ACOs are required to completely and accurately report quality data that are used to calculate and assess their quality performance. In addition, in order to be eligible to share in any savings generated, an ACO must meet the established quality performance standard that corresponds to its performance year.
In the first performance year of their first agreement period, ACOs satisfy the quality performance standard when they completely and accurately report on all quality measures (pay-for-reporting). Complete and accurate reporting in the ACO’s first performance year qualifies the ACO for the maximum sharing rate. In subsequent performance years, quality performance benchmarks are phased-in for performance measures and the quality performance standard requires ACOs to continue to completely and accurately report quality data on all measures but the ACOs final sharing rate is determined based on its performance compared to national benchmarks.
In addition, ACOs must meet minimum attainment (30th percentile benchmark) on at least 1 pay-for-performance measure in each domain in order to be eligible to share in savings. Both attainment and improvement in performance are taken into account when calculating the final sharing rate for ACOs in their second and subsequent performance years. ACOs are rewarded up to four additional points in each domain, if they demonstrate quality improvement. In this way, the ACO becomes increasingly responsible for quality performance and improvement during the first agreement period. When an ACO renews its participation in the program for a second or subsequent agreement period, the quality performance of ACOs is assessed in the same manner as ACOs in the third performance year of their first agreement period.
Quality performance benchmarks are established by the Centers for Medicare and Medicaid Services (CMS) prior to the reporting period for which they apply and are set for 2 years. This document defines and sets the quality performance benchmarks that will be used for the 2016 and 2017 reporting periods. These benchmarks will apply to all Shared Savings Program ACOs reporting quality data in 2016 and 2017.
For the 2016 reporting year, CMS will measure quality of care using 34 quality measures (32 individual measures and 1 composite measure that include 2 individual component measures). The quality measures span four quality domains: Patient/Caregiver Experience, Care Coordination/Patient Safety, Preventive Health, and At-Risk Population. Because new quality measures introduced to the Shared Savings Program are set at the level of complete and accurate reporting for the first 2 years before phasing into performance, this document will be updated prior to the 2017 reporting year to include benchmarks for 7 measures (including the Diabetes Composite) that phase into performance for the 2017 reporting year. The benchmarks for each measure along with the phase-in schedule for pay-for-performance are displayed in the Table.
It is also important to note that CMS maintains the authority to revert measures from pay-for-performance to pay-for-reporting when the measure owner determines the measure causes patient harm or no longer aligns with clinical practice. Should CMS need to make such a modification, CMS will alert the ACOs through the Spotlight newsletter.
**Benchmark Data Sources**
These 2016 and 2017 benchmarks used all available and applicable 2012, 2013 or 2014 Medicare fee-for-service (FFS) data. This includes:
– Quality data reported through the Physician Quality Reporting System (PQRS) by physicians and groups of physicians through the Web Interface, claims, or a registry for the 2012, 2013, and2014 performance years, as available;
– Quality data reported by Shared Savings Program and Pioneer Model ACOs through the Web Interface for 2012, 2013 or 2014 performance years;
– Quality measure data collected from the Consumer Assessment of Healthcare Providers and Systems (CAHPS) for ACOs, CAHPS for PQRS and Medicare FFS CAHPS surveys administered for the 2012, 2013 or 2014 performance years;
Attestation and meaningful use data collected through the Electronic Health Record (EHR) Incentive Program for 2013 and 2014.
All of the quality measure benchmarks were calculated using ACO, group practice and individual physician data aggregated to the TIN level and included if there were at least 20 cases in the denominator. Quality data for ACOs, providers or group practices that did not satisfy the reporting requirements of the Shared Savings Program or PQRS were not included in calculation of the benchmarks.
Benchmarks for ACO Quality Measures
Benchmarks for the 23 of the 34 quality measures that are pay-for-performance for the 2016 and 2017 reporting years for an ACO’s second or third year of the ACO’s first agreement period are specified in the Table. ACOs in a second agreement period should refer to performance year 3 in the Table. In addition, the following 7 measure benchmarks will be released prior to the 2017 reporting year, because they phase into performance in 2017:
– ACO-34 Stewardship of Patient Resources
– ACO-35 Skilled Nursing Facility 30-Day All-Cause Readmission Measure (SNFRM)
– ACO-36 All-Cause Unplanned Admissions for Patients with Diabetes
– ACO-37 All-Cause Unplanned Admissions for Patients with Heart Failure
– ACO-38 All-Cause Unplanned Admissions for Patients with Multiple Chronic Conditions
– ACO-39 Documentation of Current Medications in the Medical Record
– Diabetes Composite (includes 2 component measures)
A quality performance benchmark is the performance rate an ACO must achieve to earn the corresponding quality points for each measure. The Table shows the benchmark for each percentile, starting with the 30th percentile (corresponding to the minimum attainment level) and ending with the 90th percentile (corresponding to the maximum attainment level). Under the Shared Savings Program’s regulation at 42 C.F.R. 425.502, there are circumstances when the set benchmark uses flat percentages. The use of flat percentages addresses issues with measures that have an overall high level of performance and allows ACOs with high scores to be recognized for their performance and earn maximum or near maximum quality points while also recognizing a range of performance levels allowing room for improvement and rewarding that improvement in subsequent years. For 15 measures, benchmarks using flat percentages use the 60th percentile when it is equal to or greater than 80.00 percent. For 3 measures, benchmarks using flat percentages use the 90th percentile when it is equal to or greater than 95.00 percent.
