Effective March 1, 2020, CMS temporarily waived the three-day inpatient hospitalization requirement for Medicare-covered skilled nursing facility (SNF) stays. This waiver allowed beneficiaries to receive Medicare-covered SNF care without a prior hospital stay of at least three days. It expired with the end of the federal Public Health Emergency (PHE) on May 11, 2023.

The three-day inpatient stay requirement for Medicare-covered SNF stays has been discussed by policymakers for years, even prior to the COVID-19 pandemic. Several CMS initiatives have waived this requirement, such as certain accountable care organization programs and the Bundled Payments for Care Improvement initiative. In non-PHE contexts, policymakers have considered whether a three-day stay waiver could lead to increased SNF stays in the absence of current requirements.

The PHE was the first time the waiver was applied more broadly, offering an opportunity to examine its potential for wider use.

Waiver Use and COVID-19 Status

A previous Avalere insight detailed patterns in the use of the three-day waiver from March 2020 through December 2021. In this updated analysis, Avalere analyzed claims for all SNF stays from March 2020 to May 2023 and the use of the three-day waiver among beneficiaries without a diagnosis for COVID-19 (i.e., non-COVID) to examine trends by admission source and for dual-eligible beneficiaries through the end of the PHE.

Utilization of the three-day waiver generally aligned with trends in COVID-19 case rates. Non-COVID waiver stays remained relatively constant, at approximately 15% of total SNF stays throughout the pandemic (Figure 1).

Figure 1. Medicare-Covered SNF Stays, by Waiver and COVID-19 Status, March 2020–May 2023

Dotted line represents the average US COVID-19 case rate. Users can hover over each line and month for number of stays by waiver and COVID-19 status and percentage of total SNF stays.
Sources: Avalere analysis of 100% Medicare fee-for-service claims data, accessed via a research collaboration with Inovalon, Inc. and governed by a research-focused CMS data use agreement.
COVID-19 weekly case data was from mid-month rates available from the Centers for Disease Control and Prevention.

Sources of Admissions

Avalere also analyzed the sources of admission for three-day waiver stays to identify trends in utilization prior to and following the availability of COVID-19 vaccinations (Figure 2).

Nursing Facilities (Skill in Place)

In 2020, 45% of non-COVID three-day waiver stays (6.5% of total SNF stays) were admitted from nursing facilities. Additionally, the proportion of beneficiaries using the three-day waiver who were admitted from nursing facilities decreased, from 12% of total SNF stays in April 2020 to 3% in April 2021. This change corresponds with the drop in COVID-19 weekly cases following the early distribution of vaccinations to nursing facility residents. From 2021 through the end of the PHE, non-COVID waiver stays admitted from nursing facilities accounted for about 28% of all non-COVID waiver stays (4% of total SNF stays). Medicare beneficiaries initiating SNF stays from nursing facilities may have been long-term nursing home residents in need of higher-level care due to an acute medical need. In many cases, the same provider cared for these beneficiaries, but the reimbursement shifted to Medicare.

Inpatient Hospitalization <3 Days and Emergency Room

In 2020, the second and third most common admission source of three-day waiver stays were stays following an inpatient hospitalization of less than three days or stays following an emergency room visit. In mid- to late-2021, the highest proportion of three-day waiver stays were admitted following inpatient hospitalizations of less than three days. Throughout the rest of the PHE, these two admission sources accounted for an average of 50% or more of all waiver non-COVID SNF stays.

Observation and Physician Office

During the PHE waiver period, the proportion of SNF stays admitted following an observation stay or physician office visits (community entrance) was constant. The proportion of non-COVID waiver stays admitted from these sources ranged from 1.9% to 6.8% of all waiver stays during the PHE.

Figure 2. Non-COVID-19 Three-Day Waiver SNF Stays, by Admission Source

Dotted line represents the average US COVID-19 case rate. Users can hover over each line and month for number of stays by admission source and percentage of all Waiver Non-COVID stays.
ER: Emergency Room; LOS: Length of Stay
Sources: Avalere analysis of 100% Medicare fee-for-service claims data, accessed via a research collaboration with Inovalon, Inc. and governed by a research-focused CMS data use agreement.
COVID-19 weekly case data was from mid-month rates available from the Centers for Disease Control and Prevention.

Pandemic-Era Patterns: Use Among Dual-Eligible Beneficiaries

Between January 2017 and February 2020, on average, 38% of all SNF stays were for beneficiaries dually eligible for Medicare and Medicaid (dual beneficiaries) each month. During the pandemic the percentage of total SNF stays (waiver and non-waiver) for dual beneficiaries ranged from 37% to 60% each month, with the highest percentage of stays for dual beneficiaries occurring between April and July 2020, November 2020 and January 2021, and January 2022 where more than 50% of SNF stays were for duals.

On average, waiver non-COVID stays for dual beneficiaries accounted for 8% of total SNF stays throughout the PHE. The majority of waiver non-COVID SNF stays for dual beneficiaries were admitted from a nursing facility (Figure 3). Generally, many long-term care residents are eligible for both Medicare and Medicaid. With the implementation of the three-day waiver during the PHE, beneficiaries in long-term care facilities with acute medical needs were able to receive skilled nursing services without transitioning to another facility. At the end of the PHE, the proportion of non-COVID SNF stays for duals admitted from nursing facilities decreased and a higher proportion were admitted following an inpatient stay.

Figure 3. Non-COVID-19 Three-Day Waiver SNF Stays for Dual Beneficiaries, by Admission Source

Users can hover over each line and month for percentage of Waiver Non-COVID stays by admission source among dual eligible beneficiaries.
ER: Emergency Room; LOS: Length of Stay
Sources: Avalere analysis of 100% Medicare fee-for-service claims data, accessed via a research collaboration with Inovalon, Inc. and governed by a research-focused CMS data use agreement.

Conclusion

The relatively constant use of the three-day waiver throughout the pandemic, alongside changes in use by admission source tracking with COVID-19 case rates and vaccine availability, suggest that use of the three-day waiver reflected dynamics of the pandemic environment. As vaccines became available and COVID-19 case rates decreased, direct transfer to Medicare-covered SNF stays from nursing facilities decreased. Overall, SNF stays remained relatively stable over the PHE, with slight increases corresponding to months with high COVID-19 case rates. This indicates that the implementation of the waiver was a tool for SNFs to provide the necessary level of care to Medicare beneficiaries during the COVID-19 pandemic.

Funding for this research was provided by the American Health Care Association. Avalere retained full editorial control.

Methodology

Avalere used the 100% Medicare fee-for-service claims data to analyze SNF stays from January 2017 through May 2023. Claims data were accessed via a research collaboration with Inovalon, Inc. and governed by a research-focused Centers for Medicare and Medicaid Services data use agreement. Waiver stays were identified by a condition code signifying the three-day waiver (condition code = DR). In cases where SNF stays were coded as waiver but had prior inpatient stays of 3 days or more, Avalere recoded waiver status to non-waiver for this analysis as these reflect Medicare-covered SNF stays under current law. SNF stays for patients treated for COVID-19 were identified based on the presence of International Classification of Diseases, 10th Revision diagnosis code U071 (COVID-19).