Journal Logo

COVID-19

The Cost of Quarantine

Projecting the Financial Impact of Canceled Elective Surgery on the Nation's Hospitals

Bose, Sourav K. MD, MSc, MBA∗,†; Dasani, Serena MD, MBA‡,§; Roberts, Sanford E. MD§,¶; Wirtalla, Chris BA§,¶; DeMatteo, Ronald P. MD; Doherty, Gerard M. MD; Kelz, Rachel R. MD, MSCE, MBA†,§,¶

Author Information
doi: 10.1097/SLA.0000000000004766

Abstract

COVID-19 is an unprecedented shock to the US healthcare system that places patients at risk and hospitals under financial stress. Regardless of region and hospital type, elective surgeries represent a portion of comprehensive healthcare from which substantial margins are realized, accounting for up to two-thirds of hospital revenues.1 Therefore, losses related to elective surgery cancellation may have significant impact on hospitals. Disasters including Hurricane Katrina2 and the 2003 SARS outbreak3,4 had profound financial impact on hospital operations. However, there is no precedent for the duration, scale, and scope of cancellations realized during COVID-19. The pandemic thus raises questions regarding the financial impact of a 3-month cessation of elective surgeries. Moreover, the resumption of procedures requires strategic considerations that account for ongoing demand and the backlog of patients who were unable to undergo surgery. It is critical to understand the impact that elective surgery cancellations will have on hospitals’ financial solvency and ability to provide safe, high quality care.

To buffer associated losses and subsidize COVID-19 related expenses, the US Department of Health & Human Services will distribute $175B to healthcare systems through the CARES Act.5 However, funding favors hospitals with the highest proportions of private insurance revenue which are typically urban for-profit institutions.6 A consequence of this measure is to asymmetrically allocate financial risk to rural hospitals which already contend with fewer financial reserves, competition with large healthcare chains, and a high percentage of Medicare, Medicaid, and uninsured patients.7,8 Financial failure among these hospitals during the pandemic may place underserved patients at even higher medical risk by reducing their access-to-care. It is equally critical, however, to ensure the financial solvency of urban teaching hospitals which serve high volumes of patients.

Given the variable risk facing different types of hospitals, we sought to predict the financial impact of elective surgery cancellations during the COVID-19 pandemic. We forecasted short-term hospital revenues during March–May 2020 based on hospital region and type, generated sensitivity analyses to assess the backlog of cases, and determined long-term ramp-up times to market equilibrium in which available capacity accounts for ongoing and backlog demand. These data may further inform policies to subsidize hospitals and guide managerial decision-making regarding operational capacity with the aim of keeping US hospitals open and financially solvent.

METHODS

Analytic Sample

The most recent Hospital Cost and Utilization Project (HCUP) Nationwide Inpatient Sample (NIS)9 was selected as it is the largest publicly available sample of all-payers claims and contains a 20% random sample of US hospitals. The NIS was queried from January 2016 to December 2017 for patients older than 18 years-of-age and sorted for those who received a “major” elective procedure as categorized by HCUP procedure classes based on ICD-10 codes.10 Pregnancy-related procedures were excluded; it is expected that the need for these procedures continued at the same cadence throughout the pandemic. Data were matched to HCUP cost-to-charge ratios to provide estimates of case costs.9 NIS weights were applied to extrapolate national estimates from the 20% sample.

Variables

Patient, procedure, and hospital variables were analyzed (Table 1 and Table S1, http://links.lww.com/SLA/C920). Total hospital and procedural costs excluding professional fees were calculated by applying cost-to-charge ratios, adjusting by the wage index, and discounted using 2016–2017 US inflation rates.11 Hospitals were grouped into 12 strata based on location (ie, northeast, midwest/north central, south, west) and teaching status (ie, rural, urban nonteaching, urban teaching).

