Impact of skilled nursing facility (SNF) 3‐day hospitalization requirement waiver during the COVID‐19 pandemic on emergency department and inpatient SNF discharges in California

Abstract

Objective

We sought to study the impact of the Centers for Medicare & Medicaid services (CMS) waiver of the 3‐day hospitalization requirement for skilled nursing facility (SNF) care implemented as part of the Federal COVID‐19 response on emergency department (ED) and inpatient hospital SNF discharges.

Methods

We conducted a multicenter retrospective cohort study of hospital ED and inpatient visits in California during 18 months before (prewaiver, September 2018–February 2020) and 18 months after (waiver, March 2020–August 2021) waiver implementation. Data were collected from all adult ED and admitted patients utilizing California Department of Health Care Access and Information datasets from all acute care hospitals licensed in the state. Prewaiver and waiver periods were compared for SNF discharge/disposition rates stratified by patient demographic and hospital data with differences in the proportion and 95% confidence interval [CI] reported (SPSS).

Results

SNF discharges decreased from the prewaiver to waiver periods from the ED (−7.4% [CI −8.1%, −6.6%]), along with larger declines occurring from the inpatient hospital setting (−18.1% [CI −18.4%, −17.9%]). For Medicare beneficiaries, there was a smaller decrease in ED SNF rates (−3.8% [CI −4.7%, −2.9%]), and there was no significant change for SNF discharge rates for inpatient admissions with a length of stay (LOS) <3 days (+1.0% [CI 0.0%, 2.1%]).

Conclusion

In California, the CMS waiver did not result in an increase, but an actual decrease rate of SNF discharges from the ED and inpatient setting, though with smaller declines for the ED, Medicare patients, and those with a LOS <3 days.

1. INTRODUCTION

1.1. Background

In response to the worldwide COVID‐19 pandemic, the US Federal government initiated a number of measures to address health care delivery and hospital capacity challenges in the face of the unprecedented public health emergency. As part of this effort, on March 1, 2020, the Centers for Medicare & Medicaid services (CMS) waived a set of rules governing reimbursement for skilled nursing facilities (SNFs) admissions. ¹ Specifically, CMS waived the requirement for a preceding 3‐day acute care hospitalization prior to a covered or reimbursed SNF stay for beneficiaries.

1.2. Importance

Under this waiver, Medicare patients could be covered for SNF care after discharge directly from the emergency department (ED) or from an acute care hospital stay of less than 3 days. In theory, the waiver could improve hospital inpatient capacity by allowing more patients to be discharged to SNFs following shorter acute hospitalizations, or even without such hospitalizations by direct transfer from an ED (Figure 1). ² The impact of the 3‐day waiver during the pandemic on ED and hospital care and discharges has yet to be fully understood. Moreover, concerns have been raised that the elimination of the waiver with the end of the public health emergency declaration in May 2023 could exacerbate current ED and hospital capacity challenges. ³

FIGURE 1.

Skilled nursing facility (SNF) placement/coverage pathway for FFS (fee‐for‐service) Medicare patients. Abbreviations: ED, emergency department; LTC, long‐term care facility; LOS, length of stay.

1.3. Goals of this study

In this study, we evaluated the impact of the 3‐day SNF waiver on ED and inpatient hospital discharges before and after its implementation in the state of California during the pandemic. Specifically, we sought to analyze whether the waiver resulted in more patients accessing SNF care, either following a shorter inpatient hospital admission, or directly from an ED visit with no inpatient admission, and whether there were significant differences by insurance coverage or other demographic variables. We hypothesized that the waiver program should have resulted in higher and earlier utilization of SNF stays out of the ED and inpatient hospital settings, particularly for Medicare beneficiaries.

2. METHODS

2.1. Study design and setting

We conducted a multicenter retrospective cohort study of hospital ED and inpatient visits from all nonfederal acute care hospitals in the state of California during a 3‐year period before and after implementation of the CMS 3‐day SNF waiver. The Institutional Review Board of the University of California, San Diego approved this study with a waiver of informed consent because of the use of deidentified data that did not constitute human participant research in accordance with the university's Human Research Protection Program guidelines.