In efforts to maintain consistency across benchmarks, the ACO-10 Ambulatory Sensitive Conditions Admissions were also displayed: Heart Failure observed to expected ratio percentiles as percentages. ACOs can compare their previous annual performance scores for ACO-9 and ACO-10 by multiplying their observed/expected ratio performance rates with the following national means:
– ACO-9 national mean performance rate: 6.86%
– ACO-10 national mean performance rate: 18.19%
**Quality Scoring Points System**
The Table shows the maximum possible points that may be earned by an ACO in each domain and overall. An ACO achieves the maximum points for all measures designated as pay for reporting when the ACO completely and accurately reports. For measures that are pay for performance, quality scoring will be based on the ACO’s level of performance on each measure.
An ACO will earn quality points for each measure on a sliding scale based on level of performance. Performance below the minimum attainment level (the 30th percentile) for a measure will receive zero points for that measure; performance at or above the 90th percentile of the quality performance benchmark earns the maximum points available for the measure.
For most of the measures, the higher the level of performance, the higher the corresponding number of quality points. However, it is important to note that eight ACO quality measures have a reverse scoring structure, which means that a lower score represents better performance, and a higher score represents worse performance.
The following measures are scored such that a lower rate is indicative of better performance:
– ACO-8: Risk Standardized, all condition readmissions.
– ACO-9: Ambulatory Sensitive Conditions Admissions: for COPD or asthma in older adults.
– ACO-10: Ambulatory Sensitive Conditions Admissions: for heart failure (HF).
– ACO-27: Diabetes Mellitus: Hemoglobin A1c poor control.
– ACO-35: Skilled Nursing Facility 30-Day All-Cause Readmission Measure (SNFRM)
– ACO-36: All-Cause Unplanned Admissions for Patients with Diabetes
– ACO-37: All-Cause Unplanned Admissions for Patients with Heart Failure
– ACO-38: All-Cause Unplanned Admissions for Patients with Multiple Chronic Conditions
A maximum of 2 points can be earned for each scored individual or composite measure, except for the Percent of Primary Care Physicians who Successfully Met Meaningful Use Requirements measure (ACO-11). The ACO-11 measure is double weighted and is worth up to 4 points to provide incentive for greater levels of EHR adoption.
The matrix below shows the points earned for each measure at the corresponding decile value. For example, if an ACO’s performance rate for the Influenza immunization measure (ACO-14) is 72 percent or percentile, it would earn 1.70 points for that measure. Because the EHR measure (ACO-11) is double weighted, an ACO’s performance rate of 78 percent or percentile on that measure would earn 3.40 points.
**ACO Performance Level Quality points**
90+ percentile benchmark or 90+ percent 2.00 points
80+ percentile benchmark or 80+ percent 1.85 points
70+ percentile benchmark or 70+ percent 1.70 points
60+ percentile benchmark or 60+ percent 1.55 points
50+ percentile benchmark or 50+ percent 1.40 points
40+ percentile benchmark or 40+ percent 1.25 points
30+ percentile benchmark or 30+ percent 1.10 point
<30 percentile benchmark or <30+ percent No points
**Quality Improvement Reward**
Additionally, CMS will reward ACOs that demonstrate significant improvement in their quality measure performance by adding up to 4.00 points to each domain score. The total points in each domain cannot exceed the maximum points that are possible in that domain. For instance, an ACO may receive 4.00 additional points in the Preventive Health domain by demonstrating quality improvement; however, the ACO’s total points for the domain cannot exceed the maximum 18 possible points that can be earned for the Preventive Health domain.
The total points earned for measures in each domain, including any quality improvement points, will be summed and divided by the total points available for that domain to produce a domain score of the percentage of points earned relative to points available. The percentage score for each domain will be averaged together to generate a final overall quality score for each ACO that will be used to determine the amount of savings it shares or, if applicable, the amount of losses it owes.