Abstract
This cross-sectional study examines whether characteristics of hospitals differ across 5 frequently used safety-net hospital definitions using 2020 data.
Introduction
Safety-net hospitals (SNHs) provide care for many Medicaid beneficiaries and the uninsured, although the definition of SNHs varies.1 While the US Centers for Medicare & Medicaid Services (CMS) provides Disproportionate Share Hospital (DSH) funds for state allocation to SNHs for this care, there is little connection between the DSH calculation and the level of uninsured patients, uncompensated care, or safety-net services.1,2,3 Seeking to better characterize the safety-net and inform SNH funding decisions, researchers and policymakers have developed definitions that include uncompensated care costs, ownership, and Medicaid caseloads.4 Research has compared hospitals using these definitions, albeit omitting some and using less current data.5 The aim of this study is to examine whether characteristics of hospitals differ across 5 frequently used SNH definitions using 2020 data.
Methods
This cross-sectional study analyzed hospital and community characteristics captured in the 2020 American Hospital Association Annual Survey and the 2020 CMS DSH Adjustment Data using 5 SNH definitions identified in a 2021 literature review4: (1) hospitals in the top quartile of DSH payments using the DSH patient percentage formula; (2) Medicaid inpatient discharge days (1 SD above the state mean); (3) Medicaid inpatient discharge days or public hospital (a nonfederal government hospital); (4) public hospital or teaching hospital; or (5) public hospital.4 For each definition, we assessed the proportion and number of hospitals; rurality; ownership type; bed total; state Medicaid expansion status; and Case Mix Index (CMI), an indicator in which higher values indicate greater clinical complexity. The study followed the STROBE reporting guideline.
Results
Our sample of 5955 US hospitals included 31.87% rural hospitals (n = 1898), 20.44% (n = 1217) nonfederal government (defined as public) hospitals, 26.88% for-profit hospitals (n = 1601), and 52.68% nonprofit hospitals (n = 3137) (Table). More than 56% are small hospitals (6-99 beds) and 61.43% were in states that expanded Medicaid. The mean (SD) CMI was 1.63 (0.01). The first DSH definition identified 11.42% of hospitals (n = 680) as SNHs compared with 14.79% (n = 881) using the second (Medicaid inpatient discharge) definition, 31.50% (n = 1876) using the third (Medicaid discharge or public hospital) definition, 55.18% (n = 3286) using the fourth (public hospital or teaching hospital) definition, and 20.44% (n = 1217) using the fifth (public hospitals) definition. The first definition captured the fewest (n = 185, 27.21%) rural and the fewest public hospitals (n = 144, 21.18%). The fourth definition had a CMI above the mean (SD) of 1.68 (0.01), captured the greatest numbers of small hospitals (n = 1316), nonprofit hospitals (n = 1643), and hospitals in Medicaid expansion states (n = 2112).
Table. Characteristics by SNH Definition Among US Hospitals, 2020.
| Characteristic | No. (%) | |||||
|---|---|---|---|---|---|---|
| All hospitals | SNH definitiona | |||||
| Definition 1: DSH paymentb | Definition 2: Medicaid inpatient discharges | Definition 3: Medicaid inpatient discharges or public hospitalc | Definition 4: teaching hospital or public hospitalc | Definition 5: public hospitalc | ||
| Total | 5995 (100) | 680 (11.4) | 881 (14.8) | 1876 (31.5) | 3286 (55.2) | 1217 (20.4) |
| Ownership | ||||||
| Public hospital, nonfederal government | 1217 (20.44) | 144 (21.18) | 222 (25.20) | 1217 (64.87) | 1217 (37.04) | 1217 (100) |
| For profit | 1601 (26.88) | 116 (17.06) | 260 (29.51) | 260 (13.86) | 426 (12.96) | 0 |
| Nonprofit | 3137 (52.68) | 420 (61.76) | 399 (45.29) | 399 (21.27) | 1643 (50.00) | 0 |
| Rural hospital | 1898 (31.87) | 185 (27.21) | 260 (29.51) | 816 (43.50) | 946 (28.79) | 672 (55.22) |
| Bed size | ||||||
| Small (6-99 beds) | 3349 (56.24) | 171 (25.15) | 434 (49.26) | 1114 (59.38) | 1316 (40.05) | 793 (65.16) |
| Medium (100-299 beds) | 1759 (29.54) | 286 (42.06) | 321 (36.44) | 530 (28.25) | 1175 (35.76) | 277 (22.76) |
| Large (≥300 beds) | 847 (14.22) | 223 (32.79) | 126 (14.30) | 232 (12.37) | 795 (24.19) | 147 (12.08) |
| Medicaid Expansion State | 3658 (61.43) | 461 (67.79) | 552 (62.66) | 1123 (59.86) | 2112 (64.27) | 685 (56.29) |
| Case Mix Index, mean (SD) | 1.63 (0.01) | 1.57 (0.01) | 1.53 (0.02) | 1.50 (0.02) | 1.68 (0.01) | 1.48 (0.02) |
Abbreviations: DSH, Disproportionate Share Hospital; SNH, safety-net hospital.