TABLE 1 - Characteristics of US Hospitals Conducting Elective Surgery 2016–2017 Represented in the NIS Dataset
2016 (N = 777,600) 2017 (N = 772,021)
Frequency # Total Frequency # Total
Census region 777,600 772,021
 Northeast 18.4% 142,977 18.5% 142,577
 Midwest/North Central 24.2% 188,213 24.5% 188,754
 South 37.3% 289,928 37.2% 287,116
 West 20.1% 156,482 19.9% 153,574
Hospital teaching status 777,600 772,021
 Rural 6.3% 48,896 6.29% 48,567
 Urban nonteaching 23.2% 180,571 20.99% 162,023
 Urban teaching 70.5% 548,133 72.7% 561,431
Hospital bed size 777,600 772,021
 Small 20.0% 155,625 21.5% 165,947
 Medium 26.7% 207,713 26.6% 205,368
 Large 53.3% 414,262 51.9% 400,706
Hospital Ownership 772,021
 Government, nonfederal 9.4% 73,164 9.7% 74,942
 Private, not-profit 76.7% 596,221 76.7% 591,985
 Private, invest-own 13.9% 108,215 13.6% 105,094
NIS indicates nationwide inpatient sample.

Statistical Analysis

Patient, procedure, and hospital variables were tabulated. Average hospital revenue by stratum was calculated by summing case costs. Each of the 24 months represented by the dataset were defined as a month-long period. Given the low period count and the need to forecast 36 periods in the future, we selected state-space time series models (SSM). Compared to conventional time series analyses, SSMs better handle missing values and measurement errors, provide recursive and iterative expressions for prediction, and can assess complex seasonalities.12,13 SSMs interpret 3 parameters in order: error, trend, and seasonality. The period-to-period variation in these parameters are defined as non-trend, additive, or multiplicative. These features can also be damped based on second order changes in the growth rate of a given feature.

We reviewed state guidelines on elective surgery cessation and resumption which indicated that over 70% of states mandated the cancellation of elective surgical procedures and that March–May 2020 best represented the national shutdown.14 We used JMP's automated Time Series Analysis Platform to generate revenue forecasts. The system iterates thirty combinations of SSM parameters and uses the Aikaike information criterion to compare each generated model to select the best model.15 Revenue and 95% confidence intervals were forecast for each of 12 strata with a total 360 time series models generated. The best 12 models were selected to generate stratum-level revenue estimates for March–May 2020 (Table S2, http://links.lww.com/SLA/C921). The upper and lower bounds of the 95% confidence intervals were used as “worst” and “best” case scenarios respectively. Sample revenues were tabulated by region. National estimates were calculated using NIS 2016–2017 sample multipliers. Net present values of revenue calculated in 2016 USD were then inflated using 2016–2020 US rates (net r = 0.0782).11

Next, we tabulated the average number of monthly cases at the stratum level—by scaling NIS volume estimates by appropriate discharge weight—from 2016–2017 to estimate pre-COVID baseline demand. To estimate baseline demand in 2020, we generated SSMs to forecast the number of cases within strata on a monthly basis. We calculated the number of missed March-May 2020 elective surgeries, which represents the service backlog.

We then conducted stratum-level sensitivity analyses to calculate the time to both clear backlog and match expected ongoing demand. Capacity was defined by the pre-COVID average annual case volume at a given hospital. Based on industry surveys indicating elective volume ranges between 50% and 100% capacity utilization, we assumed 3 pre-COVID capacity represented utilization rates of 50%, 75%, or 95%.16 In addition, industry surveys demonstrated variable hospital resumption of elective surgery in June 2020 due to delays in staffing and bed availability.16 As such, we applied 3 post-COVID capacity utilization rates of 30%, 45%, and 60% starting in June 2020 to estimate recovery time to market equilibrium. Given that post-COVID capacity utilization is not expected to immediately return to pre-COVID utilization as patients may remain reticent to pursue elective surgery and clinic services may not be fully operational, month-to-month growth rates were utilized to represent the gradual return of activity.17 We assumed a base growth rate of 5% but also evaluated faster growth rates of 15 and 25%. In all analyses, maximum capacity was set at 100% and we assumed that once a hospital achieved 100% capacity utilization, it would continue at that rate until the demand backlog was cleared. Dataset generation was completed in Stata 15 (StataCorp, College Station, TX) and analyzed in JMP 15 (SAS, Cary, NC).

