The 2020 performance year (PY) results for two of Medicare’s accountable care organization (ACO) programs, the Medicare Shared Savings Program (MSSP) and the Next Generation ACO (NGACO) model, have been released. This Health Capital Topics article will examine the key highlights of the 2020 MSSP and NGACO model performance results.

The MSSP was established by the Centers for Medicare & Medicaid Services (CMS) to achieve savings in how providers delivered healthcare services while maintaining the quality of those services. Providers could share in those savings by participating in the MSSP through an ACO, i.e., a group of healthcare providers who come together to provide coordinated high-quality care to patients.¹ Established in 2012, the MSSP was fundamentally changed in 2018 based on the program’s first six years of experience. Under this new “Pathways to Success” model, all ACOs (both new and current) had to choose between the new Basic Track and Enhanced Track.² Under the Basic Track, which is further divided into five levels – A, B, C, D, and E – an ACO is automatically advanced to the next track level at the start of each subsequent performance year.³ While levels A and B are one-sided risk models, levels C, D, and E are two-sided risk models, which progressively increase in risk (as well as in potential shared savings) up to 50% of savings and 30% of losses.⁴ The Enhanced Track is a two-sided model with much higher financial risk – up to 75% of shared savings or losses.⁵ For each ACO, CMS sets annual financial targets (termed benchmarks) under which the ACO must fall in order to be eligible for shared savings. The MSSP is the largest alternative payment model offered by CMS, with 513 participating ACOs reaching approximately 10.6 million patients in 2020.⁶ 

The NGACO model was built upon CMS’s experience from previous ACO initiatives such as the MSSP. Established in 2016, this model sought to set predictable financial targets, give providers more opportunities to coordinate care to beneficiaries, and ensure high quality care and examine whether the combination of financial incentives coupled with increased patient care interaction/management can lower expenditures for fee-for-service recipients.⁷ Toward that end, the NGACO model allows physicians to take on higher financial risks than the MSSP. There were only 37 NGACOs in the program in 2020, serving 1.1 million patients.⁸

CMS’s report on the MSSP’s performance results for 2020 (the program’s eighth performance year) found that participating ACOs generated savings of approximately $4.145 billion above their benchmarks, the highest amount of program savings to date.⁹ This resulted in the participants receiving a total shared savings of $2.3 billion, and Medicare experiencing net savings of approximately $1.9 billion; this was Medicare’s fourth consecutive year of net savings.¹⁰ In fact, 83% of all MSSP ACOs reduced spending relative to their own benchmarks (a record high), and 67% reduced their spending sufficient to achieve shared savings.¹¹ Only six MSSP ACOs qualified for shared losses (which were waived in light of the COVID-19 public health emergency).¹² Notably, ACO benchmarks declined in 2020, requiring participants to spend less to potentially achieve savings, rendering these record-breaking figures all the more notable.¹³

CMS performance data showed that ACOs that take on financial risk are more likely to reduce costs and achieve savings. The grand majority (88%) of MSSP ACOs that operated under two-sided risk models (wherein they are on the hook for any losses) achieved shared savings, while just over half (55%) of one-sided risk models earned shared savings.¹⁴ Further, two-sided risk models saved Medicare $211 per beneficiary, while one-sided risk models saved $152 per beneficiary (38% less).¹⁵

The data also indicated that more experienced ACOs are more likely to achieve savings. Approximately 80% of mature ACO participants (i.e., those who entered the program between 2012 and 2014) earned shared savings in 2020, compared to 59% of newer ACO participants (i.e., those who entered the program between 2018 and 2020).¹⁶ This confirms that implementing the processes required to reduce costs and achieve shared savings is a years-long process.¹⁷

Further, the data showed that the type of provider leading the ACO may result in more savings. Physician-led ACOs saved Medicare $218 per beneficiary in 2020, compared to hospital-led ACOs, which saved $168 per beneficiary.¹⁸ Interestingly, ACOs led by both physicians and hospitals saved even less – only $145 per beneficiary.¹⁹ This dichotomy may indicate increased savings for Medicare going forward, as more and more ACOs are being led by physician groups.²⁰

It is important to note that, despite streamlining costs, MSSP ACOs maintained high quality care, receiving an average quality score of 97.8%, the best score to date.²¹ The 2020 MSSP performance results indicate that ACOs continue to have an impact in improving healthcare quality while lowering costs.²² The report findings also suggest that MSSP ACOs might be better positioned to deliver care than other providers during public health emergencies and advance valuable healthcare delivery trends such as telehealth.²³

Approximately two months after CMS released the MSSP performance report, the agency released 2020 performance data related to the NGACO model’s fourth performance year. Similar to the MSSP, NGACOs also saved CMS money in 2020, with 35 of the 37 NGACOs generating savings compared to their benchmark.²⁴  NGACOs generated $637 million in gross savings, but Medicare netted only $230 million after doling out shared savings payments; both of these amounts were higher than a year prior.²⁵ Even while reducing spending, NGACOs provided high quality care, receiving an average quality score of 96.5% (a slightly lower average score than the MSSP ACOs).²⁶

Further, the nonpartisan and objective research organization (NORC) at the University of Chicago utilized the 2020 NGACO data in evaluating the overall performance of the NGACO model in its first four years.²⁷ Its recently-released report found that NGACOs reduced acute care hospital stays and spending, with acute care spending declining by 0.9%.²⁸ Importantly, NGACOs’ total Medicare spending reductions were substantially larger for patients with eight or more chronic conditions (a reduction of approximately $755 per beneficiary) and for patients with prior hospitalizations (a reduction of approximately $410 per beneficiary), highlighting the success of the model’s focus on care coordination.²⁹

Despite this seemingly positive news (and the savings generated in 2020), the NGACO model has not been successful at reducing Medicare spending in aggregate.  After accounting for the shared savings and coordinated care reward³⁰ payments of $909.6 million over the last four years, the NORC report found that the NGACO model actually increased net Medicare spending by 0.4% during the first four performance years.³¹

The results of the 2020 performance year for MSSP ACOs were overwhelmingly positive, despite having a difficult year in 2020 due to the COVID-19 pandemic. However, despite gross savings and a generally positive year, the short-lived NGACO model ultimately increased Medicare spending during its four-year program. This lack of aggregate savings is largely the reason for the model ending at the conclusion of 2021.³² The model was originally set to end in 2020, but extended it for an additional year due to the pandemic, in order to give NGACO model participants sufficient time to transition to another two-sided risk payment model; in particular, CMS is encouraging NGACOs to transition to the Direct Contracting model.³³ The Direct Contracting Model is a voluntary, five-year Medicare ACO model that aims to reduce administrative burden though partially- and fully-capitated payments for Medicare Part A and B services.³⁴ Given the cancellation of the NGACO model, it is increasingly likely that CMS and its Center for Medicare and Medicaid Innovation (CMMI) could introduce a new advanced payment model for 2023, or at least significantly “tweak” the Medicare Direct Contracting model.³⁵