2.2. Selection of participants

We evaluated all ED and hospital inpatient visits in the state of California over a 3‐year period. Hospital and ED data were obtained from the California Department of Health Care Access and Information (HCAI, formerly known as the Office of Statewide Health Planning and Development). HCAI collects information from all acute care hospitals and hospital systems licensed in the state of California (excluding Veterans or Department of Defense facilities). ⁴ These data are reported real‐time in a standardized format and do not involve claims data.

2.3. Measurements and outcomes

Data reported here were from two separate HCAI encounter‐based data sources. The Patient Discharge Dataset included all hospital discharges and Emergency Department Dataset (EDD) included all ED discharges. Patients included in the Patient Discharge dataset who were admitted from the same hospital's ED were identified and combined with all ED discharges from the EDD to construct a complete ED utilization database. Utilization can be assessed over time by using a record linking number. Visits for adult patients 18 years or older during the study period were included for analysis.

To assess the impact of the 3‐day waiver, we analyzed data for ED and hospital patients over 3 years, including an 18‐month period, from September 1, 2018 through February 29, 2020, before the 3‐day waiver (prewaiver), and an 18 month period following the waiver, from March 1, 2020 through August 31, 2021 (waiver).

Outcome measures included ED and inpatient visit/utilization rates, payer mix, disposition, and diagnoses. Measures included standardized reported data on demographic information (age, gender, ethnicity/race), service date, hospital, primary source of payment or insurance coverage, and discharge disposition. Patient characteristics, including age, gender, race/ethnicity, expected payer, and discharge disposition, are reported.

2.4. Analysis

ED and inpatient utilization including admit and discharge rates overall were evaluated and compared between the 2 study periods. Visit data for prewaiver and waiver periods were stratified based on disposition, payer, and other demographic data. The difference in the proportion and 95% confidence interval (CI) was calculated assessing trends over the 36‐month time period studied. Hospital SNF discharges hospitals by payer status and hospital length of stay (LOS) between the two study periods are also reported. All statistical analyses were conducted using the IBM SPSS Statistics for Windows, Version 28.0 software package (SPSS, Inc).

3. RESULTS

During the 3‐year study, ED and hospital admission patient data were obtained from 382 non‐Federal acute care hospitals in California. Overall, from March 2018 through August 2021, there were a total 33,273,521 patients seen in the ED, of which 154,948 were discharged directly to a SNF. Total ED census decreased from the prewaiver to waiver period, from 17,882,515 to 15,391,006 visits respectively, as did discharges to SNFs, from 86,207 to 68,741 patient discharges, respectively. The mean number of monthly ED visits was 993,473 (SD 39,539) and ranged from 909,299 to 1,075,275 (January 20) during the prewaiver period. During that time, the mean monthly SNF discharges was 4789 (SD 209) and ranged from 4294 to 5091 (January 19). The mean number of monthly ED visits decreased to 855,056 (SD 101,623) during the waiver period, and ranges from 600,740 to 1,028,629 (July 21). ED discharges to SNFs followed a similar pattern. The ED admission rate during the prewaiver period was 18.2 and 19.9% during the waiver period. Figure 2 demonstrates monthly ED visits, ED SNF discharges, and SNF discharge per 100,000 ED visits during the study period.

FIGURE 2.

(A) ED visits, (B) ED SNF discharges, and (C) ED SNF discharges per 100,000 ED visits by month. Abbreviations: ED, emergency department; SNF, skilled nursing facility.

Demographic data for ED visits and ED SNF discharges for prewaiver and waiver periods are displayed in Table 1. Patients discharged from the ED to a SNF were older and more likely to have Medicare coverage than all other ED patients (Table 1).

TABLE 1.

All ED visit and SNF discharge characteristics before and during the COVID‐19 public health emergency waiver.