RESULTS

Study Population

The study sample included 1,549,621 discharges across 3776 hospitals (2016–2017) and was representative of US elective surgery patients (Supplemental Table 1, http://links.lww.com/SLA/C920). The median age of elective surgery patients was 63. The average patient had a length of stay 2 days. Demographics were: 55.5% female; 77.9% White, 10.1% Black, and 12% other. Payor mix was biased towards Medicare (48.1%) and private insurers (38.7%). Hospitals were distributed across 4 regions: northeast (18.4%), midwest/north central (24.2%), south (37.3%), and west (20.1%) (Table 2). Most hospitals in the sample were large (53.3%) and were urban teaching (70.5%). The most common procedures were orthopedic, urologic, bariatric, and neurosurgical. Strata representing the most and least patients were southern urban teaching (25.5%) and northeast rural (0.7%) respectively. In all strata, urban teaching hospitals had greater patient volumes than urban nonteaching and rural hospitals (Table 3).

TABLE 2 - Distribution of Hospitals Performing Elective Surgeries by Hospital Region and Teaching Status
Teaching Status
Region Rural Urban Nonteaching Urban Teaching
Northeast 2.5 4 8.5
Midwest/North Central 11.5 7.2 9.3
South 9.7 13.1 13.8
West 4.6 8.1 7.7
Values are reported as % unless otherwise indicated (Total = 100%).

TABLE 3 - Distribution of Patients Who Underwent Elective Surgery by Hospital Region and Teaching Status
Teaching Status
Region Rural Urban Nonteaching Urban Teaching
Northeast 0.7 2.3 15.4
Midwest/North Central 2.4 4.7 17.2
South 2.3 9.4 25.5
West 0.9 5.7 13.5
Values are reported as % unless otherwise indicated (Total = 100%).

Forecasting Revenue

Time series analyses demonstrated diminishing forecasted revenues in urban nonteaching hospitals, flat to diminishing revenues in rural hospitals, and increasing revenues in urban teaching hospitals (Fig. 1) from 2016 to 2020 in the observed NIS sample. In the weighted NIS sample, national revenue loss due complete cessation of elective surgery based on state recommendations from March to May 2020 was $22.3B (95% CI $16.1B, $35.6B) (Table 4). By region, revenue estimates were most consistent in the northeast where hospitals showed the greatest seasonality thereby yielding optimal forecasting accuracy. In contrast, estimates demonstrated the widest error bars in the south and west, where seasonality and trend were profound in urban teaching but not in rural hospitals.

FIGURE 1
FIGURE 1:
Forecasted estimates of elective surgery revenues based on the 20% NIS sample utilizing state-space time series models by hospital region and teaching status January 2016 to December 2020. NIS indicates nationwide inpatient sample.
TABLE 4 - Forecasted Estimates of Total US Elective Surgery Revenues (USD) March to May 2020 Using the Weighted NIS Sample by Hospital Region and Teaching Status Assuming No COVID-19 Related Stoppages and Using Pre-COVID Cost Data. These Forecast Values are Proxies for the Total Potential Loss Related to 100% Cessation of Elective Surgeries From March–May 2020
Estimate Rural Urban Nonteaching Urban Teaching Total
Northeast
 Median revenue 157,229,262 239,878,641 2,757,617,115 3,154,725,018
 Lower 95% CI 86,864,264 183,711,119 2,271,567,114 2,542,142,496
 Upper 95% CI 235,089,538 296,905,846 3,267,079,530 3,799,074,914
Midwest/North Central
 Median revenue 745,695,204 883,378,924 3,599,708,432 5,228,782,560
 Lower 95% CI 448,451,115 658,337,280 3,006,149,756 4,112,938,151
 Upper 95% CI 1,052,072,356 1,133,293,446 4,250,529,257 6,435,895,059
South
 Median revenue 494,607,678 1,722,345,149 6,165,569,135 8,382,521,962
 Lower 95% CI 407,855,347 276,829,359 4,873,508,145 5,558,192,851
 Upper 95% CI 588,628,531 10,051,591,223 7,579,964,845 18,220,184,599
West
 Median revenue 452,908,434 848,485,046 4,232,998,182 5,534,391,661
 Lower 95% CI (100,597,899) 366,209,202 3,688,436,745 3,954,048,048
 Upper 95% CI 995,336,860 1,386,705,014 4,792,527,917 7,174,569,790
National
 Median revenue 22,300,421,201
 Lower 95% CI 16,167,321,546
 Upper 95% CI 29,647,158,691
The median revenue was generated utilizing JMP Time Series Analysis Platform. All values are reported in 2020 US dollars unless otherwise indicated.CI indicates confidence interval; NIS, nationwide inpatient sample.

