS Innovation Center's Pioneer initiative. But so far, the CMS has released results for only five of 22 quality measures it has responsibility for publishing. The original ACO rules said that releasing these data would “hold ACOs accountable and contribute to the dialogue on how to drive improvement and innovation in healthcare.” Meanwhile, ACOs are still waiting for final guidance from the CMS on how to publish three measures on preventable hospital admissions.
The CMS collects more extensive data than it reports under ACO rules, which require hospital and medical groups to meet performance criteria for 33 measures. The CMS also internally audits utilization and spending among ACOs every three months for signs that providers are stinting on care or overtreating other patients to offset reduced revenue from ACO patients. While ACOs themselves can see the data, the public can learn little about ACO quality of care from the CMS data.
The agency's reticence about publicly releasing ACO quality information runs counter to the Medicare program's increasing boldness in publishing cost and quality information about hospitals and physicians, and to the growing pressure for cost and quality transparency in the private sector as well.
Click to enlarge
Despite the limited public reporting, however, ACO leaders say the program's quality measures have accelerated improvement efforts and strengthened care, most notably in areas not previously tracked by providers and in areas of weak performance.
The information that has been released shows that ACOs earning shared savings from having met cost targets also demonstrated better quality performance, suggesting that the two accomplishments may be linked. ACOs that received financial bonuses had stronger quality scores across four of five publicly reported measures compared with ACOs that did not slow spending enough to earn bonuses.
One Pioneer ACO in Minneapolis has voluntarily released quality performance results that opens a window onto the results for other Pioneer and Shared Savings ACOs that haven't released data. Allina Health's data release reveals sharp variations in quality performance across measures of preventive care and disease management. ACO leaders say some of that variation can be attributed to inexperience with newly introduced measures and confusion over reporting criteria during the first year.
Under the rules for the Shared Savings and Pioneer programs, ACOs that fail annual quality performance measures cannot receive shared-savings bonuses, no matter how large their savings. Both Shared Savings and Pioneer ACOs are required to meet the same quality targets.
For the first year, ACOs were simply required to submit quality results to Medicare to be eligible for possible bonus awards based on savings. Their quality performance initially was not factored into their bonus calculations. Five of 145 ACOs failed to submit the quality measures, and two were deemed ineligible for shared savings as a result.
The quality requirements grow more stringent after the first year, with ACOs scored on quality performance for an increasing number of measures during the second and third years. To be eligible for bonuses in the second year, ACOs will be scored on quality performance for 25 measures, increasing to 32 measures in the third year. The ACOs, however, will continue to be required to report their performance on all 33 measures. Second-year quality results for the 23 Pioneer ACOs, measured through last December, will be finalized this summer.
The five measures the CMS made public track whether heart disease patients receive recommended prescriptions and how they fared on diabetes management. Quality measures not yet released by the CMS or ACOs include patient-satisfaction scores, rates of preventable hospital admissions, and screening rates for colon and breast cancer, depression, and risk of falling.
ACOs and the CMS faced technical hurdles as they worked to report ACO quality measures, ACO leaders said. Confusion during 2012 created reporting discrepancies that make performance across ACOs for some measures difficult to compare. ACOs adopted different interpretations of how to collect and report some measures. Hospitals and physicians also needed the year to adapt electronic health records to capture new data and report measures they previously did not track.
Nonetheless, ACO leaders insisted that the program's quality requirements have helped providers identify and target quality efforts where performance was weakest. Doctors armed with new data culled from quality registries now focus additional effort on patients who have gaps in care. The tougher requirements for the second and third year of the program forced ACOs to move quickly to make quality improvements, said Dr. Scott Hines, an endocrinologist and co-chief transformation officer for Crystal Run Healthcare ACO in New York's Hudson Valley.
Voluntary reporting by Allina Health's Pioneer ACO may offer the most comprehensive public information available on ACOs' quality performance for 2012, the first year of the ACO demonstration. Allina, a six-hospital system, published its 2012 quality performance compared with the highest, lowest and median performance figures for all 145 ACOs in 2012.
Dr. Jen Childs-Roshak reviews an X-ray with a colleague. Childs-Roshak is a member of Atrius Health, a Pioneer ACO in Massachusetts.
The Allina data show that performance across ACOs varied significantly on many measures. Rates of influenza immunization ranged from 42% at the 30th percentile of performance to 71% at the 90th percentile. Rates for colorectal and breast cancer screening also varied. For example, rates were 43% for colorectal cancer screening at the 30th percentile and 87% at the 90th percentile.
Performance on a measure known as medication reconciliation also showed wide variation. The measure tracks the percentage of patients who see a physician to review their prescriptions within 60 days of leaving a nursing home or hospital. At the 30th percentile, 70% of patients saw a doctor who reviewed their medications within 60 days of their leaving a hospital or nursing home, compared with 99.7% at the 90th percentile. The median rate was 85%.
For 114 Shared Savings ACOs, the CMS' publicly reported results on five performance measures in 2012 show a correlation between strong quality performance and cost savings. ACOs that succeeded in earning shared savings reported higher median scores than ACOs that did not receive shared savings on three out of four measures for diabetes control. Tobacco use among diabetics was the one exception.
There are efforts to improve Medicare ACO quality measurements and integrate them with other quality programs. Some organizations, including the National Committee for Quality Assurance, are analyzing quality measures across public and private ACOs for ways to standardize and benchmark ACO quality performance inside and outside Medicare.
The Dartmouth-Brookings ACO Learning Network is reviewing where quality measures overlap across private insurers, Medicare ACOs and various federal programs. The goal is to create standard measures for ACO performance, said Dr. Tom Valuck, a consultant with Discern Health.
Meanwhile, the American Hospital Association has criticized the current Medicare ACO performance measurement system. The AHA, in a letter to Dr. Patrick Conway, acting director of the CMS Innovation Center, described the current quality measures as confusing and an obstacle to Medicare ACO expansion.
Hospitals will balk at the hefty investments required to launch an ACO without clearer financial incentives, said Nancy Foster, the AHA's vice president for quality and patient-safety policy. Follow Melanie Evans on Twitter: @MHmevans