	All ED visits		ED SNF discharges
Characteristic	Before the waiver	During the waiver	Before the waiver	During the waiver
Number of visits	17,882,515	15,391,006	86,207 (0.48)	68,741 (0.45)
Age (mean, SD)	49.6, 20.4	49.5, 20.1	75.6,14.9	75.3,14.6
Expected payer (%)
Private	4,985,754 (27.9)	4,414,558 (28.7)	7945 (9.2)	5274 (7.7)
Medicare	5,025,257 (28.1)	4,206,825 (27.3)	68,610 (79.6)	55,247 (80.4)
Medi‐Cal	6,454,228 (36.1)	5,661,831 (36.8)	8623 (10.0)	7487 (10.9)
Self‐pay	1,411,540 (7.9)	1,102,941 (7.2)	1005 (1.2)	723 (1.1)
Sex (%)
Male	8,022,998 (44.9)	7,252,462 (47.1)	36,717 (42.6)	30,233 (44.0)
Female	9,858,497 (55.1)	8,137,274 (52.9)	49,487 (57.4)	38,507 (56.0)
Race/ethnicity (%)
Hispanic or Latino	6,343,497 (35.5)	5,697,512 (37.0)	15,109 (17.5)	11,170 (16.2)
NH White	7,146,064 (40.0)	5,955,762 (38.7)	51,142 (59.3)	41,626 (60.6)
NH Black	2,035,163 (11.4)	1,698,965 (11.0)	9064 (10.5)	7085 (10.3)
NH Asian or Pac Isl	1,209,384 (6.8)	1,030,000 (6.7)	6505 (7.5)	5038 (7.3)
Other or unknown	1,148,407 (6.4)	1,008,767 (6.6)	4387 (5.1)	3822 (5.6)

Abbreviations: ED, emergency department; SD, standard deviation; SNF, skilled nursing facility.

There were a total of 8,771,439 hospital inpatient discharges, of which 894,261 were discharges to a SNF during the 3‐year period. Similar to ED discharges, there was a decline in inpatient discharges from prewaiver to waiver periods (4,601,182 to 4,170,257, respectively), as well as inpatient SNF discharges (513,349–380,912, respectively). Figure 3 demonstrates monthly Hospital discharges, Hospital inpatient SNF discharges, and SNF discharge per 100,000 Hospital discharges during the study period. The mean number of monthly inpatient discharges was 255,621 (SD 8137) and ranged from 241,667 (February 19) to 271,570 (March 19) during the prewaiver period. The mean monthly SNF discharges during that time was 28,519 (SD 1515) and ranged from 25,712 to 31,805 (January 20). The mean number of monthly inpatient discharges decreased to 231,681 (SD 16,825) during the Waiver period and ranged from 179,378 to 251,956 (August 21). Inpatient discharges to SNFs followed a similar pattern.

FIGURE 3.

(A) Hospital discharges,  (B) SNF discharges, and (C) SNF discharges per 100,000 hospital discharges by month.

Demographic data for inpatient admissions and SNF discharges for prewaiver and waiver periods are displayed in Table 2. Similar to findings in the ED, patients discharged from the inpatient setting to a SNF were older and more likely to have Medicare insurance.

TABLE 2.

All hospital and SNF discharge characteristics before and during the COVID‐19 public health emergency waiver.

	All hospital discharges		Hospital SNF discharges
Characteristic	Before the waiver	During the waiver	Before the waiver	During the waiver
Number of visits	4,601,182	4,170,257	513,349 (11.2)	380,912 (9.1)
Age (mean, SD)	57.3, 20.8	56.7, 20.6	74.5, 13.6	73.0, 13.8
Expected payer (%)
Private	1,239,921 (26.9)	1,151,115 (27.6)	39,035 (7.6)	29,221 (7.7)
Medicare	1,963,997 (42.7)	1,712,721 (41.1)	409,410 (79.8)	291,594 (76.6)
Medi‐Cal	1,293,126 (28.1)	1,231,552 (29.5)	62,482 (12.2)	58,299 (15.3)
Self‐pay	103,372 (2.2)	73,030 (1.8)	2366 (0.5)	1696 (0.4)
Sex (%)
Male	1,978,234 (43.0)	1,850,500 (44.4)	237,333 (46.2)	186,363 (48.9)
Female	2,622,767 (57.0)	2,319,543 (55.6)	275,996 (53.8)	194,535 (51.1)
Race/ethnicity (%)
Hispanic or Latino	1,354,660 (29.4)	1,328,504 (31.9)	96,800 (18.9)	76,252 (20.0)
NH White	2,106,062 (45.8)	1,803,077 (43.2)	290,535 (56.6)	209,496 (55.0)
NH Black	411,707 (8.9)	369,452 (8.9)	47,944 (9.3)	38,509 (10.1)
NH Asian or Pac Isl	438,595 (9.5)	391,132 (9.4)	47,334 (9.2)	32,628 (8.6)
Other or unknown	29,015 (6.3)	278,092 (6.7)	30,736 (6.0)	24,027 (6.3)

Abbreviations: SD, standard deviation; SNF, skilled nursing facility.