Predicted Recovery Time

Recovery time to market equilibrium was conserved across strata and was influenced by pre- and post-COVID capacity utilization (Table 5). The number of months to recovery ranged from 4 to 164 across modeled scenarios. The median case (75% pre-COVID utilization rate/45% post-COVID utilization rate) demonstrated recovery times of 12–22 months across all strata. Lower levels of pre-COVID utilization were associated with fewer months to recovery. The reduction in recovery time associated with increased rate of return to activity was highest among institutions that had lower pre-COVID rates of capacity utilization. The benefit of a faster monthly growth rate was diminished if post-COVID starting utilization rates were higher than baseline pre-COVID utilization rates.

TABLE 5 - Recovery Time in Months to Market Equilibrium Assuming Various Pre-COVID Utilization Rates, Monthly Growth Rates, and Post-COVID Utilization Rates by Hospital Region and Teaching Status
Pre-COVID Capacity Utilization Rate
50% 75% 95%
Post-COVID Starting Rate Post-COVID Starting Rate Post-COVID Starting Rate
Growth Rate 30% 45% 60% 30% 45% 60% 30% 45% 60%
Northeast rural 5% 13 9 6 29 22 16 161 122 85
15% 6 5 5 16 14 12 95 82 71
25% 5 4 4 14 12 11 83 80 68
Northeast urban nonteaching 5% 13 9 6 29 22 16 164 125 95
15% 7 5 5 17 14 12 99 85 75
25% 6 5 4 15 13 11 87 83 71
Northeast urban teaching 5% 13 9 6 30 22 16 162 123 94
15% 7 5 5 17 14 12 98 84 73
25% 5 5 4 14 12 11 85 82 69
Midwest/North Central rural 5% 13 9 6 30 22 16 161 122 92
15% 12 5 5 16 14 11 95 82 71
25% 5 4 4 14 12 11 83 80 68
Midwest/North Central urban nonteaching 5% 13 9 6 29 21 15 159 120 90
15% 12 5 5 16 13 11 94 80 68
25% 5 4 4 14 12 10 81 77 65
Midwest/North Central urban teaching 5% 13 9 6 29 22 16 160 122 91
15% 12 5 5 16 14 11 95 81 70
25% 5 4 4 14 12 11 82 79 66
South rural 5% 13 9 6 29 22 16 160 121 91
15% 7 5 5 16 14 13 95 81 71
25% 5 4 4 14 12 11 83 79 67
South urban nonteaching 5% 13 9 6 29 22 16 161 122 92
15% 6 5 5 16 14 13 96 82 71
25% 5 4 4 14 12 11 83 80 67
South urban teaching 5% 13 9 6 30 22 16 162 123 93
15% 7 5 5 17 14 13 97 84 73
25% 6 5 4 14 12 11 85 82 70
West rural 5% 13 9 6 29 22 16 160 121 91
15% 6 5 5 16 14 11 94 81 70
25% 5 4 4 14 12 11 81 78 66
West urban nonteaching 5% 13 9 6 29 21 15 158 119 89
15% 6 5 5 16 13 11 92 78 67
25% 5 4 3 13 11 10 78 76 63
West urban teaching 5% 13 9 6 30 22 16 162 123 93
15% 7 5 5 17 14 13 97 84 73
25% 5 4 4 14 12 11 85 82 69
Pre-COVID hospital capacity utilization rates were assumed to be 50%, 75%, and 95%.
Monthly growth rates were assumed to be 5%, 15%, 25%.
Post-COVID hospital capacity utilization rates were assumed to be 30%, 45%, and 60%. In all analyses, it is assumed that once 100% capacity utilization was achieved, utilization would continue at that rate until the demand backlog was cleared.

DISCUSSION

Impact in the Short-term (March–May 2020)

Total forecasted revenue lost during the pandemic due to elective surgery cancellation alone from March to May 2020 was $22.3B. Forecasted revenue loss is pervasive across all regions and hospital types regardless of teaching status. However, acute revenue losses are compounded by preexisting trends which may place urban nonteaching and rural hospitals at increased financial risk.18 Moreover, total losses over the next year will be dependent on factors such as the rate of reopening and patient willingness to schedule surgery.

