Overview

The Centers for Medicare & Medicaid Services (CMS) has added new quality information to the Physician Compare website.

The Affordable Care Act required the establishment of the Physician Compare website. The goal of the website is to help patients and caregivers make informed choices about the Medicare physicians and other clinicians they see; publicly reporting both new and updated 2016 performance information will help further that goal. The first quality measures were added to Physician Compare in February 2014. Since then, CMS has continued a phased approach to public reporting.

2016 Performance Information on Physician Compare

As the next step in CMS’s phased approach to public reporting, starting in December 2017 CMS is publicly reporting certain 2016 performance information on Physician Compare. The information was designated as available for public reporting in the 2016 Physician Fee Schedule final rule.

Data are available for public reporting on public-facing profile pages and/or via the Physician Compare Downloadable Database available on data.medicare.gov. Because of the different primary audiences, CMS publicly reports information differently in the Downloadable Database than on profile pages.

The primary audience for profile pages is patients and caregivers. On the profile pages, groups may have the following measures reported: a subset of 2016 Physician Quality Reporting System (PQRS) measures reported as star ratings; Consumer Assessment of Healthcare Providers and Systems (CAHPS) for PQRS summary survey measures; and/or non-PQRS Qualified Clinical Data Registry (QCDR) measures.

The 2016 non-PQRS QCDR measures, available for both groups and individual clinicians, and 2016 CAHPS for PQRS measures available for only groups, are being reported as a percent – not as star ratings – on the relevant profile pages.

Group CAHPS for PQRS summary survey measures are reported as top-box scores on Physician Compare. The top-box score is the percentage of responses in the most positive response categories. Based on testing with patients and caregivers and the recommendation from the Agency for Healthcare Research and Quality (AHRQ), Physician Compare publicly reports top-box scores to ensure patient experience data is interpreted correctly.

The Downloadable Database is mainly intended as a resource for clinicians and group representatives, as well as third-party data users. Groups may have 2016 PQRS measures, CAHPS for PQRS summary survey measures, and/or non-PQRS QCDR measures included in the Downloadable Database. Individual clinicians may have 2016 PQRS and non-PQRS QCDR measures, as well as 2015 utilization data reported in the Downloadable Database. The final 2016 data will be available for download in late spring or early summer of 2018 once the informal review process has been completed. At this time, the most current general information (updated every two weeks) and performance information for 2015 is available for download.

The measures now included on Physician Compare profile pages represent a variety of types of clinical care by groups representing many different specialties. The 2016 PQRS performance information is divided into eight different categories, ranging from general care to more specialized care.

The categories include:

• Preventive care: General health
• Preventive care: Cancer screening
• Patient safety
• Care planning
• Diabetes
• Heart disease
• Respiratory diseases
• Behavioral health

In addition to the measures being reported for groups and individual clinicians, 2016 data for the Shared Savings Program, Pioneer, and Next Generation Accountable Care Organizations (ACOs) are now also publicly reported on Physician Compare. ACO measures are reported as percent performance rates. CAHPS for ACO summary survey measures are reported as means. In this way, the CAHPS for ACO performance scores are the average of all responses available for each summary survey measure.

Physician Compare also now includes information about group ACO affiliation. If a group is part of an ACO, there will be a link to that ACO’s Physician Compare profile page from the group profile page.

The criteria for public reporting require that the measures must be statistically valid, reliable, accurate, and comparable across reporting mechanisms, and meet the minimum reliability threshold, to be included in the Physician Compare Downloadable Database. To be included on public facing profile pages, the data must also prove to resonate with patients and caregivers.

Star Ratings on Physician Compare

When public reporting performance information began in 2014, CMS started with a small subset of group-level Web Interface measures. Now with the use of star ratings, CMS is restarting its phased approach for the use of star ratings with just a small subset of group-level measures. For the first time this year as part of the continued phased approach to public reporting, CMS has publicly reported a small subset of 2016 PQRS group-level measures on group profile pages as star ratings.

Since star ratings are new to Physician Compare this year, it is important to understand how the star rating is constructed.

After extensive research and outreach, and hearing what stakeholders wanted to see in a benchmark and understanding the concerns and cautions raised, CMS proposed an item-level (or measure-level) benchmark using the Achievable Benchmark of Care (ABC™) methodology. This benchmark was finalized in the CY 2016 Physician Fee Schedule final rule (80 FR 71128 through 71129). The ABC™ benchmark is the “5-star rate,” serving as the anchor for the star rating methodology.

As discussed in the CY 2018 Quality Payment Program final rule (82 FR 53827 through 53829), the process for choosing a star rating attribution method built upon CMS’s work in choosing the ABC™ benchmark methodology. After reviewing feedback, conducting extensive statistical analysis, and consulting with the TEP, the equal ranges method was chosen. The equal ranges method for assigning star ratings is intuitive for patients and caregivers to interpret, and reflects the true performance on the measure rather than forcing a distribution. Additionally, CMS expects star rating assignments based on the equal ranges method to be more stable across years, allowing clinicians to better assess their performance year-to-year. The equal ranges method also provides a more reliable and meaningful classification, ensuring that a 4-star performance is statistically better than and distinct from a 3-star performance on a measure, for example.

The ABC™ methodology for the benchmark and the equal ranges method for assigning star ratings were chosen because together they allow the public reporting of a statistically sound and easy-to-interpret set of data. They also give a point of comparison to help patients and caregivers interpret the performance information on Physician Compare. For more information about star ratings, visit the CMS Physician Compare Initiative page.

More Information on Physician Compare

For more information on Physician Compare, please visit the website at http://www.medicare.gov/physiciancompare. You can also go to www.medicare.gov and click on “Find doctors & other health professionals”. Clinicians can visit the CMS Physician Compare Initiative page for information on keeping their general information current and troubleshooting problems, and to learn more about public reporting.

If you have questions about public reporting on Physician Compare or keeping your information current, contact the Physician Compare support team at PhysicianCompare@Westat.com.

###