Policy Solutions to Facilitate the New eCQM ACO Reporting and Advancement of Interoperability

,
The American Journal of Accountable Care, March 2022, Volume 10, Issue 1

This article describes the challenges associated with aggregating and reporting quality data via electronic health records and discusses corresponding policy solutions.

Am J Accountable Care. 2022;10(1):28-30. https://doi.org/10.37765/ajac.2022.88849

_____

The manner in which Medicare measures quality is undergoing a seismic shift. Currently, the CMS Web Interface is a popular option through which physician group practices and accountable care organizations (ACOs) can submit requisite quality data. Quality data submitted via the Web Interface are used to assess performance in alternative payment models (APMs) such as the Medicare Shared Savings Program in addition to calculating a Merit-based Incentive Payment System (MIPS) score, which determines adjustments to physicians’ Part B payments in subsequent years. The Web Interface requires physician groups and/or APM participants to submit data on 248 patients selected from a CMS sample of 616 Medicare patients. Starting in 2025, CMS will require APM participants to cease utilization of the Web Interface and instead submit electronic clinical quality measures (eCQMs) at the APM entity level.1

At a high level, eCQMs are quality measures that are generated and reported via certified electronic health record (EHR) technology. Many APM participants find eCQMs controversial because they include all patients who meet measure criteria, regardless of insurance status. Others argue that the measures themselves (examples in the Table) are not applicable to all specialties. This article will set aside debates regarding the appropriateness of the measure selection and specifications and instead focus on the reporting process. In theory, eCQMs are touted as reducing the manual burden associated with sampling and corresponding data abstraction while promoting the goals of interoperability. In reality, successfully building, aggregating, and reporting eCQM data at the APM entity level is rife with challenges. To its credit, CMS recognized these challenges and delayed mandatory eCQM reporting from 2023 to 2025.1 All this being said, the effort to optimize eCQM reporting should not be abandoned. The seamless exchange of health information among health systems, payers, and patients is a vital goal, and alterations to our information technology policy and infrastructure could facilitate eCQM reporting and simultaneously move the nation closer to achieving true interoperability. Policy makers should pursue a path to enabling the recommendations proposed in this article.

The Challenges of eCQM Reporting

Submitting eCQM data for a single practice on a single EHR platform is not inherently challenging. Real difficulties arise when an ACO with multiple practices on disparate EHRs attempts to submit data at the ACO level. Northwestern Medicine (NM) ACO contains 48 practices serviced by 16 EHR vendors. The 2 largest practices are NM-employed groups that both use Epic. As a result, NM ACO is aggregating all eCQM data and preparing it for submission within Epic (eAppendix [available at ajmc.com]). This ACO configuration and planned process for eCQM submission is not an uncommon one. The process steps and associated challenges are as follows:

  • Practice engagement: ACOs must collect eCQM data from member practices in the form of quality reporting document architecture (QRDA) 1 files for 3 eCQM measures (Table). Many ACO practices are small and lack substantive administrative and/or technological support. Clearly communicating the need for eCQM QRDA 1 files and how to transmit such files often requires significant at-the-elbow support. In many instances, NM communicated directly with practices’ EHR vendors on practices’ behalf to ensure that vendors had the ability to produce QRDA 1 files for all 3 eCQMs.
  • Successful file transfer: In order to submit eCQM data to CMS at the ACO level, the eCQM QRDA 1 files are ultimately aggregated within a single EHR. At present, most EHR vendors are still developing tools that will support aggregation of eCQM files from disparate EHRs. In the interim, some ACOs are transferring eCQM files via secure email or SharePoint. This manual process can result in errors: practices send QRDA 3 files instead of QRDA 1 files, files are not in an XML format, and/or data are missing. Once correctly formatted files are received, they can be ingested into a single EHR instance.
  • Aggregation and deduplication: When aggregating eCQM data from multiple ACO practices, it is quite probable that some patients will appear in multiple practices’ data. Successful eCQM measurement and submission requires the ability to complete patient matching. For example, if ACO practice A screened a patient for depression, and that patient subsequently had an encounter with ACO practice B but was not screened for depression, that patient should appear in the numerator of the eCQM depression screening measure. If aggregation and deduplication are unsuccessful, it’s possible that the patient could be counted as 2 unique patients: 1 who met numerator criteria (“pass”) and 1 who did not (“fail”). However, due to the fact that different health systems and EHR vendors use different unique patient identifiers (IDs), patient matching can be extremely challenging. The absence of a universal primary key means that developers must use patient name, date of birth, sex, and address to attempt to identify and match unique patients. However, addresses can be outdated, names can be misspelled, and some health systems may utilize a middle initial whereas others do not. As a result, this identification process is unlikely to generate 100% accurate results.

