key: cord-0756754-77w2f09o
authors: Werner, Rachel M.; Bressman, Eric
title: Trends in post-acute care utilization during the COVID-19 pandemic
date: 2021-09-07
journal: J Am Med Dir Assoc
DOI: 10.1016/j.jamda.2021.09.001
sha: 276132804c9acfe8f666fce963347281a91f5d0c
doc_id: 756754
cord_uid: 77w2f09o

Objective To examine the effect of the COVID-19 pandemic on post-acute care utilization and spending. Design We used a large national multi-payer claims dataset from January 2019 through October 2020 to examine trends in post-hospital-discharge location and spending Setting and participants We identified and included 975,179 hospital discharges who were 65 years or older. Methods We summarized post-discharge utilization and spending in each month of the study: (1) the percentage of patients discharged from the hospital to home for selfcare and to the three common post-acute care locations: home with home health, skilled nursing facility (SNF) and inpatient rehabilitation; (2) the rate of discharge to each location per 100,000 insured members in our cohort; (3) the total amount spent per month in each post-acute care location; and (4) the percentage of spending in each post-acute care location out of the total spending across the three post-acute care settings. Results The percentage of patients discharged from the hospital to home or to inpatient rehabilitation did not meaningfully change during the pandemic while the percentage discharged to SNF declined from 19% of discharges in 2019 to 14% by October 2020. Total monthly spending declined in each of the three post-acute care locations, with the largest relative decline in SNFs of 55%, from an average of $42 million per month in 2019 to $19 million in October 2020. Declines in total monthly spending were smaller in home health (a 41% decline) and inpatient rehabilitation (a 32% decline). As a percentage of all post-acute care spending, spending on SNFs declined from 39% to 31% while the percentage of post-acute care spending on home health and inpatient rehabilitation both increased. Conclusions and Implications Changes in post-hospital-discharge location of care represent a significant shift in post-acute care utilization, which persisted nine months into the pandemic. These shifts could have profound implications on the future of post-acute care.

The COVID-19 pandemic had profound effects on health care delivery, utilization, and spending. 71

Beginning in March 2020 hospital admissions declined as hospitals deferred elective admissions 72 and procedures. Physicians pivoted from in-person care to telemedicine. And many patients 73 deferred health care services. These changes were associated with large and unparalleled 74 declines in medical spending. 1 75 76 While changes in hospitalization rates and shifts to telemedicine have been well described, less 77 attention has been paid to how COVID-19 affected post-acute care, which is a very commonly 78 used after hospital discharge in the U.S. and results in high health care spending. 2 79 80

The effects of COVID-19 on utilization and spending on post-acute care may be particularly 81 important for nursing homes, one of the most common sites of post-acute care, as they have 82 been particularly hard hit during the pandemic with substantial declines in census 3 and 83 revenue, leading to financial instability and closures. 4 With higher reimbursement for Medicare-84 paid skilled nursing stays, post-acute care admissions are an important source of revenue for 85 nursing homes and has been a large and often growing part of Medicare's budget for years. In 86 2018, one-fifth of hospitalized Medicare beneficiaries were discharged to a skilled nursing 87 facility (SNF), costing Medicare $28.5 billion. 2 88 89 Understanding how the pandemic has altered use of post-acute care is critically important to 90 informing expectations and policies aimed at funding post-acute care and optimizing the use of 91 post-acute care moving forward, particularly for nursing homes. 92 93 In this paper we use a large national dataset to document COVID-19-associated changes in the 94 utilization and spending on post-acute care. 95 96

We used multi-payer deidentified non-capitated claims data from FAIR Health. FAIR Health is an 99 independent nonprofit organization that maintains a data repository which contains privately 100 billed medical and dental claims contributed by over 60 payors nationwide. We used 101 longitudinal data including over 70 million commercially insured individuals, including Medicare 102 Advantage beneficiaries. 103

We identified all hospital discharges for patients 65 years or older between January 2019 and 105

October 2020. Using their first location of care after hospital discharge, we identified each 106 discharge's subsequent post-discharge destination -including the three most common sites of 107 post-acute care (home with home health care, admission to a skilled nursing facility (SNF), or 108 inpatient rehabilitation facility) and discharge home for selfcare. We excluded the 109 approximately 2% of hospital discharges that went to long-term acute care hospitals, hospice, 110 or other institutional settings such as psychiatric hospitals. 

