eCQM Benchmarking Paper Submission Document

  • by Vini Ehsan

Research Objective:
Clinical quality measures based on claims are commonly used to determine clinician and health care facility may contribute to performance as part of a value-based payment system. A new approach uses electronic clinical quality measures (eCQMs) that are generated from electronic health records (EHRs). eCQMs may provide a more complete picture of care than claims-based measures. This study created eCQM performance benchmarks for determining whether primary care practices in the Comprehensive Primary Care (CPC) initiative should receive a shared- savings, value-based payment. CPC is a patient centered medical home model sponsored by the Center for Medicare and Medicaid Innovation at the Centers for Medicare & Medicaid Services (CMS).
No other Medicare program has used EHR reported eCQM performance data in value- based payment to date. The results of this study raise important implications for alternative payment models. Study Design: We began by considering CMS Physician Quality Reporting System (PQRS) group reporting data quality and reliability. We determined data quality by analyzing the percentage of records with missing values, invalid values, missing or invalid performance rates, and high exclusion or exception rates. We estimated reliability values for all reporting groups, which were then aggregated to estimate overall reliability for each measure. We calculated benchmarks at the 25th, 50th, and 75thpercentiles of the distribution of each measure.

Population Studied:
Calendar year 2015 was the first in which eCQM performance factored into a shared savings payment for practices participating in CPC. To be eligible for shared savings, practices were required to meet performance benchmarks for 9 of 11 possible eCQMs. To construct these benchmarks, we used eCQM performance data from group practices that reported electronically to the CMS PQRS.

eCQM Benchmarking Paper Submission Document

  • by Vini Ehsan

Principal Findings:
PQRS group data had very few missing or invalid values. For every measure, more than 90 percent of PQRS groups achieved at least 70 percent statistical reliability, indicating that the data were sufficiently reliable to benchmark. The benchmarks showed considerable variation across measures. CPC practices performed better than PQRS group reporters. For example, for 9 of 11 measures, more than 60 percent of CPC practices scored above the median PQRS score. For 8 measures, at least 40 percent of CPC practices scored in the top 25 percent of PQRS values.

Conclusions:
Performance rates from group practices that report electronically to PQRS can be used to create benchmarks for value-based payment programs. Performance on clinical quality measures by CPC practices was better than in group practices that reported to PQRS. While we could not judge the accuracy of data of groups who reported to PQRS, this limitation is not unique to eCQM data; data could be incorrectly recorded for other reporting mechanisms.

Implications for Policy or Practice:
Payers and provider organizations may use data from eCQMs to establish valid and reliable performance standards under value-based payment arrangements. For example, CMS will use eCQM data to set performance standards for Comprehensive Primary Care Plus, an advanced alternative payment model under the Quality Payment Program.