Before the onset of the pandemic, rural and urban nonteaching hospitals were already facing financial uncertainty leading to an increase in their reliance on elective surgery revenue.19–24 Consistent with this trend, we observed flat or negative growth rates in these strata with regard to operative revenue. Furthermore, urban nonteaching hospitals may also be at risk as our analyses demonstrate a rapid decline in urban nonteaching hospital revenues in the northeast and more gradual declines in other regions from 2016 to 2020. This is a trend that has been attributed to demographic shifts and market pressures such as competition and recession and has been concentrated in Medicaid non-expansion states.24–26 Excess capacity and reduced patient volumes in affected hospitals contribute to low baseline operating margins.25,26 Thus, pre-COVID financial risk due to dependence on elective surgery revenue may exacerbate the impact of elective surgery cessation on both rural and urban nonteaching hospitals.

To provide relief to hospitals, the CARES Act will distribute $175B to US hospitals and skilled nursing facilities to support the cost of COVID care and reimburse lost revenue due to the pandemic with $50B in general funds distributed to health systems through May 20205 These funds will be allocated as disbursements determined as follows. The initial $30B disbursement was ultimately based on 2019 Medicare fee-for-service (MFFS) billings to facilitate rapid payouts for COVID care.27 For example, according to published CARES Act formulae, a hospital with total annual billings of $200 million ($50 million per 3-month period) and a private/Medicare split of 50/50 in 2019 would expect an initial disbursement of approximately $6 million (disbursmentMFFSbillings($0.1B)totalMFFS($453B)× Overallsubsidy($30B)).5 An additional $20B was disbursed based on hospital gross receipts regardless of payor mix; thus, the same hospital could then receive an additional $4 million (disbursmentTotalreceipts($0.2B)$2.5T)×$50B).5 Additional targeted payments are expected to be distributed based on COVID impact, MFFS billings, total receipts, and operating expenses.

Using aggregate metrics to determine subsidy disbursements can exacerbate disparities. Over half the general disbursement is allocated based on MFFS and this disadvantages hospitals that serve uninsured populations. Moreover, as the rate of increase in 2019 Medicare FFS billings is not driven by volume but rather by price and care intensity, hospitals dealing in high volume, low complexity cases may not receive an appropriately proportionate subsidy.25 The remainder of the general disbursement is allocated based on total receipts which disadvantages hospitals with primarily public payors. One report suggests that hospitals in the top 10% with respect to private insurance revenue will receive over double the funds compared to those in the bottom 10%.6 In addition, targeted payments to COVID prevalent areas do not account for losses among hospitals in regions with low COVID patient volume but that were also required to cease elective operations. Indeed, the pandemic universally poses a solvency threat to US hospitals which, unlike in other acutely distressed industries such as airlines and hotels, puts patients directly at risk. If hospitals close because they cannot pay workers or meet debt obligations, both elective and emergency care will be less available to local communities.

Impact in the Long-term (June 2020 and Beyond)

One response to the threat of financial insolvency has been to furlough medical workers but this strategy may portend greater risk in the long run. For example, recovery time seems to be independent of hospital type and region and dependent on capacity utilization. Hospitals that were operating below capacity pre-COVID may be able to hasten recovery by rapidly and maximally increasing capacity utilization. For example, in the best-case scenario we predict a hospital that previously operated at 50% capacity but resumed activity post-COVID at 60% capacity with a month-to-month growth rate of 25% could recover volume within 4 months. This case may be most appropriate for rural and urban nonteaching hospitals that previously underutilized capacity and poses considerations for staffing, physical resource allocation, and marketing. In contrast, the highest pre-COVID capacity utilizers will have the least room to account for missed volume and may take the longest time to recover.

If the capacity ceiling has been reached, alternative strategies to increase capacity will be essential. For example, hiring additional staff, scheduling cases on nights and weekends, pursuing mergers/acquisitions, and accelerating infrastructure investments will increase a hospital's capacity and thereby reduce recovery time from years to months. Finally, marketing will be an important competitive tool for all hospitals. Pandemic-related fear may dampen patients’ willingness to undergo elective surgery.17,28,29 As such, assuaging patients’ fears will be an essential driver of post-COVID demand. Thus, identifying pre- and post-COVID capacity utilization may significantly impact operational strategy. Rather than reducing operations and furloughing workers, aggressively pursuing marketing and investment opportunities may reduce health system risk, particularly as the handicap of pandemic-related losses is likely proportional regardless of hospital type and region.