We identified and included 975,179 hospital discharges who were 65 years or older. The 127 percentage of discharges to each of the four post-discharge locations remained steady in 2019. 128

In 2020, the percentage of patients discharged home for selfcare or with home health increased 129 slightly (from an average of 52% in 2019 to 54% in October 2020 for selfcare; 20% to 21% for 130 home health; Figure 1a ). The percentage of patients discharged to inpatient rehabilitation did 131 not meaningfully change. The use of SNF declined over this time period, from an average of 132 19% of discharges in 2019 to 14% in October 2020. Medicare and other payers implemented a number of alternative payment arrangements over 168 the last decade, such as accountable care organizations and bundled payments which hold 169 providers accountable for total costs of care. With its high costs and high rates of utilization, 2 170 reducing skilled nursing facility utilization has been a common target in these alternative 171 payment models to reduce health care spending. As a result, home-based care was becoming 172 more common after hospital discharge, even before the pandemic. 5 With the pandemic, we 173 document an acceleration in these trends, particularly in SNFs, likely due to concerns about 174 COVID-19 rates in nursing homes. 175 176 These changes in post-acute care utilization have important implications for patients, families, 177 and caregivers. First, shifting post-discharge care home is often consistent with patient 178 preferences However, it may also shift the burden of care to family members or other unpaid 179 caregivers. 6 While home health care is often available after hospital discharge, its benefits may 180 not provide sufficient care to patients recovering from hospitalization, who often need short-181 term help with activities of daily living. Families and friends often must fill in these gaps in care. 182 183 J o u r n a l P r e -p r o o f Second, shifting care home may result in worse outcomes for some patients. There is increasing 184 concern that beyond limitations in the level of help for activities of daily living, current models 185 of home-based care may also provide insufficient medical care to care for ill patients recovering 186 from hospitalization. 7 Prior research has found that compared to patients who receive post-187 acute care in nursing homes, those we are discharged home have higher rehospitalization 188 rates. 8, 9 Further development of more intensive support for home-based post-acute care 189 models may be needed for some conditions, 10 particularly those higher acuity conditions that post-COVID era, as the pandemic has harmed the finances of many nursing homes due to a 203 declining census 12 and the increased costs needed to adequately respond to the pandemic in 204 terms of equipment, supplies, and staffing. The general consensus is that nursing homes use 205

Medicare-based revenue to cross-subsidize Medicaid stays. 2,13 With the decline in post-acute 206 care in nursing homes and the accompanying decline in payment to nursing homes that we 207 document, nursing homes may be losing an important source of revenue. The decline in SNF 208 admissions may hasten a change in the role of nursing homes, 11,14 causing nursing home to 209 focus more on long-term care as post-acute care admissions decline. This specialization in long-210 term care may benefit patients, but, given the lower Medicaid reimbursement rates, may also 211 cause nursing homes to become more financially unstable, and, in the worst case, forcing some 212 to close. 213 214 It is unknown whether these trends will continue as the pandemic fades and what the full 215 implications of these shifts on patients, families, and health care systems are. However, they 216 suggest there may be a fundamental shift in providing post-acute care at home, rather than in 217 nursing homes. If we are to continue to support the growth of care at home, we will need to 218 expand the availability of and access to paid personal care assistance, which is currently an 219 optional benefit in Medicaid and therefore subject to considerable state variation. We may also 220 need to consider more wide-spread implementation of more intensive home-based post-acute 221 care models that can more feasibly substitute for SNF care. 15 destinations (home with self-care, skilled nursing facility, home with home health, and inpatient 271 rehabilitation). Panel a displays these as a percentage of all hospital discharges. Panel b displays 272 these as the rate per 100,000 insured members. 273 274 Figure 2 . Trends in spending on the three most common settings for formal post-acute care 275

(skilled nursing facility (SNF), home with home health (HH), and inpatient rehabilitation (IRF)). 276

The trend lines display the total monthly spending in each setting. The stacked bars display the 277 monthly spending in each of these three settings as a percentage of total spending across these 278 three post-acute care settings. 279 280 281

Implications of Early Health Care Spending 229 Reductions for Expected Spending as the COVID-19 Pandemic Evolves

Report to the Congress: Medicare Payment Policy

Nursing Home Staffing Levels Did Not Change Significantly During 234 COVID-19: Study examines US nursing home staffing levels during COVID-19 pandemic

Nursing Home Industry Projects $34B in Revenue Losses, 1,800 Closures or 237 Mergers Due to COVID. Skilled Nursing News

Two-year evaluation of mandatory 239 bundled payments for joint replacement

Shifting the Burden? Consequences Of Postacute 242

Care Payment Reform On Informal Caregivers

Postacute care-the piggy bank for savings in 244 alternative payment models? The New England journal of medicine

Patient outcomes after hospital discharge to 246 home with home heath care versus to skilled nursing facility

The Effects of Skilled Nursing Facility Care: Regression Discontinuity Evidence 249 from Medicare

The Time Of Covid-19, We Should Move High-Intensity

Turning Away From Nursing Homes, to What? New York Times

Firm Finances and the Spread of COVID-19: Evidence from 255

Georgia Tech Scheller College of Business Research Paper No 3659480

Cross-Subsidization in Nursing Homes: Explaining Rate Differentials among 258

Long-term care policy after Covid-19-solving the 260 nursing home crisis

Implementation of post-acute 262 rehabilitation at home: a skilled nursing facility-substitutive model