By Sharon Bee Cheng, Strategic Healthcare
Cash bonuses for ACOs will be funded by the savings they generate for the Medicare program.  However, ACOs will not receive the entire portion of savings they are entitled to unless their performance on quality measures is among the best in the country.  Each ACO will be measured on how well they improve care, safety, and population health and will be compared to their national peers.  ACOs which out-perform a third of their peers will get some of the savings they generated; only those which out-perform 90 percent of their peers will receive a bonus equal to the entire savings pool for which they were eligible.
ACOs will be required to accurately and completely report quality measures to qualify for the full bonus amount, if they successfully generate savings. In the first year of the agreement, successful reporting will be sufficient to earn bonuses.  In the second and subsequent years, ACOs will also be gauged on a performance scale and earn more or less of their total potential bonus based on their quality.
Several of the quality measures are claims-based and will be calculated by CMS.  Others will require a survey, chart abstraction and the collection of lab values.  The physicians participating in an ACO could consider themselves a group and the ACO could use the group practice reporting option on their behalf to consolidate information and reduce reporting burden somewhat.
Eligible professionals participating in ACOs would be eligible for incentives through the Physician Quality Reporting (PQR), e-prescribing and Meaningful Use programs in addition to the bonuses they earn through the ACO.
The quality domains and the number of measures in each are

  • Experience of care, 7 measures;
  • Care coordination, 16 measures;
  • Patient safety, 2 measures;
  • Preventive health, 9 measures;
  • Population health, 31 measures.

The Measures

CMS has drawn from several familiar and several not-so-familiar sources for the ACO quality measures.   The experience of care measures are derived from the Consumer Assessment of Health Providers and Systems(CAHPS) for clinicians/groups and the Medicare Advantage (MA) CAHPS.  The care coordination domain includes CMS’s all-condition readmission rate, a 30-day post discharge physician visit measure, medication reconciliation, care transition, several of the Agency for Healthcare Research and Quality’s Prevention Quality Indicators (AHRQ’s PQIs), and some measures based on participation in the Meaningful Use incentive program.  Patient safety consists of a Health Care Acquired Conditions composite and an infection rate (the CLASBI bundle).  Immunizations, vaccinations, cancer screenings, and other health management processes from the PQR System are included in the preventive health domain.
The population health measures focus on six conditions/populations: diabetes, heart failure, coronary artery disease (CAD), hypertension, chronic obstructive pulmonary disorder, and the frail elderly. Many measures for population health are based on PQR processes.  Two of the population health measures (diabetes and CAD) are scored with an “all or nothing” method which measures how frequently a patient received all of the indicated care for their condition.  For example, if a patient receives four of the five indicated processes for CAD, the ACO gets no credit for this patient under this measure.
Quality Benchmarks
The benchmarks for each quality score will be based on the rates for Medicare’s fee for service (FFS) and MA populations.  To earn 1.1 point, an ACO’s performance would need to be equal to or better than the 30th percentile for the FFS and MA population scores.  Better performance will be scored on a linear points scale, reaching 2.0 points for performance at or above the 90th percentile for FFS and MA.
Translating Quality Scores into Bonus Payments 
The intermediate step in the quality scoring process is to roll up into 5 scores for each of the 5 domains. Finally, the percentage of points earned for each domain will be aggregated using an equal weighting method to arrive at a single percentage for the ACO annually.  Thus, each of the domains is given equal weight in the final score, though the number of measures in each domain varies widely.  The two measures in the patient safety domain will have a much greater impact on an ACO’s final score than any of the 31 measures of population health.
Each ACO’s quality information will reported publically, along with some identifying data on the ACO, the amount of shared savings bonuses earned, and the use of that bonus money to improve quality.
Other Performance Monitoring
CMS acknowledges that exposing some ACOs to the risk of losses might incentivize them to stint on care or avoid some at-risk beneficiaries.  ACO quality data will also be monitored by CMS to assess the performance of ACOs, along with site visits, analysis of beneficiary and provider complaints, audits, and an analysis of trends and patterns in the patients treated by ACO participants suggestive of avoidance of at-risk beneficiaries.  The results of this monitoring may require further investigation and could ultimately result in suspension or termination of the ACO’s contract with the shared savings program.