Late 2020 changes make accountable care organizations a canary in the coal mine.
Am J Accountable Care. 2021;9(2):4-6
CMS is expected to make updates to ACO quality reporting in the proposed 2022 Medicare Physician Fee Schedule. The information and arguments in this piece are up to date at the time of publication.
Current Health Information Technology Realities Affecting Health Care and Quality Assessments
In less than a decade, accountable care organizations (ACOs) have grown from nothing to caring for nearly 20% of all Medicare beneficiaries.1 Throughout this period of remarkable growth, quality improvement has been a critical component of Medicare ACO programs. A set of predefined quality measures incentivizes doctors, hospitals, and other providers in an ACO to provide optimal care coordination, preventive services, and chronic disease management. Shared savings rates—and whether an ACO receives any shared savings at all—are determined by quality performance.
Those incentives are working.2 The HHS Office of Inspector General found that ACOs outperformed fee-for-service providers on 81% of quality measures and improved quality over time in the program.3 ACOs also hit an average quality score of more than 94% in 2019, the latest year for which Medicare data are available.4
But despite ACOs’ positive track record on quality improvement, CMS is dramatically overhauling quality measurement within the Medicare Shared Savings Program (MSSP), CMS’ largest and most successful alternative payment model. In the waning weeks of the Trump administration, CMS finalized several changes, including the mechanism for which ACOs submit data to CMS, what is being measured, on whom quality is measured, ways in which shared savings are determined, the number of patients on which data are submitted—and the list goes on.5
In a survey of MSSP ACOs this spring, nearly 75% of respondents noted that they are either “extremely concerned” or “very concerned” about implementing the changes in the relatively short time allotted for the overhaul. Additionally, these new requirements come in the midst of a global pandemic, which providers and ACOs continue to fight on the front lines.
While this article outlines key issues with the ACO quality overhaul, the bigger picture is the poor state of interoperability in electronic health records (EHRs). The gap between the current state of EHR interoperability and what is needed to implement CMS’ changes is vast. Not nearly enough progress has been made in recent years to close that gap. This is a much larger issue across the health care industry, and EHRs’ interoperability problems cause a variety of hurdles impeding progress toward realizing the full benefits of EHRs and electronic data. The broader challenge of interoperability needs to be addressed before downstream changes can be successfully implemented, such as reasonably requiring ACOs to report electronic clinical quality measures (eCQMs).
In response to the MSSP quality overhaul, the National Association of ACOs (NAACOS) was recently joined by 10 other leading health care associations asking HHS Secretary Xavier Becerra for a delay in several aspects of these quality changes, along with other significant updates.6 As part of that request, the letter urged more time—at least 3 years—during which policy changes could be implemented and hopefully progress can be made to improve EHR interoperability to more easily shift to eCQMs, among other advancements to benefit patient care and ease provider burden. EHR vendors need time to make necessary changes to their products to smoothly implement new requirements. It takes at least 18 months to make updates to EHRs, and CMS provided far less than that. Additionally, CMS has yet to provide the industry with guidance on how certain issues will be addressed, including around patient matching and how data completeness will be calculated.
Specific Challenges With the ACO Quality Overhaul
Use of eCQMs
CMS’ stated goal of the quality measurement overhaul is to streamline the quality reporting process for ACOs. To do this, the agency is moving to focus on quality measures that are collected within EHRs and submitted through those same EHRs. Previously, much of the quality data were collected and submitted manually, which sounds onerous, but ACOs were increasingly efficient at this process.
Unlike in the Merit-based Incentive Payment System, or MIPS, which already uses eCQMs and measures quality for single clinicians, ACOs are large collections of providers who often do not otherwise have incentives to work together. Almost 40% of ACOs told NAACOS in a recent survey that they have more than 15 EHRs across their participants (Figure). EHRs cannot aggregate eCQM data across systems. That means ACOs will need to add another layer of technology to their EHRs to gather and combine eCQMs across their participants. In that spring survey, nearly 70% of ACOs reported that they do not have software in place to assist with integrating and extracting quality data from their participants’ EHRs.
Lack of standardized data
Related to EHRs’ interoperability problems, many health information technology systems collect and determine the same quality measures in different ways. EHRs can be highly individualized, with users allowed to customize their screens, prompts, and what documentation must be supplied in certain clinical situations. Because EHRs are so individualized, so too are the outputs on different quality measures. This results in large variability in quality scores across practices, even if their performance is basically the same.
Reporting data from all payers
CMS’ new reporting requirements have ACOs report data on all the patients seen by the ACO’s providers, not just those in Medicare or those assigned to their ACO. Since the 2012 inception of the MSSP, ACOs have been required to report on—and be evaluated on—only their ACO-assigned beneficiaries, making this a notable change and exponentially complicating ACO quality reporting. This will create more burdensome reporting, which is the opposite of CMS’ intended goal of these changes, and may show differences in groups of patients that may widen disparities, going against a top administrative priority.
Because ACO performances will be measured against one another, ACOs that serve sicker and more vulnerable patients will likely score more poorly. In other words, ACOs treating vulnerable populations have a different mix of payers and patients, which could cause them to appear to have lower quality. For example, many ACOs have relationships with federally qualified health centers (FQHCs) to provide care. FQHCs often provide care to individuals with multiple risk factors such as food insecurity, housing instability, or medical complexity. Including their all-payer data will likely not drive meaningful change and could penalize ACOs for treating more vulnerable populations.
