Part three in a series of client advisories focusing on the new ACO regulations

This is the third in our client advisory series on accountable care organizations (ACOs). CMS published the final ACO regulations in the Federal Register on Nov. 2, 2011. One of the primary components of the rules is the quality standards by which CMS plans to evaluate the performance of ACOs. That performance will be a key part of the formula used to share savings (or losses) between CMS and each ACO.

Quality Measures Reduced by Half

Comments received by CMS to the proposed regulations resulted in significant changes, both large and small. Big changes include CMS’s decision to reduce quality measures almost exactly in half, from 65 to 33. Those 33 measures are grouped in to four “domains,” which are used as intermediate steps in calculating an ACO’s quality score. The four domains are:

  • Patient experience
  • Care coordination and safety
  • Preventive health
  • At-risk populations
    • Diabetes
    • Coronary heart disease (CAD)
    • Hypertension
    • Ischemic vascular disease
    • Heart failure  

A copy of CMS’s table of quality measures is attached here.

Year One: Pay for Reporting

Quality measures will be determined from claims, EHR reporting, a web interface dubbed “Group Practice Reporting Option” (GPRO), and a survey of the ACO’s patients. For an ACO’s first year, CMS finalized its proposal to treat accurate reporting on the quality measures—regardless of what performance the measures indicate—as entitling the ACO to the highest quality score. The survey will be based on the Consumer Assessment of Healthcare Providers and Systems Clinician & Group Surveys (CG-CAHPS). To relieve some of the administrative burdens that start-up ACOs will encounter, and to achieve higher consistency in the methods and results, CMS will contract with third parties to perform the surveys. For the first two years, CMS will also pay for the work, but ACOs must choose an approved vendor and pay thereafter.

Because there is no baseline for comparison, the survey data will be used for informational purposes only. It will not count as part of the shared savings/loss calculation in any part of the ACO’s first three year commitment.

The quality measures reporting data will be collected from a variety of sources, with seven measures coming from the patient survey, three abstracted from Medicare claims, one from the EHR incentive program, and 22 from the GPRO web interface, which was created as part of the Care Management Demonstration Program. That web interface is capable of either accepting data that the ACO inputs manually or that the ACO uploads from its EHR system.

Years Two and Three: Pay for Performance

Using the measures to affect shared savings will be phased in, with Year 1 being “pay for reporting.” In the ACO’s second year, 25 measures become “pay for performance.” In year three, all save one measure are pay for performance. As noted above, the data gathered through the survey of patients will not affect payments for an ACO’s first three years.

When performance matters, each of the four domains will be weighted equally, with a 4.0 being the highest grade. Within the Care Coordination/Patient Safety domain, the quality measure regarding the percentage of primary care physicians that qualify for an EHR incentive program payment will be double-weighted. Two of the disease-related measures—diabetes and CAD--will be scored on an “all or nothing” basis.

In order to receive points for a measure, the ACO must be above the 30th percentile. It is theoretically possible that an ACO could receive no points for one or more measures and still receive some portion of the shared savings.

CMS plans to audit the data submitted. ACOs that do not audit well will receive a second, and potentially a third, level of scrutiny. Egregious violators are likely, at a minimum, to be tossed out of the ACO program.

In sum, CMS asked for comments about its ACO quality proposals, and it received plenty. As an indication of the level of commitment that CMS has to the concept, the regulators took to heart the admonition that the measures were too many and too burdensome. It is still not clear whether—and if so, how quickly—ACOs will form and participate in the Medicare program. Challenges remain, but the quality standards published in the final regulations appear not to pose as formidable a barrier to participation as initially believed under the proposed regulations.