The 5 frequently used SNH definitions are given as identified in Hefner et al.4
DSH defined using the DSH Patient Percentage calculation, using total Supplemental Security Income days, Medicare days, Medicaid days, and total patient days.2
Public hospital defined as a nonfederal government owned hospital.
Discussion
Different SNH definitions captured varying hospital characteristics, which has implications for funding and health care access. For instance, the University of Florida Jacksonville Health Hospital and Orlando Health-Health Central Hospital shared 4 of the 5 examined characteristics (nonprofit ownership, nonrural, large bed size, and non-Medicaid expansion) but were excluded from different SNH definitions, illustrating how differing definitions can exclude similar hospitals.
Given these differences, a universal SNH definition or an index that incorporates a wider range of measures may be needed. For example, a multi-stakeholder collaborative in California developed a Hospital Social Needs Index using publicly available geographic social needs measures combined with hospital specific information.6 An SNH index, while limited, could allow researchers and policymakers to methodically identify SNHs across multiple relevant characteristics. Further, a definition or index that includes hospital services in addition to Medicaid days would identify more hospitals as SNHs in nonexpansion states. Limitations to this study include the use of self-reported data, the absence of a patient social need indicator, and the survey’s 75% response rate. Future studies should examine these and identify how different definitions are associated with patient outcomes.
Data Sharing Statement
References
- 1.Chatterjee P, Sommers BD, Joynt Maddox KE. Essential but undefined—reimagining how policymakers identify safety-net hospitals. N Engl J Med. 2020;383(27):2593-2595. doi: 10.1056/NEJMp2030228 [DOI] [PubMed] [Google Scholar]
- 2.Centers for Medicare & Medicaid Services . Disproportionate Share Hospital (DSH). Updated June 6, 2023. Accessed July 19, 2023. https://www.cms.gov/medicare/medicare-fee-for-service-payment/acuteinpatientpps/dsh
- 3.Medicaid and CHIP Payment and Access Commission . Report to the Congress on Medicaid and CHIP. March 2011. Accessed August 3, 2023. https://www.macpac.gov/wp-content/uploads/2011/03/March-2011-Report-to-Congress-on-Medicaid-and-CHIP.pdf
- 4.Hefner JL, Hogan TH, Opoku-Agyeman W, Menachemi N. Defining safety net hospitals in the health services research literature: a systematic review and critical appraisal. BMC Health Serv Res. 2021;21(1):278. doi: 10.1186/s12913-021-06292-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Popescu I, Fingar KR, Cutler E, Guo J, Jiang HJ. Comparison of 3 safety-net hospital definitions and association with hospital characteristics. JAMA Netw Open. 2019;2(8):e198577. doi: 10.1001/jamanetworkopen.2019.8577 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Senathirajah M, Dankwa-Mullan I, Pickens G, Benevent R, Spurlock B. A hospital social needs index would help hospitals collaborate to address social needs and health equity. Health Affairs Forefront. Published online August 2, 2022. doi: 10.1377/forefront.20220729.222359 [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Sharing Statement