There are several limitations to this study. First, forecasting was conducted based on 2016–2017 claims data which provides a relatively low number of periods. In addition, NIS sampling variation at the hospital level may affect the accuracy of volume estimates at the stratum level. However, the NIS is the most inclusive all-payers database and 2017 is the most recent available dataset. Second, our sensitivity analyses assume a single shutdown affecting all elective surgeries equally and that utilization will return to 100% of pre-COVID capacity; however, new regulations regarding social distancing, environmental policies, and clinic access may dampen recovery. Third, the backlog of patients who did not undergo elective surgery may not be completely cleared due to patient factors. For example, some patients may have suffered disease progression in the interim resulting in inoperable pathology or death, some may be lost to follow-up, and others may be fearful of undergoing surgery in urban teaching hospitals and contracting COVID-19. Finally, the data analyzed here reflects elective surgery alone; however, assessment of total revenue losses related to COVID-19 should also consider other profitable service-lines such as radiology, interventional services, and infusions.

In conclusion, US elective surgery cessation from March to May 2020 is predicted to result in a revenue loss of $22.3B. Recovery to pre-COVID supply-demand equilibrium will require rapid increase in capacity utilization and may benefit from capacity expansion at the hospital level. Distributions from the CARES Act may be inadequate to buffer losses observed by rural and urban nonteaching hospitals, which may face disproportionate financial solvency risk thereby exacerbating care disparities.