Overall, the rate of ED SNF discharges declined by −7.4% [CI −8.1%, −6.6%] from prewaiver to waiver periods (482.1–446.6 per 100,000 ED discharges, respectively). The greatest decline occurred in patients with Private insurance (−25.0% [CI −72.2%, −22.8%]), with a much smaller decline in patients with Medicare (−3.8% [CI −4.7%, −2.9%]) (Table 3).

TABLE 3.

ED and inpatient SNF discharge data by payer status prewaiver and waiver periods.

	Prewaiver			Waiver
Characteristic	ED SNF discharges	All ED visits	ED SNF discharges per 100K visits	SNF discharges	All ED visits	ED SNF discharges per 100K visits	Percent change	95% CI
ED DISCHARGES TO SNF	86,207	17,882,515	482.1	68,741	15,391,006	446.6	−7.4	(−8.1, −6.6)
Expected payer
Private	7945	4,985,754	159.4	5274	4,414,558	119.5	−25.0	(−27.2, −22.8)
Medicare	68,610	5,025,257	1,365.3	55,247	4,206,825	1313.3	−3.8	(−4.7, −2.9)
Medi‐Cal	8623	6,454,228	133.6	7487	5,661,831	132.2	−1.0	(−3.6, 1.6)
Self‐Pay	1005	1,411,540	71.2	723	1,102,941	65.6	−7.9	(−15.3, −0.5)
Hospital discharges to SNF	513,349	4,601,182	11,156.9	380,912	4,170,257	9134.0	−18.1	(−18.4, −17.9)
Expected payer
Private	39,035	1,239,921	3148.2	29,221	1,151,115	2538.5	−19.4	(−20.4, −18.4)
Medicare	409,410	1,963,997	20,845.8	291,594	1,712,721	17,025.2	−18.3	(−18.6, −18.0)
Medi‐Cal	62,482	1,293,126	4831.9	58,299	1,231,552	4733.8	−2.0	(−2.9, −1.1)
Self‐Pay	2,366	103,372	2288.8	1696	73,030	2322.3	1.5	(−3.8, 6.7)

Abbreviations: CI, confidence interval; ED, emergency department; SNF, skilled nursing facility.

For hospital discharges, the rate of SNF discharges decreased by a greater proportion, from 11.157 to 9,134 per 100,000 hospital discharges (−18.1% [CI −18.4%, −17.9%]). Decrease rates were similar for Private and Medicare insurance coverage (−19.4% [CI −20.4%, −18.4%] and −18.3% [CI −18.6%, −18.0%], respectively) (Table 3).

In stratifying by inpatient LOS, hospital inpatient SNF discharges declined after the implementation of the waiver for patients with LOS < 3 days, as well as those with stays >/= 3 days, but the drop was less for those with shorter inpatient stays (−7.9% [CI −8.8%, −7.0%] vs. −20.5% [CI −20.8%, −20.3%], respectively). This trend was even more pronounced for patients with Medicare where there was no significant decrease in inpatient SNF discharge rates for those patients with a LOS <3 days after waiver implementation, but decreased by −21.8% [CI −22.1%, −21.5%] for those with a longer hospital stays (Table 4).

TABLE 4.

Hospital SNF discharges rates by payer status and hospital length of stay prewaiver and waiver periods.