Furthermore, this all-payer requirement will cause ACOs to need quality data from non-ACO providers and patients. The providers outside the ACO have no obligation or incentive to work with ACOs on reporting these data. Additionally, ACOs’ inability to directly access data will cause significant gaps in reporting along with substantial time and resources to collect data, all of which would be better spent improving patient care elsewhere.
Numbers of patients on whom data are submitted
Under the new rules, ACOs will have to report quality data on 70% of all patients. This is vastly different from previous requirements involving smaller samples of patients. ACOs can have tens of thousands of ACO-assigned patients just from Medicare, let alone other payers. Small ACOs can have upward of 100,000 patients across payers, and large ones can have more than 1 million patients. Asking for data on 70% of them is overwhelming and unnecessary. At the sizes at which ACOs operate, CMS can accurately judge quality using a defined sample of an ACO’s patients. Quality scores would be largely the same when asking for data on a sample as they would be when asking for data on 70% of patients. This is akin to pollsters asking a few thousand potential voters about their preference for political candidates vs 70% of the state’s or country’s voters.
Measures being scored
In 2019, ACOs were judged on quality in 23 different measures. Under CMS’ changes, that drops to 6 this year. Although the lower number of measures will reduce the reporting burden, we question if these new measures are too narrow or if they are the optimal measures for the program. They are focused on diabetes and blood pressure control, depression screening, and preventable admissions and readmissions. Although all are critical to patients’ health, other measures would also be appropriate, and these measures do not necessarily include specialists who may be involved in care. For example, if a patient has an annual skin exam and their diabetes is noted in the medical record, then the dermatologist they go to will be judged for the control of their diabetes, regardless of whether diabetes falls outside the dermatologist’s care.
ACOs can continue to be a leader in driving improvement in the overall health of patients, but these quality changes must be revisited to allow ACOs to continue to provide high-quality care. Moving to the process of digitally capturing quality data is necessary. But the state of EHRs today does not allow measures to be captured uniformly, data cannot be easily aggregated from disparate EHRs, and we have yet to identify an optimal set of measures that accurately reflect the performance of a wide array of clinicians. eCQMs simply are not ready for prime time. Delaying these changes by 3 years, as provider groups asked for in May, will give ACOs more time to prepare for changes, CMS more time to address implementation questions, and EHR vendors more time to make it possible to aggregate data from different EHRs.
With a new administration, a waning pandemic, and more focus on how to improve value-based care, now is a good time to reexamine quality measurement in health care. For the reasons we state here, there is a major gap between where the health care industry is with EHRs and interoperability and where it needs to be to reasonably require ACOs to report quality electronically. NAACOS hopes to work with CMS and other key stakeholders to find a better path for the short term while larger long-term challenges are overcome to pave the way for meaningful digital advancements to support quality and the overall shift to value-based care.
Author Affiliations: National Association of ACOs (DP, AB, HB), Washington, DC.
Source of Funding: None.
Author Disclosures: Mr Pittman is a senior policy adviser for the National Association of ACOs (NAACOS); the issues discussed in this article directly deal with accountable care organizations and the association’s membership. Ms Brennan is employed by NAACOS and has attended NAACOS conferences. Ms Bossley has several clients, including NAACOS, with members directly affected by Medicare Shared Savings Program changes; she provides consulting expertise on these issues and how they may potentially affect their members. Ms Bossley also received reimbursement from NAACOS for her time to analyze information and draft the manuscript.
Authorship Information: Concept and design (DP, AB, HB); acquisition of data (AB); analysis and interpretation of data (AB, HB); drafting of the manuscript (DP, AB, HB); critical revision of the manuscript for important intellectual content (AB); and administrative, technical, or logistic support (DP).
Send Correspondence to: David Pittman, BS, NAACOS, 601 13th Street NW, Ste 900 South, Washington, DC 20005. Email: email@example.com.
1. Shared Savings Program fast facts. CMS. January 1, 2021. Accessed May 28, 2021. https://www.cms.gov/files/document/2021-shared-savings-program-fast-facts.pdf
2. Bleser WK, Saunders RS, Muhlestein DB, Morrison SQ, Pham HH, McClellan MB. ACO quality over time: the MSSP experience and opportunities for system-wide improvement. Am J Accountable Care. 2018;6(1):e1-e15. doi:10.37765/ajac.2018.87439
3. Medicare Shared Savings Program accountable care organizations have shown potential for reducing spending and improving quality. HHS Office of Inspector General. August 2017. Accessed May 28, 2021. https://oig.hhs.gov/oei/reports/oei-02-15-00450.pdf
4. Highlights of the 2019 ACO program results. National Association of ACOs. Accessed May 28, 2021. https://www.naacos.com/highlights-of-the-2019-aco-program-results
5. Final policy, payment, and quality provisions changes to the Medicare Physician Fee Schedule for calendar year 2021. CMS. December 1, 2020. Accessed May 28, 2021. https://www.cms.gov/newsroom/fact-sheets/final-policy-
6. Re: request to delay mandatory reporting of eCQMs and MIPS CQMs and implement changes to the Medicare Shared Savings Program quality overhaul. National Association of ACOs. May 4, 2021. Accessed May 28, 2021. https://www.naacos.com/aco-coalition-letter-on-mssp-quality-overhaul