REFERENCES

1. Jackson RL. The business of surgery. Managing the OR as a profit center requires more than just IT. It requires a profit-making mindset, too. Health Manag Technol 2002; 23:20–22.
2. Breaux JA, French M, Richardson WS. Effect of natural disaster on routine surgery. J Am Coll Surg 2009; 209:352–355.
3. Chang H-J, Huang N, Lee C-H, et al. The impact of the SARS epidemic on the utilization of medical services: SARS and the fear of SARS. Am J Public Health 2004; 94:562–564.
4. The Impact of the SARS Epidemic on the Utilization of Medical Services: SARS and the Fear of SARS | AJPH | Vol. 94 Issue 4 [Internet]. [Cited 2020 Jun 18]. Available at: https://ajph.aphapublications.org/doi/full/10.2105/AJPH.94.4.562. Accessed July 8, 2020
5. CARES Act Provider Relief Fund: For Providers | HHS.gov [Internet]. [Cited 2020 Jun 18]. Available at: https://www.hhs.gov/coronavirus/cares-act-provider-relief-fund/for-providers/index.html. Accessed July 8, 2020
6. May 13 ADP, 2020. Distribution of CARES Act Funding Among Hospitals [Internet]. KFF. 2020 [Cited 2020 Jun 18]. Available at: https://www.kff.org/coronavirus-covid-19/issue-brief/distribution-of-cares-act-funding-among-hospitals/. Accessed July 8, 2020
7. Kaufman BG, Thomas SR, Randolph RK, et al. The rising rate of rural hospital closures. J Rural Health 2016; 32:35–43.
8. Edmiston KD. Rural Hospital Closures and Growth in Employment and Wages. Economic Bulletin [Internet] 2019 [Cited 2020 Jun 18]. Available at: https://ideas.repec.org/a/fip/fedkeb/00006.html. Accessed July 8, 2020
9. HCUP National Inpatient Sample (NIS). Healthcare Cost and Utilization Project (HCUP). 2012. Agency for Healthcare Research and Quality, Rockville, MD. [Internet]. Available at: https://www.hcup-us.ahrq.gov/db/ccr/ip-ccr/ip-ccr.jsp. Accessed July 8, 2020
10. Procedure Classes for ICD-10-PCS (beta version) [Internet]. [Cited 2020 Sep 7]. Available at: https://www.hcup-us.ahrq.gov/toolssoftware/procedureicd10/procedure_icd10.jsp. Accessed July 8, 2020
11. Current US Inflation Rates: 2009–2020 [Internet]. US Inflation Calculator. 2008 [Cited 2020 Jun 18]. Available at: https://www.usinflationcalculator.com/inflation/current-inflation-rates/. Accessed July 8, 2020
12. OUP Oxford, Durbin J, Koopman SJ. Time Series Analysis by State Space Methods. Second Edition2012.
13. Maunder MN, Deriso RB, Hanson CH. Use of state-space population dynamics models in hypothesis testing: advantages over simple log-linear regressions for modeling survival, illustrated with application to longfin smelt (Spirinchus thaleichthys). Fisheries Res 2015; 164:102–111.
14. State Guidance on Elective Surgeries [Internet]. [Cited 2020 Jun 18]. Available at: https://www.ascassociation.org/covid-19-state. Accessed July 8, 2020
15. Liew VK-S. On Autoregressive Order Selection Criteria [Internet]. 2004 [Cited 2020 Jun 18]. Available at: /paper/On-Autoregressive-Order-Selection-Criteria-Liew/8560eb7c6ca0d9c7d3fc94520093142571f65405. Accessed July 8, 2020
16. The Impact of COVID-19 on US Hospitals — Edition 10 | L.E.K. Consulting [Internet]. [Cited 2020 Jun 18]. Available at: https://www.lek.com/insights/impact-covid-19-us-hospitals-edition-10. Accessed July 8, 2020
17. Vanni G, Materazzo M, Pellicciaro M, et al. Breast cancer and COVID-19: the effect of fear on patients’ decision-making process. In Vivo 2020; 34: (3 Suppl): 1651–1659.
18. Khullar D, Bond AM, Schpero WL. COVID-19 and the financial health of US hospitals. JAMA 2020; 323:2127–2128.
19. 171 Rural Hospital Closures: January 2005 - Present (129 Since 2010) - Sheps Center [Internet]. [Cited 2020 Jun 18]. Available at: https://www.shepscenter.unc.edu/programs-projects/rural-health/rural-hospital-closures/. Accessed July 8, 2020
20. New Risks at Rural Hospitals - WSJ [Internet]. [Cited 2020 Jun 18]. Available at: https://www.wsj.com/articles/new-risks-at-rural-hospitals-1451088096. Accessed July 8, 2020
21. Wishner J, Solleveld P, Rudowitz R, et al. A look at rural hospital closures and implications for access to care: three case studies. Kaiser Family Foundation [Internet]. 2016.
22. Understanding The Relationship Between Medicaid Expansions And Hospital Closures | Health Affairs [Internet]. [Cited 2020 Jun 18]. Available at: https://www.healthaffairs.org/doi/full/10.1377/hlthaff.2017.0976?casa_token=hCm_CH1d7cMAAAAA%3AhPlVV81s-I1HozXT-ObTFdNhflMT5ab4CRaF-QPtem_TDiPqrgsbbwGnW421evAvAJzzv24csk4. Accessed July 8, 2020
23. National Bureau of Economic Research, Gujral K, Basu A. Impact of Rural and Urban Hospital Closures on Inpatient Mortality [Internet]. 2019.
24. Mullner RM, McNeil D. Rural and urban hospital closures: a comparison. Health Aff 1986; 5:131–141.
25. Crosson FJ, Ginsburg P, Buto K, et al. (n.d.). COMMISSIONERS PRESENT: 412. http://www.medpac.gov/docs/default-source/meeting-materials/medpac-december-2019-transcript_rev121219_sec.pdf?sfvrsn=0. Accessed March 13, 21
26. Public Meeting Agenda for December 5–6, 2019 [Internet]. [Cited 2020 Jun 18]. Available at: http://www.medpac.gov/-public-meetings-/meeting-details/december-2019-public-meeting. Accessed July 8, 2020
27. Affairs (ASPA) AS for P. CARES Act Provider Relief Fund: General Information [Internet]. HHS.gov. 2020 [Cited 2020 Jun 18]. Available at: https://www.hhs.gov/coronavirus/cares-act-provider-relief-fund/general-information/index.html. Accessed July 8, 2020
28. Rosenbaum L. The untold toll — the pandemic's effects on patients without Covid-19. N Engl J Med 2020; 382:2368–2371.
29. Søreide K, Hallet J, Matthews JB, et al. Immediate and long-term impact of the COVID-19 pandemic on delivery of surgical services. Br J Surg [Internet] 2020.
Keywords:

COVID-19; econometrics; elective surgery; hospital management; hospital solvency; recovery; SARS-COV2; strategy; surgery

Supplemental Digital Content

Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.