	Prewaiver			Waiver
Characteristic	SNF discharges	All hospital discharges	SNF discharges per 100K visits	SNF discharges	All hospital discharges	SNF discharges per 100K visits	Percent change	95% CI
Overall
<3 days LOS	60,223	1,954,208	3081.7	49,119	1,731,224	2837.2	−7.9	(−8.8, −7.0)
>=3 days LOS	453,126	2,646,974	17,118.6	331,793	2,439,033	13,603.5	−20.5	(−20.8, −20.3)
Private Insurance
<3 days LOS	5495	635,208	865.1	3141	574,318	546.9	−36.8	(−39.1, −34.5)
>=3 days LOS	33,540	604,713	5546.4	26,080	576,797	4521.5	−18.5	(−19.6, −17.4)
Medicare Insurance
<3 days LOS	46,993	684,178	6868.5	39,995	576,264	6940.4	1.0	(0.0, 2.1)
>=3 days LOS	362,417	1,279,819	28,317.8	251,599	1,136,457	22,138.9	−21.8	(−22.1, −21.5)
Medi‐Cal Insurance
<3 days LOS	7483	577,460	1295.8	5800	541,560	1071.0	−17.4	(−19.7, −15.0)
>=3 days LOS	54,999	715,666	7,685.0	52,499	689,992	7608.6	−1.0	(−2.0, −0.0)
Self‐pay
<3 days LOS	245	56,982	430.0	164	38,104	430.4	0.1	(−16.5, 16.7)
>=3 days LOS	60,223	1,954,208	3081.7	49,119	1,731,224	2837.2	−7.9	(−8.8, −7.0)

Abbreviation: LOS, length of stay; SNF, skilled nursing facility.

4. LIMITATIONS

This study has a number of limitations. First, data were extracted from ED and hospital administrative data collected by a large state regulatory agency (HCAI) and is limited to the specific data elements required and determined by HCAI.

Second, while this study included a large number of ED visits and inpatient admissions from nearly 400 hospitals, our study only included hospitals and EDs in California. Given the varying healthcare landscape by state in terms of types of healthcare institutions, patient and community demographics, and regulatory oversight, our findings may not be generalizable to other regions of the country.

Third, while our study was conducted over a 3‐year period, this duration may be too short to determine the longer term impact of waiving the 3‐day hospital stay rule for SNF coverage. Importantly, the Federal public health emergency declaration, as well as the CMS waiver, expired in May 2023.

Finally, because the waiver was implemented as part of the Federal response to the public health emergency, it is difficult to determine what changes and impact are attributable primarily to the waiver versus other aspects of the pandemic and healthcare response. There were marked changes in public health, individual health and behaviors, larger epidemiologic trends and shifts, and transformation of healthcare delivery and healthcare systems as part of the pandemic that likely impacted the findings reported in this 3‐year study.

Importantly, the waiver period occurred during the height of the pandemic. During that time, many healthcare facilities such as SNFs faced operating challenges, difficulties, and dysfunction that limited access, and likely impacted the findings of our study during the waiver period.

5. DISCUSSION

In this cohort study of all nonfederal hospital EDs in California, the waiver of the CMS 3‐day requirement for SNF placement did not significantly increase rates of SNF discharges from the ED or from patients who were admitted and subsequently discharged from an inpatient hospital setting. In fact, during the 18 months following the implementation of the waiver, rates of SNF discharges from the ED and inpatient setting actually declined overall.

There are a variety of potential reasons why this may have occurred despite widespread promulgation and knowledge of the waiver. First, the study period occurred during the COVID‐19 pandemic (as the waiver itself was in fact part of CMS's response to the public health emergency) and likely impacted by dramatic changes in healthcare access and delivery during the unprecedented pandemic. The myriad of changes that occurred within the healthcare system as a result of the pandemic, such as access restrictions, resource limitations, and staffing shortages, likely affected ED, hospital and SNF utilization during the study period with unintended consequences that undoubtedly impacted our findings. For example, during the initial months following the declaration of the public health emergency, ED census across the nation fell markedly, including in California. ⁵ , ⁶ Similarly, COVID had a major impact on SNF and other long‐term care facilities that adopted efforts to reduce disease spread within their facilities such as limiting access and admissions. ⁷

Second, the 3‐day hospitalization rule primarily applied, and thus the waiver had the most potential impact, on fee‐for‐service (FFS) beneficiaries of Medicare. Other alternative Medicare coverage, such as most Medical Advantage plans and Accountable Care Organizations may not have had the 3‐day requirement under their agreements with CMS during the study period. While FFS remains the largest proportion of Medicare beneficiaries, these alternative coverages have grown dramatically over the last decade. ⁸ Moreover, patients covered by private insurance likely did not have similar requirements and resulting variable SNF coverage.