What Can Be Done?

To reduce the long-term burden associated with ACO-level eCQM reporting, CMS and the Office of the National Coordinator for Health Information Technology (ONC) should implement policies that (1) support real-time data exchange between disparate EHRs and (2) enable patient identification.

Facilitate real-time transfer. As previously stated, NM ACO’s current eCQM data exchange process involves multiple manual steps and significant time spent educating and supporting ACO practices. This process will have to be repeated each time an ACO wishes to collect new eCQM data for each performance period. However, if all ACO practices use the HL7 Fast Health Interoperability Resources (FHIR) standard, a mapping exercise of data elements between systems would need to occur only once. FHIR is essentially an interface that enables real-time data exchange. Once direct connections and eCQM data mapping are completed, any time a patient has an eCQM denominator–eligible encounter with an ACO practice, the encounter data will be sent in real time to the EHR system (in this instance, NM’s Epic) conducting the aggregation and deduplication.

The primary barriers to broad utilization of the FHIR standard are the time and cost associated with implementing FHIR. This resource requirement is especially difficult for small physician practices. However, broad utilization of FHIR is an undertaking worth pursuing due to the critical gains that extend well beyond eCQMs. FHIR would bring the country closer to achieving true interoperability: Patients’ health care data would seamlessly follow them from physician to physician, regardless of the practice’s EHR vendor. ONC should establish a timeline for the requisite adoption of the FHIR standard, and CMS could facilitate implementation through multiple policy levers. For example, adoption of FHIR could translate to automatic full credit for the Improvement and Promoting Interoperability categories of MIPS.

Establish a universal health care ID. As previously stated, patient matching is necessary to successfully aggregate and deduplicate data at the ACO level. When ACO practices operate on disparate EHRs, a common patient ID does not exist and ACOs must conduct an imperfect matching process utilizing patient name, sex, date of birth, and address. In the near term, CMS could enhance the QRDA format to include additional identifying fields such as email address. Although the incorporation of supplementary data fields would bring ACOs closer to complete patient matching, gaps would still exist.

The existence of a universal patient health care ID would eliminate this burdensome matching process and lead to the calculation of more accurate eCQM results. This concept is not a new one, and in the past it has been met with concerns regarding patient privacy. The issue of ensuring patient privacy should be addressed, and any unique patient ID should be protected like any other form of personal health information. Moreover, it is important to note that a unique health care ID would not link to patient financial information and may enable patients to avoid sharing their Social Security number in certain instances.2

Conclusions

Altering quality measurement methodology to more closely align with the goals of interoperability is a laudable goal, but additional policies must be implemented to enable real-time data exchange and efficient, accurate measure calculations. Employing the policy solutions presented in this article is admittedly a highly complex endeavor, but an important one that would benefit from collaboration among ONC, CMS, EHR vendors, and the ACO community.

Acknowledgments

The authors acknowledge the contributions of Jessica Prieskorn, Bijal Desai, and Adam Sloane.

Author Affiliation: Northwestern Medicine (JW, DB), Chicago, IL.

Source of Funding: None.

Author Disclosures: The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (JW, DB); drafting of the manuscript (JW, DB); and critical revision of the manuscript for important intellectual content (JW, DB).

Send Correspondence to: Jessica Walradt, MS, BA, Northwestern Medicine, 541 N Fairbanks Ct, Ste 1500, Chicago, IL 60611. Email: Jessica.Walradt@nm.org.

REFERENCES

1. CMS, HHS. Medicare program; CY 2022 payment policies under the physician fee schedule and other changes to Part B payment policies; Medicare Shared Savings Program requirements; provider enrollment regulation updates; and provider and supplier prepayment and post-payment medical review requirements. Fed Regist. 2021;86(221):64996-66031.

2. VanHouten JP, Brandt CA. Universal patient identification: what it is and why the US needs it. Health Affairs. July 7, 2021. Accessed January 24, 2022. https://bit.ly/3uKdXIf