The CMS 3‐day prior hospital stay requirement for use of the SNF benefit for Medicare beneficiaries was adopted early in its history primarily to ensure appropriate use of the benefit following medical evaluation, treatment, and hospitalization. ⁹ In recent decades, concerns about rising health care costs has led to a reexamination of the 3‐day SNF rule and its effect on acute care hospitalization and SNF utilization. Demonstration projects waiving the 3‐day rule in two states (Oregon and Massachusetts) had variable results with no impact on overall health outcome. ¹⁰ In the late 1980s, the waiver was also suspended briefly under the Medicare Catastrophic Coverage Act, with reports suggesting no impact on hospitalizations, but resulting in increased SNF utilization. ¹¹ , ¹²

It is possible the waiver did have an impact in the ED given the smaller decline in SNF discharge rates compared with the inpatient setting after implementation. In addition, as a CMS initiative, the waiver may have limited the decline in SNF charges for Medicare patients more so when compared with those covered by other private or commercial insurance. For example, we found that for ED SNF discharges, the decrease in the SNF rate was markedly lower compared with private insurance. Similarly, with hospital discharges, and in particular for those patients with an inpatient stay less than 3 days potentially most impacted by the waiver, overall SNF discharges did not change as opposed to the decrease seen with all other payers (with the exception of self‐pay) regardless of LOS duration.

Ulyte et al ¹³ studied Medicare FFS claims data for inpatient hospitalizations and SNF care episodes from 2017 through 2021 to evaluate the impact of the waiver. They reported that SNF care increased significantly for LTC patients without a preceding acute ED visit or hospitalization, but did not significantly change for beneficiaries who were seen in the ED or hospitalized (and in fact, had decreased overall SNF care episodes). This appears to have been driven primarily by LTC patients diagnosed with COVID and transitioned to SNF care at the same facility without evaluation in the ED or acute hospitalization. ¹³

Our findings focused on ED and acute care hospital data appear to be consistent with the claims data as we found no increase (but an actual decrease) in the rates of SNF disposition from both the ED and acute inpatient care settings after the implementation of the waiver. Importantly, overall ED and hospital admission census decreased during the initial period of the pandemic, and it is possible the waiver's impact on LTC patients transitioning to SNF without a direct ED or inpatient visit may have contributed to this decrease. However, it is challenging to differentiate what may have been attributable to the waiver as opposed to other large‐scale changes that were occurring as a result of pandemic and its impact on patient health, behavior, and healthcare utilization.

In conclusion, in this study of all nonfederal acute care hospitals in California, the CMS waiver of the requirement for a preceding 3‐day acute care hospitalization for SNF coverage did not result in an increase in SNF discharges either from the ED or inpatient setting. In fact, there was a decrease in SNF rates from both settings. However, smaller declines were seen for direct ED SNF discharges, Medicare beneficiaries, and those with inpatient stays less than 3 days.

AUTHOR CONTRIBUTION

All authors conceived the study and design methodology. J. J. B. led the data collection and data review. E. C. C. took the lead role in analyzing the data. T. C. C. drafted the manuscript and all authors contributed substantially to its revision. T. C. C. takes responsibility for the paper as a whole.

CONFLICT OF INTEREST STATEMENT

The authors have no commercial, financial, or other relationships related to the subject of this article as per ICMJE conflict of interest guidelines to report.

Biography

Theodore C. Chan, MD, is the chair and professor of the Department of Emergency Medicine at the University of California San Diego Medical Center in Hillcrest and Jacobs Medical Center in La Jolla, California. He also leads the department's development of academic and research programs at UC San Diego School of Medicine.

Chan TC, Brennan JJ, Castillo EM. Impact of skilled nursing facility (SNF) 3‐day hospitalization requirement waiver during the COVID‐19 pandemic on emergency department and inpatient SNF discharges in California. JACEP Open. 2024;5::e13094. 10.1002/emp2.13094